FTOS Configuration Guide for the S60 System FTOS 8.3.3.

Notes, Cautions, and Warnings

NOTE: A NOTE indicates important information that helps you make better use of your computer. CAUTION: A CAUTION indicates either potential damage to hardware or loss of data and tells you how to avoid the problem.
WARNING: A WARNING indicates a potential for property damage, personal injury, or death.

Information in this publication is subject to change without notice. 2013 Dell Force10. All rights reserved. Reproduction of these materials in any manner whatsoever without the written permission of Dell Inc. is strictly forbidden. Trademarks used in this text: Dell, the Dell logo, Dell Boomi, Dell Precision , OptiPlex, Latitude, PowerEdge, PowerVault, PowerConnect, OpenManage, EqualLogic, Compellent, KACE, FlexAddress, Force10 and Vostro are trademarks of Dell Inc. Intel, Pentium, Xeon, Core and Celeron are registered trademarks of Intel Corporation in the U.S. and other countries. AMDis a registered trademark and AMD Opteron, AMD Phenom and AMD Sempron are trademarks of Advanced Micro Devices, Inc. Microsoft, Windows, Windows Server, Internet Explorer, MS-DOS, Windows Vista and Active Directory are either trademarks or registered trademarks of Microsoft Corporation in the United States and/or other countries. Red Hat and Red HatEnterprise Linux are registered trademarks of Red Hat, Inc. in the United States and/or other countries. Novell and SUSE are registered trademarks of Novell Inc. in the United States and other countries. Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Citrix, Xen, XenServer and XenMotion are either registered trademarks or trademarks of Citrix Systems, Inc. in the United States and/or other countries. VMware, Virtual SMP, vMotion, vCenter and vSphere are registered trademarks or trademarks of VMware, Inc. in the United States or other countries. IBM is a registered trademark of International Business Machines Corporation. Other trademarks and trade names may be used in this publication to refer to either the entities claiming the marks and names or their products. Dell Inc. disclaims any proprietary interest in trademarks and trade names other than its own. April 2013

1 About this Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Information Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Related Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

2 Configuration Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Accessing the Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 CLI Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Navigating CLI Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 The do Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Undoing Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Obtaining Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Entering and Editing Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Command History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Filtering show Command Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Multiple Users in Configuration mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3 Getting Started. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Console access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Serial console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 USB-B console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Default Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Configure a Host Name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Access the System Remotely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Access the C-Series and E-Series and the S60 Remotely . . . . . . . . . . . . . . . . . . . . 39 Access the S-Series Remotely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Configure the Enable Password . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Configuration File Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Copy Files to and from the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Save the Running-configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 View Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 File System Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 View command history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Upgrading and Downgrading FTOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

4 Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Configure Privilege Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Create a Custom Privilege Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Apply a Privilege Level to a Username . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Apply a Privilege Level to a Terminal Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

| 3

www.dell.com | support.dell.com

Configure Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Log Messages in the Internal Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Configuration Task List for System Log Management . . . . . . . . . . . . . . . . . . . . . . . . 54 Disable System Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Send System Messages to a Syslog Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Configure a Unix System as a Syslog Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Change System Logging Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Display the Logging Buffer and the Logging Configuration . . . . . . . . . . . . . . . . . . . . . . . 56 Configure a UNIX logging facility level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Synchronize log messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Enable timestamp on syslog messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 File Transfer Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Configuration Task List for File Transfer Services . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Terminal Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Deny and Permit Access to a Terminal Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Configure Login Authentication for Terminal Lines . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Time out of EXEC Privilege Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Telnet to Another Network Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Lock CONFIGURATION mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Viewing the Configuration Lock Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Recovering from a Forgotten Password on the S60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Recovering from a Forgotten Enable Password on the S60 . . . . . . . . . . . . . . . . . . . 68 Recovering from a Failed Start on the S60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

5 802.1ag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Ethernet CFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Maintenance Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Maintenance Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Maintenance End Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Configure CFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Enable Ethernet CFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Create a Maintenance Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Create a Maintenance Association . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Create Maintenance Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Create a Maintenance End Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Create a Maintenance Intermediate Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 MP Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Continuity Check Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Enable CCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Enable Cross-checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Loopback Message and Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Linktrace Message and Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Link Trace Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Enable CFM SNMP Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Display Ethernet CFM Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

6 Access Control Lists (ACL), Prefix Lists, and Route-maps . . . . . . . . . . . . . . . . . . 85

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 IP Access Control Lists (ACLs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 CAM Profiling, CAM Allocation, and CAM Optimization . . . . . . . . . . . . . . . . . . . . . . 86 Implementing ACLs on FTOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 IP Fragment Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Configure a standard IP ACL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Configure an extended IP ACL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Configuring Layer 2 and Layer 3 ACLs on an Interface . . . . . . . . . . . . . . . . . . . . . . . . . 98 Assign an IP ACL to an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Counting ACL Hits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Configuring Ingress ACLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Configuring Egress ACLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Egress Layer 3 ACL Lookup for Control-plane IP Traffic . . . . . . . . . . . . . . . . . . . . 102 Configuring ACLs to Loopback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Applying an ACL on Loopback Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 IP Prefix Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Configuration Task List for Prefix Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 ACL Resequencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Resequencing an ACL or Prefix List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Route Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Configuration Task List for Route Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

7 Border Gateway Protocol IPv4 (BGPv4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Autonomous Systems (AS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Sessions and Peers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Route Reflectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Confederations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 BGP Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Best Path Selection Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Local Preference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Multi-Exit Discriminators (MEDs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

| 5

www.dell.com | support.dell.com

AS Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Next Hop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Multiprotocol BGP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Implementing BGP with FTOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 4-Byte AS Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 AS4 Number Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 AS Number Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 BGP4 Management Information Base (MIB) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 BGP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Configuration Task List for BGP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 MBGP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 BGP Regular Expression Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Debugging BGP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Storing Last and Bad PDUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 Capturing PDUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 PDU Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Sample Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

8 Bare Metal Provisioning 2.0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 JumpStart mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 DHCP Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 File Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Domain Name Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Switch boot and set-up behavior in JumpStart Mode . . . . . . . . . . . . . . . . . . . . . . . 200

9 Content Addressable Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

Content Addressable Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 CAM Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .204 Microcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 CAM Profiling for ACLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Boot Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 When to Use CAM Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Select CAM Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 CAM Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Test CAM Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 View CAM Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 View CAM-ACL settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

View CAM Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Configure IPv4Flow Sub-partitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 Configure Ingress Layer 2 ACL Sub-partitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 Return to the Default CAM Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 CAM Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Applications for CAM Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 LAG Hashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 LAG Hashing based on Bidirectional Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 CAM profile for the VLAN ACL group feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Troubleshoot CAM Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 CAM Profile Mismatches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 QoS CAM Region Limitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

10 Debugging and Diagnostics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

Offline diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Running Offline Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Trace logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Auto Save on Crash or Rollover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Last restart reason (S60) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 show hardware commands (S60) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Hardware watchdog timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 Buffer tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 Deciding to tune buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 Buffer tuning commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Sample buffer profile configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Multicast Buffering on the S60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 Troubleshooting packet loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 Displaying Drop Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 Dataplane Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 Displaying Stack Port Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Displaying Stack Member Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Application core dumps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Mini core dumps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242

11 Dynamic Host Configuration Protocol (DHCP) . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 DHCP Packet Format and Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Assigning an IP Address using DHCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Configure the System to be a DHCP Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

| 7

www.dell.com | support.dell.com

Configure the Server for Automatic Address Allocation . . . . . . . . . . . . . . . . . . . . . 250 Specify a Default Gateway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Enable DHCP Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Configure a Method of Hostname Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252 Create Manual Binding Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Debug DHCP server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 DHCP Clear Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Configure the System to be a Relay Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 Configure the System for User Port Stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Configure Secure DHCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Option 82 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 DHCP Snooping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Drop DHCP packets on snooped VLANs only . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 Dynamic ARP Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Source Address Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

12 Force10 Resilient Ring Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 Ring Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 Multiple FRRP Rings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 Important FRRP Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Important FRRP Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Implementing FRRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 FRRP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Troubleshooting FRRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Configuration Checks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Sample Configuration and Topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

13 GARP VLAN Registration Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Configuring GVRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 Enabling GVRP Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 Enabling GVRP on a Layer 2 Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Configuring GVRP Registration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Configuring a GARP Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

14 Internet Group Management Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

IGMP Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 IGMP Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 IGMP version 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286

IGMP version 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 Configuring IGMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 Viewing IGMP Enabled Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 Selecting an IGMP Version . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 Viewing IGMP Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 Adjusting Timers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Adjusting Query and Response Timers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Adjusting the IGMP Querier Timeout Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Configuring a Static IGMP Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Enabling IGMP Immediate-leave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 IGMP Snooping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 IGMP Snooping Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 Configuring IGMP Snooping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 Enabling IGMP Immediate-leave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 Disabling Multicast Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Specifying a Port as Connected to a Multicast Router . . . . . . . . . . . . . . . . . . . . . . 295 Configuring the Switch as Querier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Fast Convergence after MSTP Topology Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296 Designating a Multicast Router Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296

15 Interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Interface Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 View Basic Interface Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Enable a Physical Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 Physical Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Configuration Task List for Physical Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Overview of Layer Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Configure Layer 2 (Data Link) Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Configure Layer 3 (Network) Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Management Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Configure Management Interfaces on the E-Series and C-Series and on the S60 . 304 Configure Management Interfaces on the S-Series . . . . . . . . . . . . . . . . . . . . . . . . 306 VLAN Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Loopback Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 Null Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 Port Channel Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Bulk Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Interface Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Bulk Configuration Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 Interface Range Macros . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Define the Interface Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 Choose an Interface-range Macro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324

| 9

www.dell.com | support.dell.com

Monitor and Maintain Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 Maintenance using TDR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 Link Debounce Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Important Points to Remember about Link Debounce Timer . . . . . . . . . . . . . . . . . 327 Assign a debounce time to an interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328 Show debounce times in an interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328 Disable ports when one only SFM is available (E300 only) . . . . . . . . . . . . . . . . . . 328 Disable port on one SFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Link Dampening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Enable Link Dampening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Ethernet Pause Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Threshold Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Enable Pause Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Configure MTU Size on an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 Port-pipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Auto-Negotiation on Ethernet Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336 View Advanced Interface Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 Display Only Configured Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 Configure Interface Sampling Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 Dynamic Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340

16 IPv4 Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343

IP Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 Configuration Task List for IP Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 Directed Broadcast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 Resolution of Host Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 ARP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 Configuration Task List for ARP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 ARP Learning via Gratuitous ARP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 ARP Learning via ARP Request . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354 Configurable ARP Retries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Configuring UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Important Points to Remember about UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Enabling UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 Configuring a Broadcast Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 Configurations Using UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 UDP Helper with Broadcast-all Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 UDP Helper with Subnet Broadcast Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . 358 UDP Helper with Configured Broadcast Addresses . . . . . . . . . . . . . . . . . . . . . . . . 358 UDP Helper with No Configured Broadcast Addresses . . . . . . . . . . . . . . . . . . . . . 359

10

Troubleshooting UDP Helper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

17 IPv6 Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362 Extended Address Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362 Stateless Autoconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362 IPv6 Headers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363 Extension Header fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365 Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 Implementing IPv6 with FTOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 Path MTU Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 IPv6 Neighbor Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 IPv6 Neighbor Discovery of MTU packets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 QoS for IPv6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372 IPv6 Multicast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372 SSH over an IPv6 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373 Configuration Task List for IPv6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373 Change your CAM-Profile on an E-Series system . . . . . . . . . . . . . . . . . . . . . . . . . 374 Adjust your CAM-Profile on an C-Series or S-Series . . . . . . . . . . . . . . . . . . . . . . . 375 Assign an IPv6 Address to an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376 Assign a Static IPv6 Route . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Telnet with IPv6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 SNMP over IPv6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 Show IPv6 Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 Show an IPv6 Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Show IPv6 Routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380 Show the Running-Configuration for an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 382 Clear IPv6 Routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382

18 iSCSI Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385

iSCSI Optimization Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385 Detection and Auto-configuration for Dell EqualLogic Arrays . . . . . . . . . . . . . . . . . 386 Detection and Port Configuration for Dell Compellent Arrays . . . . . . . . . . . . . . . . . 387 Enabling and Disabling iSCSI Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388 Default iSCSI Optimization Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 iSCSI Optimization Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 Configuring iSCSI Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 Displaying iSCSI Optimization Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

19 Link Aggregation Control Protocol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391

Introduction to Dynamic LAGs and LACP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392

| 11

www.dell.com | support.dell.com

LACP modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 LACP Configuration Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 LACP Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 Monitor and Debugging LACP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Shared LAG State Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Configure Shared LAG State Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 Important Points about Shared LAG State Tracking . . . . . . . . . . . . . . . . . . . . . . . . 398 Configure LACP as Hitless . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 LACP Basic Configuration Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

20 Layer 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
Managing the MAC Address Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 Clear the MAC Address Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 Set the Aging Time for Dynamic Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410 Configure a Static MAC Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410 Display the MAC Address Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 MAC Learning Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 mac learning-limit dynamic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 mac learning-limit station-move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 mac learning-limit no-station-move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 Learning Limit Violation Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 Station Move Violation Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 Recovering from Learning Limit and Station Move Violations . . . . . . . . . . . . . . . . . 414 Per-VLAN MAC Learning Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414 NIC Teaming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416 MAC Move Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417 Microsoft Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417 Default Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 418 Configuring the Switch for Microsoft Server Clustering . . . . . . . . . . . . . . . . . . . . . . 419 Enable and Disable VLAN Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 Configuring Redundant Pairs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420 Important Points about Configuring Redundant Pairs . . . . . . . . . . . . . . . . . . . . . . . 420 Restricting Layer 2 Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422 Far-end Failure Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422 FEFD state changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424 Configuring FEFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424 Debugging FEFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426

21 Link Layer Discovery Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429

802.1AB (LLDP) Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429 Protocol Data Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429 Optional TLVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430

12

Management TLVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431 TIA-1057 (LLDP-MED) Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 TIA Organizationally Specific TLVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 Configuring LLDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 LLDP Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 CONFIGURATION versus INTERFACE Configurations . . . . . . . . . . . . . . . . . . . . . . . . 437 Enabling LLDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438 Disabling and Undoing LLDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438 Advertising TLVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438 Viewing the LLDP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440 Viewing Information Advertised by Adjacent LLDP Agents . . . . . . . . . . . . . . . . . . . . . . 440 Configuring LLDPDU Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Configuring Transmit and Receive Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442 Configuring a Time to Live . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 Debugging LLDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444 Relevant Management Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

22 Multiple Spanning Tree Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452 Configure Multiple Spanning Tree Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452 Enable Multiple Spanning Tree Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 Add and Remove Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 Create Multiple Spanning Tree Instances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 Influence MSTP Root Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 Interoperate with Non-FTOS Bridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 Modify Global Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 Modify Interface Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457 Configure an EdgePort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458 Flush MAC Addresses after a Topology Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 MSTP Sample Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460 Debugging and Verifying MSTP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464

23 Multicast Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467

Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467 Enable IP Multicast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467 Multicast with ECMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 First Packet Forwarding for Lossless Multicast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 Multicast Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470

| 13

www.dell.com | support.dell.com

IPv4 Multicast Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 IPv6 Multicast Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475 Multicast Traceroute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477 Multicast Quality of Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477 Optimize the E-Series for Multicast Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478 Allocate More Buffer Memory for Multicast WRED . . . . . . . . . . . . . . . . . . . . . . . . . 478 Allocate More Bandwidth to Multicast using Egress WFQ . . . . . . . . . . . . . . . . . . . 478 Tune the Central Scheduler for Multicast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478

24 Open Shortest Path First (OSPFv2 and OSPFv3) . . . . . . . . . . . . . . . . . . . . . . . 481

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482 Autonomous System (AS) Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482 Networks and Neighbors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484 Router Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484 Designated and Backup Designated Routers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486 Link-State Advertisements (LSAs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 487 Virtual Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 Router Priority and Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 Implementing OSPF with FTOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 Graceful Restart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 Fast Convergence (OSPFv2, IPv4 only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 Multi-Process OSPF (OSPFv2, IPv4 only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 RFC-2328 Compliant OSPF Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 OSPF ACK Packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492 OSPF Adjacency with Cisco Routers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492 Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493 Configuration Task List for OSPFv2 (OSPF for IPv4) . . . . . . . . . . . . . . . . . . . . . . . 493 Troubleshooting OSPFv2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 Configuration Task List for OSPFv3 (OSPF for IPv6) . . . . . . . . . . . . . . . . . . . . . . . 513 Troubleshooting OSPFv3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518 Sample Configurations for OSPFv2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 Basic OSPFv2 Router Topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519

25 PIM Sparse-Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521

Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 Requesting Multicast Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522 Refusing Multicast Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522 Sending Multicast Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523 Configure PIM-SM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523 Enable PIM-SM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524

14

Configurable S,G Expiry Timers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525 Configure a Static Rendezvous Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526 Override Bootstrap Router Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527 Configure a Designated Router . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527 Create Multicast Boundaries and Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528 PIM-SM Graceful Restart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528 Monitoring PIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529

26 PIM Source-Specific Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531

Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 Configure PIM-SM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 Enable PIM-SSM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 Use PIM-SSM with IGMP version 2 Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534

27 Port Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539

Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539 Port Monitoring on E-Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540 E-Series TeraScale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540 E-Series ExaScale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541 Port Monitoring on C-Series and S-Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541 Configuring Port Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544 Flow-based Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546

28 Private VLANs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549

Private VLAN Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 550 Private VLAN Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551 Private VLAN Configuration Task List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552 Private VLAN Configuration Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555

29 Per-VLAN Spanning Tree Plus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560 Configure Per-VLAN Spanning Tree Plus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560 Enable PVST+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561 Disable PVST+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561 Influence PVST+ Root Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561 Modify Global PVST+ Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563 Modify Interface PVST+ Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564 Configure an EdgePort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565

| 15

www.dell.com | support.dell.com

PVST+ in Multi-vendor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566 PVST+ Extended System ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566 PVST+ Sample Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567

30 Quality of Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 571

Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 Port-based QoS Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 Set dot1p Priorities for Incoming Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574 Honor dot1p Priorities on Ingress Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574 Configure Port-based Rate Policing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575 Configure Port-based Rate Limiting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 576 Configure Port-based Rate Shaping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577 Policy-based QoS Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577 Classify Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 578 Create a QoS Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582 Create Policy Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584 QoS Rate Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589 Strict-priority Queueing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 590 Weighted Random Early Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 590 Create WRED Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591 Apply a WRED profile to traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591 Configure WRED for Storm Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591 Display Default and Configured WRED Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 Display WRED Drop Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 Allocating Bandwidth to Multicast Queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593 Pre-calculating Available QoS CAM Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594 Viewing QoS CAM Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595 Configure Quality of Service for an Office VOIP Deployment . . . . . . . . . . . . . . . . . . . . 595 Honor the incoming DSCP value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595 Honor the incoming dot1p value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 596 Classifying VOIP traffic and applying QoS policies . . . . . . . . . . . . . . . . . . . . . . . . . 596

31 Routing Information Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 599

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 599 RIPv1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 599 RIPv2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600 Configuration Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600 Configuration Task List for RIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 601 RIP Configuration Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 608

16

32 Remote Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 615

Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 615 Fault Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 616

33 Rapid Spanning Tree Protocol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621 Configuring Rapid Spanning Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 622 Configure Interfaces for Layer 2 Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 622 Enable Rapid Spanning Tree Protocol Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623 Add and Remove Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626 Modify Global Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626 Modify Interface Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627 Configure an EdgePort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 628 Influence RSTP Root Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629 SNMP Traps for Root Elections and Topology Changes . . . . . . . . . . . . . . . . . . . . . . . . 630 Fast Hellos for Link State Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630

34 Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
AAA Accounting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631 Configuration Task List for AAA Accounting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 632 AAA Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634 Configuration Task List for AAA Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634 AAA Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637 Privilege Levels Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637 Configuration Task List for Privilege Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 638 RADIUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643 RADIUS Authentication and Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643 Configuration Task List for RADIUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 644 TACACS+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647 Configuration Task List for TACACS+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647 TACACS+ Remote Authentication and Authorization . . . . . . . . . . . . . . . . . . . . . . . 649 Command Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651 Protection from TCP Tiny and Overlapping Fragment Attacks . . . . . . . . . . . . . . . . . . . 651 SCP and SSH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651 Using SCP with SSH to copy a software image . . . . . . . . . . . . . . . . . . . . . . . . . . . 653 Secure Shell Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654 Troubleshooting SSH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 657 Telnet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 657 Trace Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658 Configuration Tasks for Trace Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658

| 17

www.dell.com | support.dell.com

VTY Line and Access-Class Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664 VTY Line Local Authentication and Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . 664 VTY Line Remote Authentication and Authorization . . . . . . . . . . . . . . . . . . . . . . . . 665 VTY MAC-SA Filter Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665

35 Service Provider Bridging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667

VLAN Stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668 Configure VLAN Stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668 Create Access and Trunk Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669 Enable VLAN-Stacking for a VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669 Configure the Protocol Type Value for the Outer VLAN Tag . . . . . . . . . . . . . . . . . . 670 FTOS Options for Trunk Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671 Debug VLAN Stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672 VLAN Stacking in Multi-vendor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672 VLAN Stacking Packet Drop Precedence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 678 Enable Drop Eligibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 Honor the Incoming DEI Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 Mark Egress Packets with a DEI Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 680 Dynamic Mode CoS for VLAN Stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 680 Layer 2 Protocol Tunneling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685 Enable Layer 2 Protocol Tunneling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685 Specify a Destination MAC Address for BPDUs . . . . . . . . . . . . . . . . . . . . . . . . . . . 686 Rate-limit BPDUs on the E-Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 686 Rate-limit BPDUs on the C-Series and S-Series . . . . . . . . . . . . . . . . . . . . . . . . . . 686 Debug Layer 2 Protocol Tunneling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 687 Provider Backbone Bridging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 687

36 sFlow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690 Enable and Disable sFlow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 Enable and Disable on an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 sFlow Show Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692 Show sFlow Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692 Show sFlow on an Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692 Show sFlow on a Line Card . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693 Specify Collectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 Polling Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .694 Sampling Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 Sub-sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695

18

Back-off Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696 sFlow on LAG ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696 Extended sFlow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697

37 Simple Network Management Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 Configure Simple Network Management Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700 Create a Community . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700 Read Managed Object Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701 Write Managed Object Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 702 Configure Contact and Location Information using SNMP . . . . . . . . . . . . . . . . . . . . . . 702 Subscribe to Managed Object Value Updates using SNMP . . . . . . . . . . . . . . . . . . . . . 703 Copy Configuration Files Using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706 Manage VLANs using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712 Create a VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712 Assign a VLAN Alias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712 Display the Ports in a VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 713 Add Tagged and Untagged Ports to a VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715 Enable and Disable a Port using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716 Fetch Dynamic MAC Entries using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716 Deriving Interface Indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 718 Viewing the System Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 719 Viewing Alarm LED Status Using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 720 Using SNMP for Entity MIB Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 721

38 Spanning Tree Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723

Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723 Configuring Spanning Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723 Related Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 724 Configuring Interfaces for Layer 2 Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 725 Enabling Spanning Tree Protocol Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 726 Adding an Interface to the Spanning Tree Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 728 Removing an Interface from the Spanning Tree Group . . . . . . . . . . . . . . . . . . . . . . . . . 728 Modifying Global Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 729 Modifying Interface STP Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 730 Enabling PortFast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 730 Preventing Network Disruptions with BPDU Guard . . . . . . . . . . . . . . . . . . . . . . . . 731 STP Root Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 733

| 19

www.dell.com | support.dell.com

SNMP Traps for Root Elections and Topology Changes . . . . . . . . . . . . . . . . . . . . . . . . 734 Configuring Spanning Trees as Hitless . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 734

39 Stacking S-Series Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735

S-Series Stacking Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735 High Availability on S-Series Stacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735 MAC Addressing on S-Series Stacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 737 Management Access on S-Series Stacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 741 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742 S-Series Stacking Installation Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742 Create an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742 Add a Unit to an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745 Remove a Unit from an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 748 Replace a Stack Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750 Merge Two S-Series Stacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 751 Split an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 751 S-Series Stacking Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 752 Assign Unit Numbers to Units in an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . 752 Create a Virtual Stack Unit on an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . 752 Display Information about an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 752 Influence Management Unit Selection on an S-Series Stack . . . . . . . . . . . . . . . . . 755 Manage Redundancy on an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 755 Reset a Unit on an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 755 Monitor an S-Series Stack with SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756 Troubleshoot an S-Series Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756 Recover from Stack Link Flaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756 Recover from a Card Problem State on an S-Series Stack . . . . . . . . . . . . . . . . . . . 757 Recover from a Card Mismatch State on an S-Series Stack . . . . . . . . . . . . . . . . . . 757

40 Storm Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 759

Configure Storm Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 759 Configure storm control from INTERFACE mode . . . . . . . . . . . . . . . . . . . . . . . . . . 759 Configure storm control from CONFIGURATION mode . . . . . . . . . . . . . . . . . . . . . 760

41 System Time and Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761

Network Time Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761 Protocol Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762 Implementation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763 Configuring Network Time Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763 Enable NTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764 Set the Hardware Clock with the Time Derived from NTP . . . . . . . . . . . . . . . . . . . 765 Configure NTP broadcasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765

20

Disable NTP on an interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765 Configure a source IP address for NTP packets . . . . . . . . . . . . . . . . . . . . . . . . . . . 766 Configure NTP authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767 FTOS Time and Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769 Configuring time and date settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769 Set daylight savings time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 772

42 Uplink Failure Detection (UFD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 777

Feature Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 777 How Uplink Failure Detection Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 778 UFD and NIC Teaming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779 Important Points to Remember . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 780 Configuring Uplink Failure Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 781 Clearing a UFD-Disabled Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 782 Displaying Uplink Failure Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 784 Sample Configuration: Uplink Failure Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 787

43 Upgrade Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789

Find the upgrade procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789 Get Help with upgrades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789

44 Virtual LANs (VLAN) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 791

Default VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 792 Port-Based VLANs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 793 VLANs and Port Tagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 793 Configuration Task List for VLANs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 794 VLAN Interface Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798 Native VLANs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798 Enable Null VLAN as the Default VLAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 799

45 Virtual Router Redundancy Protocol (VRRP) . . . . . . . . . . . . . . . . . . . . . . . . . . . 801

VRRP Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 801 VRRP Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803 VRRP Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803 VRRP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804 Configuration Task List for VRRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804 Sample Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 812

46 Standards Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815

IEEE Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815 RFC and I-D Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 816 MIB Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .829

| 21

www.dell.com | support.dell.com

22

1
About this Guide
Objectives
This guide describes the protocols and features supported by the Dell Networking operating system (FTOS) and provides configuration instructions and examples for implementing them. It supports the system platforms E-Series, C-Series, and S-Series. Though this guide contains information on protocols, it is not intended to be a complete reference. This guide is a reference for configuring protocols on Dell Networking systems. For complete information on protocols, refer to other documentation including IETF Requests for Comment (RFCs). The instructions in this guide cite relevant RFCs, and Chapter 46 contains a complete list of the supported RFCs and Management Information Base files (MIBs).

Audience
This document is intended for system administrators who are responsible for configuring and maintaining networks and assumes you are knowledgeable in Layer 2 and Layer 3 networking technologies.

About this Guide | 23

www.dell.com | support.dell.com

Conventions
This document uses the following conventions to describe command syntax:
Convention
keyword parameter

Description Keywords are in bold and should be entered in the CLI as listed. Parameters are in italics and require a number or word to be entered in the CLI. Keywords and parameters within braces must be entered in the CLI. Keywords and parameters within brackets are optional. Keywords and parameters separated by bar require you to choose one.

{X} [X] x|y

Information Symbols
Table 1-1 describes symbols contained in this guide.
Table 1-1. Symbol Information Symbols Warning Note FTOS Behavior Description This symbol informs you of important operational information. This symbol informs you of an FTOS behavior. These behaviors are inherent to the Dell Networking system or FTOS feature and are non-configurable. This symbol informs you of a feature that supported on one or two platforms only: e is for E-Series, c is for C-Series, s is for S-Series. If a feature or command applies to only one of the E-Series platforms, a separate symbol calls this to attention: et for the TeraScale or e x for the ExaScale. This symbol is a note associated with some other text on the page that is marked with an asterisk.

ces
et ex

Platform Specific Feature E-Series Specific Feature/Command Exception

Related Documents
For more information about the Dell Networking E-Series, C-Series, and S-Series refer to the following documents: FTOS Command Reference Installing and Maintaining the S60 System FTOS Release Notes

24

About this Guide

2
Configuration Fundamentals
The FTOS Command Line Interface (CLI) is a text-based interface through which you can configure interfaces and protocols. The CLI is largely the same for the E-Series, C-Series, and S-Series with the exception of some commands and command outputs. The CLI is structured in modes for security and management purposes. Different sets of commands are available in each mode, and you can limit user access to modes using privilege levels. In FTOS, after a command is enabled, it is entered into the running configuration file. You can view the current configuration for the whole system or for a particular CLI mode. To save the current configuration copy the running configuration to another location.
Note: Due to a differences in hardware architecture and the continued system development, features may occasionally differ between the platforms. These differences are identified by the information symbols shown on Table 1-1 on page 24.

Accessing the Command Line

Access the command line through a serial console port or a Telnet session (Figure 2-1). When the system successfully boots, you enter the command line in the EXEC mode.
Note: You must have a password configured on a virtual terminal line before you can Telnet into the system. Therefore, you must use a console connection when connecting to the system for the first time. Figure 2-1. Logging into the System using Telnet

telnet 172.31.1.53 Trying 172.31.1.53... Connected to 172.31.1.53. Escape character is '^]'. Login: username EXEC mode Password:

prompt

Configuration Fundamentals | 25

www.dell.com | support.dell.com

CLI Modes
Different sets of commands are available in each mode. A command found in one mode cannot be executed from another mode (with the exception of EXEC mode commands preceded by the command do; see The do Command). You can set user access rights to commands and command modes using privilege levels; for more information on privilege levels and security options, refer to Chapter 9, Security, on page 627. The FTOS CLI is divided into three major mode levels: EXEC mode is the default mode and has a privilege level of 1, which is the most restricted level. Only a limited selection of commands is available, notably show commands, which allow you to view system information. EXEC Privilege mode has commands to view configurations, clear counters, manage configuration files, run diagnostics, and enable or disable debug operations. The privilege level is 15, which is unrestricted. You can configure a password for this mode; see Configure the Enable Password. CONFIGURATION mode enables you to configure security features, time settings, set logging and SNMP functions, configure static ARP and MAC addresses, and set line cards on the system.

Beneath CONFIGURATION mode are sub-modes that apply to interfaces, protocols, and features. Figure 2-2 illustrates this sub-mode command structure. Two sub-CONFIGURATION modes are important when configuring the chassis for the first time: INTERFACE sub-mode is the mode in which you configure Layer 2 and Layer 3 protocols and IP services specific to an interface. An interface can be physical (Management interface, 1-Gigabit Ethernet, or 10-Gigabit Ethernet, or SONET) or logical (Loopback, Null, port channel, or VLAN). LINE sub-mode is the mode in which you to configure the console and virtual terminal lines.
Note: At any time, entering a question mark (?) will display the available command options. For example, when you are in CONFIGURATION mode, entering the question mark first will list all available commands, including the possible sub-modes.

26

Configuration Fundamentals

Figure 2-2.

CLI Modes in FTOS

EXEC EXEC Privilege CONFIGURATION ARCHIVE AS-PATH ACL INTERFACE GIGABIT ETHERNET 10 GIGABIT ETHERNET INTERFACE RANGE LOOPBACK MANAGEMENT ETHERNET NULL PORT-CHANNEL SONET VLAN VRRP IP IPv6 IP COMMUNITY-LIST IP ACCESS-LIST STANDARD ACCESS-LIST EXTENDED ACCESS-LIST LINE AUXILIARY CONSOLE VIRTUAL TERMINAL MAC ACCESS-LIST MONITOR SESSION MULTIPLE SPANNING TREE Per-VLAN SPANNING TREE PREFIX-LIST RAPID SPANNING TREE REDIRECT ROUTE-MAP ROUTER BGP ROUTER ISIS ROUTER OSPF ROUTER RIP SPANNING TREE TRACE-LIST

Navigating CLI Modes

The FTOS prompt changes to indicate the CLI mode. Table 2-1 lists the CLI mode, its prompt, and information on how to access and exit this CLI mode. You must move linearly through the command modes, with the exception of the end command which takes you directly to EXEC Privilege mode; the exit command moves you up one command mode level.
Note: Sub-CONFIGURATION modes all have the letters conf in the prompt with additional modifiers to identify the mode and slot/port information. These are shown in Table 2-1.

Configuration Fundamentals | 27

www.dell.com | support.dell.com

Table 2-1.

FTOS Command Modes

Prompt

CLI Command Mode

Access Command Access the router through the console or Telnet. From EXEC mode, enter the command enable. From any other mode, use the command end. From EXEC privilege mode, enter the command configure. From every mode except EXEC and EXEC Privilege, enter the command exit.

EXEC EXEC Privilege CONFIGURATION

FTOS> FTOS# FTOS(conf)#

Note: Access all of the following modes from CONFIGURATION mode. ARCHIVE AS-PATH ACL Gigabit Ethernet Interface 10 Gigabit Ethernet Interface INTERFACE modes Interface Range Loopback Interface Management Ethernet Interface Null Interface Port-channel Interface SONET Interface VLAN Interface IP ACCESS-LIST STANDARD ACCESS- LIST EXTENDED ACCESS- LIST FTOS(conf-archive) FTOS(config-as-path)# FTOS(conf-if-gi-0/0)# FTOS(conf-if-te-0/0)# FTOS(conf-if-range)# FTOS(conf-if-lo-0)# FTOS(conf-if-ma-0/0)# FTOS(conf-if-nu-0)# FTOS(conf-if-po-0)# FTOS(conf-if-so-0/0)# FTOS(conf-if-vl-0)# FTOS(config-std-nacl)# FTOS(config-ext-nacl)#
ip access-list extended ip access-list standard interface archive ip as-path access-list

IP COMMUNITY-LIST AUXILIARY LINE CONSOLE VIRTUAL TERMINAL

FTOS(config-community-list)# FTOS(config-line-aux)# FTOS(config-line-console)# FTOS(config-line-vty)#

ip community-list

line

28

Configuration Fundamentals

Table 2-1.

FTOS Command Modes

Prompt

CLI Command Mode

Access Command
mac access-list standard

MAC ACCESS-LIST

STANDARD ACCESS- LIST EXTENDED ACCESS- LIST

FTOS(config-std-macl)# FTOS(config-ext-macl)#

mac access-list extended

MULTIPLE SPANNING TREE Per-VLAN SPANNING TREE Plus PREFIX-LIST RAPID SPANNING TREE REDIRECT ROUTE-MAP ROUTER BGP ROUTER ISIS ROUTER OSPF ROUTER RIP SPANNING TREE TRACE-LIST

FTOS(config-mstp)# FTOS(config-pvst)#

protocol spanning-tree mstp

protocol spanning-tree pvst

FTOS(conf-nprefixl)# FTOS(config-rstp)# FTOS(conf-redirect-list)# FTOS(config-route-map)# FTOS(conf-router_bgp)# FTOS(conf-router_isis)# FTOS(conf-router_ospf)# FTOS(conf-router_rip)# FTOS(config-span)# FTOS(conf-trace-acl)#

ip prefix-list protocol spanning-tree rstp ip redirect-list route-map router bgp router isis router ospf router rip protocol spanning-tree 0 ip trace-list

Figure 2-3 illustrates how to change the command mode from CONFIGURATION mode to PROTOCOL SPANNING TREE.
Figure 2-3. Changing CLI Modes

FTOS(conf)#protocol spanning-tree 0 FTOS(config-span)# New command

prompt

Configuration Fundamentals | 29

www.dell.com | support.dell.com

The do Command
Enter an EXEC mode command from any CONFIGURATION mode (CONFIGURATION, INTERFACE, SPANNING TREE, etc.) without returning to EXEC mode by preceding the EXEC mode command with the command do. Figure 2-4 illustrates the do command.
Note: The following commands cannot be modified by the do command: enable, disable, exit, and configure. Figure 2-4. Using the do Command

FTOS(conf)#do show linecard all

do form of show command

-- Line cards -Slot Status NxtBoot ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 not present 1 not present 2 online online E48TB E48TB 1-1-463 48 3 not present 4 not present 5 online online E48VB E48VB 1-1-463 48

Undoing Commands
When you enter a command, the command line is added to the running configuration file. Disable a command and remove it from the running-config by entering the original command preceded by the command no. For example, to delete an ip address configured on an interface, use the no ip address ip-address command, as shown in Figure 2-5.
Note: Use the help or ? command as discussed in Obtaining Help command to help you construct the no form of a command. Figure 2-5. Undoing a command with the no Command

FTOS(conf)#interface gigabitethernet 4/17 FTOS(conf-if-gi-4/17)#ip address 192.168.10.1/24 FTOS(conf-if-gi-4/17)#show config ! IP address assigned interface GigabitEthernet 4/17 ip address 192.168.10.1/24 no form of no shutdown FTOS(conf-if-gi-4/17)#no ip address FTOS(conf-if-gi-4/17)#show config IP address removed ! interface GigabitEthernet 4/17

IP address command

Layer 2 protocols are disabled by default. Enable them using the no disable command. For example, in PROTOCOL SPANNING TREE mode, enter no disable to enable Spanning Tree.

30

Configuration Fundamentals

Obtaining Help
Obtain a list of keywords and a brief functional description of those keywords at any CLI mode using the ? or help command: Enter ? at the prompt or after a keyword to list the keywords available in the current mode. ? after a prompt lists all of the available keywords. The output of this command is the same for the help command.
? Command Example

Figure 2-6.
FTOS#? calendar cd change clear clock configure copy debug --More--

? at prompt for list of commands

Manage the hardware calendar Change current directory Change subcommands Reset functions Manage the system clock Configuring from terminal Copy from one file to another Debug functions

after a partial keyword lists all of the keywords that begin with the specified letters.
Keyword? Command Example

Figure 2-7.

FTOS(conf)#cl? class-map clock FTOS(conf)#cl

partial keyword plus [space]? for matching keywords

A keyword followed by [space]? lists all of the keywords that can follow the specified keyword.
Keyword ? Command Example

Figure 2-8.

FTOS(conf)#clock ? summer-time timezone FTOS(conf)#clock

keyword plus [space]? for compatible keywords

Configure summer (daylight savings) time Configure time zone

Entering and Editing Commands

When entering commands: The CLI is not case sensitive. You can enter partial CLI keywords. You must enter the minimum number of letters to uniquely identify a command. For example, cl cannot be entered as a partial keyword because both the clock and class-map commands begin with the letters cl. clo, however, can be entered as a partial keyword because only one command begins with those three letters. The TAB key auto-completes keywords in commands. You must enter the minimum number of letters to uniquely identify a command.

Configuration Fundamentals | 31

www.dell.com | support.dell.com

Table 2-2.

The UP and DOWN arrow keys display previously entered commands (see Command History). The BACKSPACE and DELETE keys erase the previous letter. Key combinations are available to move quickly across the command line, as described in Table 2-2.
Short-Cut Keys and their Actions
Action

Key Combination

CNTL-A CNTL-B CNTL-D CNTL-E CNTL-F CNTL-I CNTL-K CNTL-L CNTL-N CNTL-P CNTL-R CNTL-U CNTL-W CNTL-X CNTL-Z Esc B Esc F Esc D

Moves the cursor to the beginning of the command line. Moves the cursor back one character. Deletes character at cursor. Moves the cursor to the end of the line. Moves the cursor forward one character. Completes a keyword. Deletes all characters from the cursor to the end of the command line. Re-enters the previous command. Return to more recent commands in the history buffer after recalling commands with CTRL-P or the UP arrow key. Recalls commands, beginning with the last command Re-enters the previous command. Deletes the line. Deletes the previous word. Deletes the line. Ends continuous scrolling of command outputs. Moves the cursor back one word. Moves the cursor forward one word. Deletes all characters from the cursor to the end of the word.

Command History
FTOS maintains a history of previously-entered commands for each mode. For example: When you are in EXEC mode, the UP and DOWN arrow keys display the previously-entered EXEC mode commands. When you are in CONFIGURATION mode, the UP or DOWN arrows keys recall the previously-entered CONFIGURATION mode commands.

32

Configuration Fundamentals

Filtering show Command Outputs

Filter the output of a show command to display specific information by adding | [except | find | grep | no-more | save] specified_text after the command. The variable specified_text is the text for which you are filtering and it IS case sensitive unless the ignore-case sub-option is implemented. Starting with FTOS 7.8.1.0, the grep command accepts an ignore-case sub-option that forces the search to case-insensitive. For example, the commands: returns a search result with instances containing a capitalized Ethernet, such as interface GigabitEthernet 0/0. show run | grep ethernet would not return that search result because it only searches for instances containing a non-capitalized ethernet.
show run | grep Ethernet

Executing the command show run | grep Ethernet ignore-case would return instances containing both Ethernet and ethernet.
grep displays only the lines containing specified text. Figure 2-9 shows this command used in combination with the command show linecard all.

Figure 2-9.

Filtering Command Outputs with the grep Command

FTOS(conf)#do show linecard all | grep 0 0 not present

Note: FTOS accepts a space or no space before and after the pipe. To filter on a phrase with spaces, underscores, or ranges, enclose the phrase with double quotation marks.

except displays text that does not match the specified text. Figure 2-10 shows this command used in

combination with the command show linecard all.

Figure 2-10. Filtering Command Outputs with the except Command
FTOS#show linecard all | except 0 -- Line cards -Slot Status NxtBoot ReqTyp CurTyp Version Ports --------------------------------------------------------------------------2 not present 3 not present 4 not present 5 not present 6 not present

Configuration Fundamentals | 33

www.dell.com | support.dell.com

find displays the output of the show command beginning from the first occurrence of specified text

Figure 2-11 shows this command used in combination with the command show linecard all.
Figure 2-11. Filtering Command Outputs with the find Command

FTOS(conf)#do show linecard all | find 0 0 not present 1 not present 2 online online E48TB E48TB 3 not present 4 not present 5 online online E48VB E48VB 6 not present 7 not present

1-1-463

48

1-1-463

48

display displays additional configuration information. no-more displays the output all at once rather than one screen at a time. This is similar to the command terminal length except that the no-more option affects the output of the specified command only. save copies the output to a file for future reference.

Note: You can filter a single command output multiple times. The save option should be the last option entered. For example: FTOS# command | grep regular-expression | except regular-expression | grep other-regular-expression | find regular-expression | save

Multiple Users in Configuration mode

FTOS notifies all users in the event that there are multiple users logged into CONFIGURATION mode. A warning message indicates the username, type of connection (console or vty), and in the case of a vty connection, the IP address of the terminal on which the connection was established. For example: On the system that telnets into the switch, Message 1 appears:

Message 1 Multiple Users in Configuration mode Telnet Message

% Warning: The following users are currently configuring the system: User "<username>" on line console0

On the system that is connected over the console, Message 2 appears:

Message 2 Multiple Users in Configuration mode Telnet Message

% Warning: User "<username>" on line vty0 "10.11.130.2" is in configuration mode

If either of these messages appears, Dell Networking recommends that you coordinate with the users listed in the message so that you do not unintentionally overwrite each others configuration changes.

34

Configuration Fundamentals

3
Getting Started
This chapter contains the following major sections: Default Configuration Configure a Host Name Access the System Remotely Configure the Enable Password Configuration File Management File System Management

When you power up the chassis, the system performs a Power-On Self Test (POST) during which Route Processor Module (RPM), Switch Fabric Module (SFM), and line card status LEDs blink green.The system then loads FTOS and boot messages scroll up the terminal window during this process. No user interaction is required if the boot process proceeds without interruption. When the boot process is complete, the RPM and line card status LEDs remain online (green), and the console monitor displays the EXEC mode prompt. For details on using the Command Line Interface (CLI), see the Accessing the Command Line section in Chapter 2, Configuration Fundamentals.

Console access
The S60 has 2 management ports available for system access: a serial console port and a USB-B port. The USB-B ports acts exactly as the console port. By default, the S60 sends system messages to the serial console port. However, only one console connection can be active at a time. When both the serial console port and the USB-B port are connected, the system defaults to the USB-B port. The console connection is considered inactive if the USB-B port is also connected. You can switch between console connections by physically connecting or disconnecting the cables. A system message is displayed on the serial console prior to switching to the USB-B console. When the USB-B cable is detached, the system returns to the serial console default.

Getting Started | 35

www.dell.com | support.dell.com

Serial console
The RS-232 console port is labeled on the S60 chassis. It is in the upper right-hand side, as you face the rear of the chassis.

Console Port

To access the console port, follow the procedures below. Refer to Table 3-1 for the console port pinout.
Step Task

1 2 3

Install an RJ-45 copper cable into the console port.Use a rollover (crossover) cable to connect the S60 console port to a terminal server. Connect the other end of the cable to the DTE terminal server. Terminal settings on the console port cannot be changed in the software and are set as follows: 9600 baud rate No parity 8 data bits 1 stop bit No flow control

Accessing the RJ-45 console port with a DB-9 adapter

You can connect to the console using a RJ-45 to RJ-45 rollover cable and a RJ-45 to DB-9 female DTE adapter to a terminal server (for example, PC). Table 3-1 lists the pin assignments.
Table 3-1. Pin Assignments Between the E-Series Console and a DTE Terminal Server RJ-45 to DB-9 Adapter DB-9 Pin 8 6 2 5 5 3 Terminal Server Device Signal CTS DSR RxD GND GND TxD

E-Series Console Port Signal RTS NC TxD GND GND RxD

RJ-45 to RJ-45 Rollover Cable RJ-45 pinout 1 2 3 4 5 6 RJ-45 Pinout 8 7 6 5 4 3

36

Getting Started

Table 3-1.

Pin Assignments Between the E-Series Console and a DTE Terminal Server (continued) RJ-45 to DB-9 Adapter DB-9 Pin 4 7 Terminal Server Device Signal DTR RTS

E-Series Console Port Signal NC CTS

RJ-45 to RJ-45 Rollover Cable RJ-45 pinout 7 8 RJ-45 Pinout 2 1

USB-B console
The USB-B connector port is labeled on the the S60 chassis. It is to the left of the management ports, as you face the rear of the chassis

USB-B Port

Step

Task

1 2 3 4 5 6

Power on the PC (XP operating system recommended) Connect the USB-A end of cable (supplied) into an available USB port on the PC Connect the USB-B end of cable into the USB-B console port on the S60 (see Figure 7) Power on the S60. Install necessary USB device drivers (internet connection required). Contact Dell Networking TAC for CD, if necessary. Open your terminal software emulation program to access the S60.

Getting Started | 37

www.dell.com | support.dell.com

Step

Task (continued)

Using the terminal settings shown here, set the terminal connection settings. 9600 baud rate, No parity, 8 data bits, 1 stop bit, No flow control

You will see the message below when you are connected to the S60.

Default Configuration
A version of FTOS is pre-loaded onto the chassis, however the system is not configured when you power up for the first time (except for the default hostname, which is FTOS). You must configure the system using the CLI.

38

Getting Started

Configure a Host Name

The host name appears in the prompt. The default host name is FTOS. Host names must start with a letter and end with a letter or digit. Characters within the string can be letters, digits, and hyphens.

To configure a host name:

Step 1 Task Create a new host name. Command Syntax
hostname name

Command Mode CONFIGURATION

Figure 3-1 illustrates the hostname command.

Figure 3-1. Configuring a Hostname

Default Hostname
FTOS(conf)#hostname R1 R1(conf)#

New Hostname

Access the System Remotely

You can configure the system to access it remotely by Telnet. The method for configuring the C-Series and E-Series for Telnet access is different from S-Series. The C-Series, E-Series and the S60 have a dedicated management port and a management routing table that is separate from the IP routing table. The S-Series (except the S60) does not have a dedicated management port, but is managed from any port. It does not have a separate management routing table.

Access the C-Series and E-Series and the S60 Remotely

Configuring the system for Telnet is a three-step process: 1. Configure an IP address for the management port. See Configure the Management Port IP Address. 2. Configure a management route with a default gateway. See Configure a Management Route. 3. Configure a username and password. See Configure a Username and Password.

Getting Started | 39

www.dell.com | support.dell.com

Configure the Management Port IP Address

Assign IP addresses to the management ports in order to access the system remotely.

Note: Assign different IP addresses to each RPMs management port.

To configure the management port IP address:

Step 1 Task Enter INTERFACE mode for the Management port. Command Syntax
interface ManagementEthernet slot/port

Command Mode CONFIGURATION

slot range: 0 to 1 port range: 0

Assign an IP address to the interface.

ip address ip-address/mask

INTERFACE

ip-address: an address in dotted-decimal format (A.B.C.D). mask: a subnet mask in /prefix-length format (/ xx).

Enable the interface.

no shutdown

INTERFACE

Configure a Management Route

Define a path from the system to the network from which you are accessing the system remotely. Management routes are separate from IP routes and are only used to manage the system through the management port. To configure a management route:
Step 1 Task Configure a management route to the network from which you are accessing the system. Command Syntax
management route ip-address/mask gateway

Command Mode CONFIGURATION

ip-address: the network address in dotted-decimal format (A.B.C.D). mask: a subnet mask in /prefix-length format (/ xx). gateway: the next hop for network traffic originating from the management port.

Configure a Username and Password

Configure a system username and password to access the system remotely.

40

Getting Started

To configure a username and password:

Step 1 Task Configure a username and password to access the system remotely. Command Syntax
username username password [encryption-type] password encryption-type specifies how you are inputting the

Command Mode CONFIGURATION

password, is 0 by default, and is not required. 0 is for inputting the password in clear text. 7 is for inputting a password that is already encrypted using a Type 7 hash. Obtaining the encrypted password from the configuration of another Dell Networking system.

Access the S-Series Remotely

The S-Series does not have a dedicated management port nor a separate management routing table. Configure any port on the S-Series to be the port through which you manage the system and configure an IP route to that gateway.
Note: The S60 system uses management ports and should be configured similar to the C-Series and E-Series systems. Refer to Access the C-Series and E-Series and the S60 Remotely

Configuring the system for Telnet access is a three-step process: 1. Configure an IP address for the port through which you will manage the system using the command ip address from INTERFACE mode, as shown in Figure 3-2. 2. Configure a IP route with a default gateway using the command ip route from CONFIGURATION mode, as shown in Figure 3-2. 3. Configure a username and password using the command username from CONFIGURATION mode, as shown in Figure 3-2.
Figure 3-2. Configuring the S-Series for Remote Access

R5(conf)#int gig 0/48 R5(conf-if-gi-0/48)#ip address 10.11.131.240 R5(conf-if-gi-0/48)#show config ! interface GigabitEthernet 0/48 ip address 10.11.131.240/24 no shutdown R5(conf-if-gi-0/48)#exit R5(conf)#ip route 10.11.32.0/23 10.11.131.254 R5(conf)#username admin pass FTOS

Getting Started | 41

www.dell.com | support.dell.com

Configure the Enable Password

Access the EXEC Privilege mode using the enable command. The EXEC Privilege mode is unrestricted by default. Configure a password as a basic security measure. There are two types of enable passwords:
enable password

stores the password in the running/startup configuration using a DES encryption

method.
enable secret

is stored in the running/startup configuration in using a stronger, MD5 encryption

method. Dell Networking recommends using the enable secret password. To configure an enable password:
Task Create a password to access EXEC Privilege mode. Command Syntax
enable [password | secret] [level level] [encryption-type]
password level is the privilege level, is 15 by default, and is not required. encryption-type specifies how you are inputting the password, is 0 by default, and is not required.

Command Mode CONFIGURATION

0 is for inputting the password in clear text. 7 is for inputting a password that is already encrypted using a DES hash. Obtain the encrypted password from the configuration file of another Dell Networking system. 5 is for inputting a password that is already encrypted using an MD5 hash. Obtain the encrypted password from the configuration file of another Dell Networking system.

Configuration File Management

Files can be stored on and accessed from various storage media. Rename, delete, and copy files on the system from the EXEC Privilege mode. The E-Series TeraScale and ExaScale platforms architecture use Compact Flash for the internal and external Flash memory. It has a space limitation but does not limit the number of files it can contain.
Note: Using flash memory cards in the system that have not been approved by Dell Networking can cause unexpected system behavior, including a reboot.

42

Getting Started

Copy Files to and from the System

The command syntax for copying files is similar to UNIX. The copy command uses the format copy source-file-url destination-file-url.
Note: See the FTOS Command Reference for a detailed description of the copy command.

To copy a local file to a remote system, combine the file-origin syntax for a local file location with the file-destination syntax for a remote file location shown in Table 3-2. To copy a remote file to Dell Networking system, combine the file-origin syntax for a remote file location with the file-destination syntax for a local file location shown in Table 3-2.
Forming a copy Command source-file-url Syntax destination-file-url Syntax

Table 3-2.

Local File Location Internal flash: primary RPM standby RPM External flash: primary RPM standby RPM
copy rpm{0|1}slot0://filename copy rpm{0|1}slot0://filename rpm{0|1}slot0://filename rpm{0|1}slot0://filename copy flash://filename copy rpm{0|1}flash://filename flash://filename rpm{0|1}flash://filename

USB Drive (E-Series ExaScale and S60) USB drive on RPM0 External USB drive Remote File Location FTP server TFTP server SCP server
copy ftp://username:password@{hostip | hostname}/filepath/filename copy tftp://{hostip | hostname}/filepath/ filename copy scp://{hostip | hostname}/filepath/ filename ftp://username:password@{hostip | hostname}/ filepath/filename tftp://{hostip | hostname}/filepath/filename scp://{hostip | hostname}/filepath/filename copy rpm0usbflash://filepath copy usbflash://filepath

rpm0usbflash://filename
usbflash://filename

Important Points to Remember

You may not copy a file from one remote system to another. You may not copy a file from one location to the same location. The internal flash memories on the RPMs are synchronized whenever there is a change, but only if both RPMs are running the same version of FTOS. When copying to a server, a hostname can only be used if a DNS server is configured. The usbflash and rpm0usbflash commands are supported on E-Series ExaScale and S60 systems. Refer to your systems Release Notes for a list of approved USB vendors.

Getting Started | 43

www.dell.com | support.dell.com

Figure 3-3 shows an example of using the copy command to save a file to an FTP server.
Figure 3-3. Saving a file to a Remote System

Local Location Remote Location

FTOS#copy flash://FTOS-EF-8.2.1.0.bin ftp://myusername:[email protected]//FTOS/FTOS-EF-8.2.1.0 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 27952672 bytes successfully copied

Figure 3-4 shows an example of using the copy command to import a file to the Dell Networking system from an FTP server.
Figure 3-4. Saving a file to a Remote System

Remote Location Local Location

core1#$//copy ftp://myusername:[email protected]//FTOS/FTOS-EF-8.2.1.0.bin flash:// Destination file name [FTOS-EF-8.2.1.0.bin.bin]: !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 26292881 bytes successfully copied

Save the Running-configuration

The running-configuration contains the current system configuration. Dell Networking recommends that you copy your running-configuration to the startup-configuration. The system uses the startup-configuration during boot-up to configure the system. The startup-configuration is stored in the internal flash on the primary RPM by default, but it can be saved onto an external flash (on an RPM) or a remote server. To save the running-configuration:
Note: The commands in this section follow the same format as those in Copy Files to and from the System but use the filenames startup-configuration and running-configuration. These commands assume that current directory is the internal flash, which is the system default.

44

Getting Started

Task Save the running-configuration to: the startup-configuration on the internal flash of the primary RPM the internal flash on an RPM

Command Syntax

Command Mode

copy running-config startup-config

copy running-config rpm{0|1}flash://filename

Note: The internal flash memories on the RPMs are synchronized whenever there is a change, but only if the RPMs are running the same version of FTOS. the external flash of an RPM an FTP server
copy running-config rpm{0|1}slot0://filename copy running-config ftp:// username:password@{hostip | hostname}/filepath/ filename copy running-config tftp://{hostip | hostname}/ filepath/filename copy running-config scp://{hostip | hostname}/ filepath/filename

EXEC Privilege

a TFTP server an SCP server

Note: When copying to a server, a hostname can only be used if a DNS server is configured. Save the running-configuration to the startup-configuration on the internal flash of the primary RPM. Then copy the new startup-config file to the external flash of the primary RPM.
copy running-config startup-config duplicate

EXEC Privilege

FTOS Behavior: If you create a startup-configuration on an RPM and then move the RPM to another chassis, the startup-configuration is stored as a backup file (with the extension .bak), and a new, empty startup-configuration file is created. To restore your original startup-configuration in this situation, overwrite the new startup-configuration with the original one using the command copy startup-config.bak startup-config.

View Files
File information and content can only be viewed on local file systems. To view a list of files on the internal or external Flash:
Step 1 Task View a list of files on: the internal flash of an RPM the external flash of an RPM
dir flash: dir slot:

Command Syntax

Command Mode

EXEC Privilege

Getting Started | 45

www.dell.com | support.dell.com

The output of the command dir also shows the read/write privileges, size (in bytes), and date of modification for each file, as shown in Figure 3-5.
Figure 3-5. Viewing a List of Files in the Internal Flash

FTOS#dir Directory of flash: 1 drw2 drwx 3 drw4 drw5 drw6 drw7 d--8 -rw9 -rw10 -rw11 drw12 -rw13 -rw14 -rw15 -rw--More-32768 512 8192 8192 8192 8192 8192 33059550 27674906 27674906 8192 7276 7341 27674906 27674906 Jan Jul Mar Mar Mar Mar Mar Jul Jul Jul Jan Jul Jul Jul Jul 01 23 30 30 30 30 30 11 06 06 01 20 20 06 06 1980 2007 1919 1919 1919 1919 1919 2007 2007 2007 1980 2007 2007 2007 2007 00:00:00 00:38:44 10:31:04 10:31:04 10:31:04 10:31:04 10:31:04 17:49:46 00:20:24 19:54:52 00:18:28 01:52:40 15:34:46 19:52:22 02:23:22 . .. TRACE_LOG_DIR CRASH_LOG_DIR NVTRACE_LOG_DIR CORE_DUMP_DIR ADMIN_DIR FTOS-EF-7.4.2.0.bin FTOS-EF-4.7.4.302.bin boot-image-FILE diag startup-config.bak startup-config boot-image boot-flash

To view the contents of a file:

Step 1 Task View the: contents of a file in the internal flash of an RPM contents of a file in the external flash of an RPM running-configuration startup-configuration
show file rpm{0|1}flash://filename

Command Syntax

Command Mode

show file rpm{0|1}slot0://filename

EXEC Privilege

show running-config show startup-config

View Configuration Files

Configuration files have three commented lines at the beginning of the file, as shown in Figure 3-6, to help you track the last time any user made a change to the file, which user made the changes, and when the file was last saved to the startup-configuration. In the running-configuration file, if there is a difference between the timestamp on the Last configuration change, and Startup-config last updated, then you have made changes that have not been saved and will not be preserved upon a system reboot.

46

Getting Started

Figure 3-6.

Tracking Changes with Configuration Comments

FTOS#show running-config Current Configuration ... ! Version 8.2.1.0 ! Last configuration change at Thu Apr 3 23:06:28 2008 by admin ! Startup-config last updated at Thu Apr 3 23:06:55 2008 by admin ! boot system rpm0 primary flash://FTOS-EF-8.2.1.0.bin boot system rpm0 secondary flash://FTOS-EF-7.8.1.0.bin boot system rpm0 default flash://FTOS-EF-7.7.1.1.bin boot system rpm1 primary flash://FTOS-EF-7.8.1.0.bin boot system gateway 10.10.10.100 --More--

File System Management

The Dell Networking system can use the internal Flash, external Flash, or remote devices to store files. It stores files on the internal Flash by default but can be configured to store files elsewhere. To view file system information:
Task View information about each file system. Command Syntax
show file-systems

Command Mode EXEC Privilege

The output of the command show file-systems (Figure 3-7) shows the total capacity, amount of free memory, file structure, media type, read/write privileges for each storage device in use.
Figure 3-7. show file-systems Command Example
Prefixes rw flash: rw slot0: rw ftp: rw tftp: rw scp:

FTOS#show file-systems Size(b) Free(b) Feature Type Flags 520962048 213778432 dosFs2.0 USERFLASH 127772672 21936128 dosFs2.0 USERFLASH network network network

You can change the default file system so that file management commands apply to a particular device or memory. To change the default storage location:
Task Change the default directory. Command Syntax
cd directory

Command Mode EXEC Privilege

In Figure 3-8, the default storage location is changed to the external Flash of the primary RPM. File management commands then apply to the external Flash rather than the internal Flash.

Getting Started | 47

www.dell.com | support.dell.com

Figure 3-8.

Alternative Storage Location

FTOS#cd slot0: FTOS#copy running-config test FTOS#copy run test ! 7419 bytes successfully copied FTOS#dir Directory of slot0: 1 2 3 4 5 6 7 8 9 drwdrwx ----rw---------------32768 512 0 7419 0 0 0 0 0 Jan Jul Jan Jul Jan Jan Jan Jan Jan 01 23 01 23 01 01 01 01 01 1980 2007 1970 2007 1970 1970 1970 1970 1970 00:00:00 00:38:44 00:00:00 20:44:40 00:00:00 00:00:00 00:00:00 00:00:00 00:00:00

No File System Specified

. .. DCIM test BT 200702~1VSN G F F

File Saved to External Flash

slot0: 127772672 bytes total (21927936 bytes free)

View command history

The command-history trace feature captures all commands entered by all users of the system with a time stamp and writes these messages to a dedicated trace log buffer. The system generates a trace message for each executed command. No password information is saved to the file. To view the command-history trace, use the show command-history command, as shown in Figure 487.
Figure 3-9. Command Example show command-history

FTOS#show command-history [12/5 10:57:8]: CMD-(CLI):service password-encryption [12/5 10:57:12]: CMD-(CLI):hostname FTOS [12/5 10:57:12]: CMD-(CLI):ip telnet server enable [12/5 10:57:12]: CMD-(CLI):line console 0 [12/5 10:57:12]: CMD-(CLI):line vty 0 9 [12/5 10:57:13]: CMD-(CLI):boot system rpm0 primary flash://FTOS-CB-1.1.1.2E2.bin

Upgrading and Downgrading FTOS

Note: To upgrade or downgrade FTOS, see the release notes for the version you want to load on the system.

48

Getting Started

4
Management
Management is supported on platforms:

ces

This chapter explains the different protocols or services used to manage the Dell Networking system including: Configure Privilege Levels Configure Logging File Transfer Services Terminal Lines Lock CONFIGURATION mode Recovering from a Forgotten Password on the S60 Recovering from a Failed Start on the S60

Configure Privilege Levels

Privilege levels restrict access to commands based on user or terminal line. There are 16 privilege levels, of which three are pre-defined. The default privilege level is 1. Level 0Access to the system begins at EXEC mode, and EXEC mode commands are limited to enable, disable, and exit. Level 1Access to the system begins at EXEC mode, and all commands are available. Level 15Access to the system begins at EXEC Privilege mode, and all commands are available.

Create a Custom Privilege Level

Custom privilege levels start with the default EXEC mode command set. You can then customize privilege levels 2-14 by: restricting access to an EXEC mode command moving commands from EXEC Privilege to EXEC mode restricting access

A user can access all commands at his privilege level and below.

Management | 49

www.dell.com | support.dell.com

Removing a command from EXEC mode

Remove a command from the list of available commands in EXEC mode for a specific privilege level using the command privilege exec from CONFIGURATION mode. In the command, specify a level greater than the level given to a user or terminal line, followed by the first keyword of each command to be restricted.

Move a command from EXEC privilege mode to EXEC mode

Move a command from EXEC Privilege to EXEC mode for a privilege level using the command privilege exec from CONFIGURATION mode. In the command, specify the privilege level of the user or terminal line, and specify all keywords in the command to which you want to allow access.

Allow Access to CONFIGURATION mode commands

Allow access to CONFIGURATION mode using the command privilege exec level level configure from CONFIGURATION mode. A user that enters CONFIGURATION mode remains at his privilege level, and has access to only two commands, end and exit. You must individually specify each CONFIGURATION mode command to which you want to allow access using the command privilege configure level level. In the command, specify the privilege level of the user or terminal line, and specify all keywords in the command to which you want to allow access.

Allow Access to INTERFACE, LINE, ROUTE-MAP, and ROUTER mode

1. Similar to allowing access to CONFIGURATION mode, to allow access to INTERFACE, LINE, ROUTE-MAP, and ROUTER modes, you must first allow access to the command that enters you into the mode. For example, allow a user to enter INTERFACE mode using the command privilege
configure level level interface gigabitethernet

2. Then, individually identify the INTERFACE, LINE, ROUTE-MAP or ROUTER commands to which you want to allow access using the command privilege {interface | line | route-map | router} level level. In the command, specify the privilege level of the user or terminal line, and specify all keywords in the command to which you want to allow access. The following table lists the configuration tasks you can use to customize a privilege level:
Task Remove a command from the list of available commands in EXEC mode. Move a command from EXEC Privilege to EXEC mode. Allow access to CONFIGURATION mode. Command Syntax
privilege exec level level {command ||...|| command} privilege exec level level {command ||...|| command} privilege exec level level configure

Command Mode CONFIGURATION CONFIGURATION

CONFIGURATION

50

Management

Task Allow access to INTERFACE, LINE, ROUTE-MAP, and/or ROUTER mode. Specify all keywords in the command. Allow access to a CONFIGURATION, INTERFACE, LINE, ROUTE-MAP, and/or ROUTER mode command.

Command Syntax
privilege configure level level

Command Mode CONFIGURATION

{interface | line | route-map | router} {command-keyword ||...|| command-keyword}

privilege {configure |interface | line | route-map | router} level level {command ||...|| command}

CONFIGURATION

The configuration in Figure 4-1 creates privilege level 3. This level: removes the resequence command from EXEC mode by requiring a minimum of privilege level 4, moves the command capture bgp-pdu max-buffer-size from EXEC Privilege to EXEC mode by, requiring a minimum privilege level 3, which is the configured level for VTY 0, allows access to CONFIGURATION mode with the banner command, and allows access to INTERFACE and LINE modes are allowed with no commands.

Management | 51

www.dell.com | support.dell.com

Figure 4-1.

Create a Custom Privilege Level

FTOS(conf)#do show run priv ! privilege exec level 3 capture privilege exec level 3 configure privilege exec level 4 resequence privilege exec level 3 capture bgp-pdu privilege exec level 3 capture bgp-pdu max-buffer-size privilege configure level 3 line privilege configure level 3 interface FTOS(conf)#do telnet 10.11.80.201 [telnet output omitted] FTOS#show priv Current privilege level is 3. FTOS#? capture Capture packet configure Configuring from terminal disable Turn off privileged commands enable Turn on privileged commands exit Exit from the EXEC ip Global IP subcommands monitor Monitoring feature mtrace Trace reverse multicast path from destination to source ping Send echo messages quit Exit from the EXEC show Show running system information [output omitted] FTOS#config [output omitted] FTOS(conf)#do show priv Current privilege level is 3. FTOS(conf)#? end Exit from configuration mode exit Exit from configuration mode interface Select an interface to configure line Configure a terminal line linecard Set line card type FTOS(conf)#interface ? fastethernet Fast Ethernet interface gigabitethernet Gigabit Ethernet interface loopback Loopback interface managementethernet Management Ethernet interface null Null interface port-channel Port-channel interface range Configure interface range sonet SONET interface tengigabitethernet TenGigabit Ethernet interface vlan VLAN interface FTOS(conf)#interface gigabitethernet 1/1 FTOS(conf-if-gi-1/1)#? end Exit from configuration mode exit Exit from interface configuration mode FTOS(conf-if-gi-1/1)#exit FTOS(conf)#line ? aux Auxiliary line console Primary terminal line vty Virtual terminal FTOS(conf)#line vty 0 FTOS(config-line-vty)#? exit Exit from line configuration mode FTOS(config-line-vty)#

52

Management

Apply a Privilege Level to a Username

To set a privilege level for a user:
Task Configure a privilege level for a user. Command Syntax
username username privilege level

Command Mode CONFIGURATION

Apply a Privilege Level to a Terminal Line

To set a privilege level for a terminal line:
Task Configure a privilege level for a terminal line. Command Syntax
privilege level level

Command Mode LINE

Note: When you assign a privilege level between 2 and 15, access to the system begins at EXEC mode, but the prompt is hostname#, rather than hostname>.

Configure Logging
FTOS tracks changes in the system using event and error messages. By default, FTOS logs these messages on: the internal buffer console and terminal lines, and any configured syslog servers

Disable Logging
To disable logging:
Task Disable all logging except on the console. Disable logging to the logging buffer. Disable logging to terminal lines. Disable console logging. Command Syntax
no logging on

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION CONFIGURATION

no logging buffer no logging monitor no logging console

Management | 53

www.dell.com | support.dell.com

Log Messages in the Internal Buffer

All error messages, except those beginning with %BOOTUP (Message), are log in the internal buffer.
Message 1 BootUp Events
%BOOTUP:RPM0:CP %PORTPIPE-INIT-SUCCESS: Portpipe 0 enabled

Configuration Task List for System Log Management

The following list includes the configuration tasks for system log management: Disable System Logging Send System Messages to a Syslog Server

Disable System Logging

By default, logging is enabled and log messages are sent to the logging buffer, all terminal lines, console, and syslog servers. Enable and disable system logging using the following commands:
Task Disable all logging except on the console. Disable logging to the logging buffer. Disable logging to terminal lines. Disable console logging. Command Syntax
no logging on

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION CONFIGURATION

no logging buffer no logging monitor no logging console

54

Management

Send System Messages to a Syslog Server

Send system messages to a syslog server by specifying the server with the following command:
Task Specify the server to which you want to send system messages. You can configure up to eight syslog servers. Command Syntax
logging {ip-address | hostname}

Command Mode CONFIGURATION

Configure a Unix System as a Syslog Server

Configure a UNIX system as a syslog server by adding the following lines to /etc/syslog.conf on the Unix system and assigning write permissions to the file. on a 4.1 BSD UNIX system, add the line: local7.debugging /var/log/force10.log on a 5.7 SunOS UNIX system, add the line: local7.debugging /var/adm/force10.log

In the lines above, local7 is the logging facility level and debugging is the severity level.

Change System Logging Settings

You can change the default settings of the system logging by changing the severity level and the storage location. The default is to log all messages up to debug level, that is, all system messages. By changing the severity level in the logging commands, you control the number of system messages logged.
Task Specify the minimum severity level for logging to the logging buffer. Specify the minimum severity level for logging to the console. Specify the minimum severity level for logging to terminal lines. Specifying the minimum severity level for logging to a syslog server. Specify the minimum severity level for logging to the syslog history table. Command Syntax
logging buffered level logging console level logging monitor level logging trap level logging history level

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION CONFIGURATION CONFIGURATION

Management | 55

www.dell.com | support.dell.com

Task Specify the size of the logging buffer. Note: When you decrease the buffer size, FTOS deletes all messages stored in the buffer. Increasing the buffer size does not affect messages in the buffer. Specify the number of messages that FTOS saves to its logging history table.

Command Syntax
logging buffered size

Command Mode CONFIGURATION

logging history size size

CONFIGURATION

To change one of the settings for logging system messages, use any or all of the following commands in the CONFIGURATION mode: To view the logging buffer and configuration, use the show logging command (Figure 35) in the EXEC privilege mode. To change the severity level of messages logged to a syslog server, use the following command in the CONFIGURATION mode: To view the logging configuration, use the show running-config logging command (Figure 37) in the EXEC privilege mode.

Display the Logging Buffer and the Logging Configuration

Display the current contents of the logging buffer and the logging settings for the system, use the show logging command (Figure 4-2) in the EXEC privilege mode.

56

Management

Figure 4-2.

show logging Command Example

FTOS#show logging syslog logging: enabled Console logging: level Debugging Monitor logging: level Debugging Buffer logging: level Debugging, 40 Messages Logged, Size (40960 bytes) Trap logging: level Informational %IRC-6-IRC_COMMUP: Link to peer RPM is up %RAM-6-RAM_TASK: RPM1 is transitioning to Primary RPM. %RPM-2-MSG:CP1 %POLLMGR-2-MMC_STATE: External flash disk missing in 'slot0:' %CHMGR-5-CARDDETECTED: Line card 0 present %CHMGR-5-CARDDETECTED: Line card 2 present %CHMGR-5-CARDDETECTED: Line card 4 present %CHMGR-5-CARDDETECTED: Line card 5 present %CHMGR-5-CARDDETECTED: Line card 8 present %CHMGR-5-CARDDETECTED: Line card 10 present %CHMGR-5-CARDDETECTED: Line card 12 present %TSM-6-SFM_DISCOVERY: Found SFM 0 %TSM-6-SFM_DISCOVERY: Found SFM 1 %TSM-6-SFM_DISCOVERY: Found SFM 2 %TSM-6-SFM_DISCOVERY: Found SFM 3 %TSM-6-SFM_DISCOVERY: Found SFM 4 %TSM-6-SFM_DISCOVERY: Found SFM 5 %TSM-6-SFM_DISCOVERY: Found SFM 6 %TSM-6-SFM_DISCOVERY: Found SFM 7 %TSM-6-SFM_SWITCHFAB_STATE: Switch Fabric: UP %TSM-6-SFM_DISCOVERY: Found SFM 8 %TSM-6-SFM_DISCOVERY: Found 9 SFMs %CHMGR-5-CHECKIN: Checkin from line card 5 (type EX1YB, 1 ports) %TSM-6-PORT_CONFIG: Port link status for LC 5 => portpipe 0: OK portpipe 1: N/A %CHMGR-5-LINECARDUP: Line card 5 is up %CHMGR-5-CHECKIN: Checkin from line card 12 (type S12YC12, 12 ports) %TSM-6-PORT_CONFIG: Port link status for LC 12 => portpipe 0: OK portpipe 1: N/A %CHMGR-5-LINECARDUP: Line card 12 is up %IFMGR-5-CSTATE_UP: changed interface Physical state to up: So 12/8 %IFMGR-5-CSTATE_DN: changed interface Physical state to down: So 12/8

To view any changes made, use the show running-config logging command (Figure 4-3) in the EXEC privilege mode.

Management | 57

www.dell.com | support.dell.com

Configure a UNIX logging facility level

You can save system log messages with a UNIX system logging facility. To configure a UNIX logging facility level, use the following command in the CONFIGURATION mode:
Command Syntax
logging facility [facility-type]

Command Mode CONFIGURATION

Purpose Specify one of the following parameters. auth (for authorization messages) cron (for system scheduler messages) daemon (for system daemons) kern (for kernel messages) local0 (for local use) local1 (for local use) local2 (for local use) local3 (for local use) local4 (for local use) local5 (for local use) local6 (for local use) local7 (for local use). This is the default. lpr (for line printer system messages) mail (for mail system messages) news (for USENET news messages) sys9 (system use) sys10 (system use) sys11 (system use) sys12 (system use) sys13 (system use) sys14 (system use) syslog (for syslog messages) user (for user programs) uucp (UNIX to UNIX copy protocol) The default is local7.

To view nondefault settings, use the show running-config logging command (Figure 4-3) in the EXEC mode.
Figure 4-3. show running-config logging Command Example

FTOS#show running-config logging ! logging buffered 524288 debugging service timestamps log datetime msec service timestamps debug datetime msec ! logging trap debugging logging facility user logging source-interface Loopback 0 logging 10.10.10.4 FTOS#

58

Management

Synchronize log messages

You can configure FTOS to filter and consolidate the system messages for a specific line by synchronizing the message output. Only the messages with a severity at or below the set level appear. This feature works on the terminal and console connections available on the system. To synchronize log messages, use these commands in the following sequence starting in the CONFIGURATION mode:
Step 1 Command Syntax
line {console 0 | vty number [end-number] | aux 0}

Command Mode CONFIGURATION

Purpose Enter the LINE mode. Configure the following parameters for the virtual terminal lines: number range: zero (0) to 8. end-number range: 1 to 8. You can configure multiple virtual terminals at one time by entering a number and an end-number. Configure a level and set the maximum number of messages to be printed. Configure the following optional parameters: level severity-level range: 0 to 7. Default is 2. Use the all keyword to include all messages. limit range: 20 to 300. Default is 20.

logging synchronous [level severity-level | all] [limit]

LINE

To view the logging synchronous configuration, use the show config command in the LINE mode.

Enable timestamp on syslog messages

syslog messages, by default, do not include a time/date stamp stating when the error or message was created.

Management | 59

www.dell.com | support.dell.com

To have FTOS include a timestamp with the syslog message, use the following command syntax in the CONFIGURATION mode:
Command Syntax
service timestamps [log | debug] [datetime [localtime] [msec] [show-timezone] | uptime]

Command Mode CONFIGURATION

Purpose Add timestamp to syslog messages. Specify the following optional parameters: datetime: You can add the keyword localtime to include the localtime, msec, and show-timezone. If you do not add the keyword localtime, the time is UTC. uptime. To view time since last boot. If neither parameter is specified, FTOS configures uptime.

To view the configuration, use the show running-config logging command in the EXEC privilege mode. To disable time stamping on syslog messages, enter no service timestamps [log | debug].

File Transfer Services

With FTOS, you can configure the system to transfer files over the network using File Transfer Protocol (FTP). One FTP application is copying the system image files over an interface on to the system; however, FTP is not supported on VLAN interfaces. For more information on FTP, refer to RFC 959, File Transfer Protocol.

Configuration Task List for File Transfer Services

The following list includes the configuration tasks for file transfer services: Enable FTP server (mandatory) Configure FTP server parameters (optional) Configure FTP client parameters (optional)

For a complete listing of FTP related commands, refer to .

60

Management

Enable FTP server

To enable the system as an FTP server, use the following command in the CONFIGURATION mode:
Command Syntax
ftp-server enable

Command Mode CONFIGURATION

Purpose Enable FTP on the system.

To view FTP configuration, use the show running-config ftp command (Figure 4-4) in the EXEC privilege mode.
Figure 4-4. show running-config ftp Command Output

FTOS#show running ftp ! ftp-server enable ftp-server username nairobi password 0 zanzibar FTOS#

Configure FTP server parameters

After the FTP server is enabled on the system, you can configure different parameters. To configure FTP server parameters, use any or all of the following commands in the CONFIGURATION mode:
Command Syntax
ftp-server topdir dir

Command Mode CONFIGURATION

Purpose Specify the directory for users using FTP to reach the system. The default is the internal flash directory. Specify a user name for all FTP users and configure either a plain text or encrypted password. Configure the following optional and required parameters: username: Enter a text string encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a text string.

ftp-server username username password [encryption-type]

password

CONFIGURATION

Note: You cannot use the change directory (cd) command until ftp-server topdir has been configured.

To view the FTP configuration, use the show running-config ftp command in EXEC privilege mode.

Management | 61

www.dell.com | support.dell.com

Configure FTP client parameters

To configure FTP client parameters, use the following commands in the CONFIGURATION mode:
Command Syntax
ip ftp source-interface interface

Command Mode CONFIGURATION

Purpose Enter the following keywords and slot/port or number information: For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a loopback interface, enter the keyword loopback followed by a number between 0 and 16383. For a port channel interface, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a VLAN interface, enter the keyword vlan followed by a number from 1 to 4094. E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS. Configure a password. Enter username to use on FTP client.

ip ftp password password ip ftp username name

CONFIGURATION CONFIGURATION

To view FTP configuration, use the show running-config ftp command (Figure 4-4) in the EXEC privilege mode.

Terminal Lines
You can access the system remotely and restrict access to the system by creating user profiles. The terminal lines on the system provide different means of accessing the system. The console line (console) connects you through the Console port in the RPMs. The virtual terminal lines (VTY) connect you through Telnet to the system. The auxiliary line (aux) connects secondary devices such as modems.

Deny and Permit Access to a Terminal Line

Dell Networking recommends applying only standard ACLs to deny and permit access to VTY lines. Layer 3 ACL deny all traffic that is not explicitly permitted, but in the case of VTY lines, an ACL with no rules does not deny any traffic. You cannot use show ip accounting access-list to display the contents of an ACL that is applied only to a VTY line.

62

Management

To apply an IP ACL to a line:

Task Apply an ACL to a VTY line. Command Syntax
ip access-class access-list

Command Mode LINE

To view the configuration, enter the show config command in the LINE mode, as shown in Figure 4-5.
Figure 4-5. Applying an Access List to a VTY Line

FTOS(config-std-nacl)#show config ! ip access-list standard myvtyacl seq 5 permit host 10.11.0.1 FTOS(config-std-nacl)#line vty 0 FTOS(config-line-vty)#show config line vty 0 access-class myvtyacl

FTOS Behavior: Prior to FTOS version 7.4.2.0, in order to deny access on a VTY line, you must apply an ACL and AAA authentication to the line. Then users are denied access only after they enter a username and password. Beginning in FTOS version 7.4.2.0, only an ACL is required, and users are denied access before they are prompted for a username and password.

Configure Login Authentication for Terminal Lines

You can use any combination of up to 6 authentication methods to authenticate a user on a terminal line. A combination of authentication methods is called a method list. If the user fails the first authentication method, FTOS prompts the next method until all methods are exhausted, at which point the connection is terminated. The available authentication methods are: for the enable password. linePrompt for the e password you assigned to the terminal line. You must configure a password for the terminal line to which you assign a method list that contains the line authentication method. Configure a password using the command password from LINE mode. localPrompt for the the system username and password. noneDo not authenticate the user. radiusPrompt for a username and password and use a RADIUS server to authenticate. tacacs+Prompt for a username and password and use a TACACS+ server to authenticate.
enablePrompt

Management | 63

www.dell.com | support.dell.com

To configure authentication for a terminal line:

Step 1 Task Create an authentication method list. You may use a mnemonic name or use the keyword default. The default authentication method for terminal lines is local, and the default method list is empty. Apply the method list from Step 1 to a terminal line. If you used the line authentication method in the method list you applied to the terminal line, configure a password for the terminal line. Command Syntax
aaa authentication login {method-list-name | default} [method-1] [method-2] [method-3] [method-4] [method-5] [method-6]

Command Mode CONFIGURATION

2 3

login authentication {method-list-name | default}

password

CONFIGURATION LINE

In Figure 4-6 VTY lines 0-2 use a single authentication method, line.
Figure 4-6. Configuring Login Authentication on a Terminal Line

FTOS(conf)#aaa authentication login myvtymethodlist line FTOS(conf)#line vty 0 2 FTOS(config-line-vty)#login authentication myvtymethodlist FTOS(config-line-vty)#password myvtypassword FTOS(config-line-vty)#show config line vty 0 password myvtypassword login authentication myvtymethodlist line vty 1 password myvtypassword login authentication myvtymethodlist line vty 2 password myvtypassword login authentication myvtymethodlist FTOS(config-line-vty)#

Time out of EXEC Privilege Mode

EXEC timeout is a basic security feature that returns FTOS to the EXEC mode after a period of inactivity on terminal lines.

64

Management

To change the timeout period or disable EXEC timeout.

Task Set the number of minutes and seconds. Default: 10 minutes on console, 30 minutes on VTY. Disable EXEC timeout by setting the timeout period to 0. Return to the default timeout values. Command Syntax
exec-timeout minutes [seconds]

Command Mode LINE

no exec-timeout

LINE

View the configuration using the command show config from LINE mode.
Figure 4-7. Configuring EXEC Timeout

FTOS(conf)#line con 0 FTOS(config-line-console)#exec-timeout 0 FTOS(config-line-console)#show config line console 0 exec-timeout 0 0 FTOS(config-line-console)#

Telnet to Another Network Device

To telnet to another device:
Task Telnet to the peer RPM. You do not need to configure the management port on the peer RPM to be able to telnet to it. Telnet to a device with an IPv4 or IPv6 address. If you do not enter an IP address, FTOS enters a Telnet dialog that prompts you for one. Enter an IPv4 address in dotted decimal format (A.B.C.D). Enter an IPv6 address in the format 0000:0000:0000:0000:0000:0000:0000:0000. Elision of zeros is supported. Command Syntax
telnet-peer-rpm

Command Mode EXEC Privilege EXEC Privilege

telnet [ip-address]

Management | 65

www.dell.com | support.dell.com

Figure 4-8.

Telnet to Another Network Device

FTOS# telnet 10.11.80.203 Trying 10.11.80.203... Connected to 10.11.80.203. Exit character is '^]'. Login: Login: admin Password: FTOS>exit FTOS#telnet 2200:2200:2200:2200:2200::2201 Trying 2200:2200:2200:2200:2200::2201... Connected to 2200:2200:2200:2200:2200::2201. Exit character is '^]'. FreeBSD/i386 (freebsd2.force10networks.com) (ttyp1) login: admin FTOS#

Lock CONFIGURATION mode

FTOS allows multiple users to make configurations at the same time. You can lock CONFIGURATION mode so that only one user can be in CONFIGURATION mode at any time (Message 2). A two types of locks can be set: auto and manual. Set an auto-lock using the command configuration mode exclusive auto from CONFIGURATION mode. When you set an auto-lock, every time a user is in CONFIGURATION mode all other users are denied access. This means that you can exit to EXEC Privilege mode, and re-enter CONFIGURATION mode without having to set the lock again. Set a manual lock using the command configure terminal lock from CONFIGURATION mode. When you configure a manual lock, which is the default, you must enter this command time you want to enter CONFIGURATION mode and deny access to others.
Locking CONFIGURATION mode

Figure 4-9.

FTOS(conf)#configuration mode exclusive auto BATMAN(conf)#exit 3d23h35m: %RPM0-P:CP %SYS-5-CONFIG_I: Configured from console by FTOS#config ! Locks configuration mode exclusively. FTOS(conf)#

console

If another user attempts to enter CONFIGURATION mode while a lock is in place, Message 1 appears on their terminal.
Message 1 CONFIGURATION mode Locked Error
% Error: User "" on line console0 is in exclusive configuration mode

66

Management

If any user is already in CONFIGURATION mode when while a lock is in place, Message 2 appears on their terminal.
Message 2 Cannot Lock CONFIGURATION mode Error
% Error: Can't lock configuration mode exclusively since the following users are currently configuring the system: User "admin" on line vty1 ( 10.1.1.1 )

Note: The CONFIGURATION mode lock corresponds to a VTY session, not a user. Therefore, if you configure a lock and then exit CONFIGURATION mode, and another user enters CONFIGURATION mode, when you attempt to re-enter CONFIGURATION mode, you are denied access even though you are the one that configured the lock.

Note: If your session times out and you return to EXEC mode, the CONFIGURATION mode lock is unconfigured.

Viewing the Configuration Lock Status

If you attempt to enter CONFIGURATION mode when another user has locked it, you may view which user has control of CONFIGURATION mode using the command show configuration lock from EXEC Privilege mode. You can then send any user a message using the send command from EXEC Privilege mode. Alternatively you can clear any line using the command clear from EXEC Privilege mode. If you clear a console session, the user is returned to EXEC mode.

Recovering from a Forgotten Password on the S60

If you configure authentication for the console and you exit out of EXEC mode or your console session times out, you are prompted for a password to re-enter. If you forget your password:
Step 1 2 3 Task Log onto the system via console. Power-cycle the chassis by switching off all of the power modules and then switching them back on. Press any key to abort the boot process. You enter uBoot immediately, as indicated by the => prompt. press any key (during bootup) Command Syntax Command Mode

Management | 67

www.dell.com | support.dell.com

Step 4

Task Set the system parameters to ignore the startup configuration file when the system reloads. To save the changes, use the saveenv command Save the running-config. Reload the system. Copy startup-config.bak to the running config. Remove all authentication statements you might have for the console. Save the running-config. Set the system parameters to use the startup configuration file when the system reloads. Save the running-config.

Command Syntax
setenv stconfigignore true

Command Mode uBoot

5 6 7 8 9 10 11

saveenv

uBoot EXEC Privilege uBoot EXEC Privilege LINE EXEC Privilege uBoot

copy running-config startup-config reset copy flash://startup-config.bak running-config no authentication login no password copy running-config startup-config setenv stconfigignore false

12

copy running-config startup-config

EXEC Privilege

Recovering from a Forgotten Enable Password on the S60

If you forget the enable password:
Step 1 2 3 Task Log onto the system via console. Power-cycle the chassis by switching off all of the power modules and then switching them back on. Press any key to abort the boot process. You enter uBoot immediately, as indicated by the => prompt. Set the system parameters to ignore the enable password when the system reloads. To save the changes, use the saveenv command Save the running-config. Reload the system. Configure a new enable password. Press any key (during bootup) Command Syntax Command Mode

setenv enablepwdignore true

uBoot

5 6 7 8

saveenv

uBoot EXEC Privilege uBoot CONFIGURATION

copy running-config startup-config reset enable {secret | password}

68

Management

Step 9

Task Save the running-config to the startup-config.

Command Syntax
copy running-config startup-config

Command Mode EXEC Privilege

Recovering from a Failed Start on the S60

A system that does not start correctly might be attempting to boot from a corrupted FTOS image or from a mis-specified location. In that case, you can restart the system and interrupt the boot process to point the system to another boot location. Use the setenv command, as described below. For details on the setenv command, its supporting commands, and other commands that can help recover from a failed start, see the BuBoot chapter in the FTOS Command Line Reference for the S60.
Step 1 2 Task Command Syntax Command Mode

Power-cycle the chassis (pull the power cord and reinsert it). Press any key to abort the boot process. You enter uBoot immediately, as indicated by the => prompt. Assign the new location to the FTOS image to be used when the system reloads. Assign an IP address to the Management Ethernet interface. Assign an IP address as the default gateway for the system. Reload the system. Press any key (during bootup)

setenv [primary_image f10boot location | secondary_image f10boot location | default_image f10boot location] setenv ipaddre address

uBoot

4 5 6

uBoot uBoot uBoot

setenv gatewayip address

reset

Management | 69

www.dell.com | support.dell.com

70

Management

5
802.1ag
802.1ag is available only on platform:

Ethernet Operations, Administration, and Maintenance (OAM) is a set of tools used to install, monitor, troubleshoot and manage Ethernet infrastructure deployments. Ethernet OAM consists of three main areas: 1. Service Layer OAM: IEEE 802.1ag Connectivity Fault Management (CFM) 2. Link Layer OAM: IEEE 802.3ah OAM 3. Ethernet Local management Interface (MEF-16 E-LMI)

Ethernet CFM
Ethernet CFM is an end-to-end per-service-instance Ethernet OAM scheme which enables: proactive connectivity monitoring, fault verification, and fault isolation. The service-instance with regard to OAM for Metro/Carrier Ethernet is a VLAN. This service is sold to an end-customer by a network service provider. Typically the service provider contracts with multiple network operators to provide end-to-end service between customers. For end-to-end service between customer switches, connectivity must be present across the service provider through multiple network operators. Layer 2 Ethernet networks usually cannot be managed with IP tools such as ICMP Ping and IP Traceroute. Traditional IP tools often fail because: there are complex interactions between various Layer 2 and Layer 3 protocols such as STP, LAG, VRRP and ECMP configurations. Ping and traceroute are not designed to verify data connectivity in the network and within each node in the network (such as in the switching fabric and hardware forwarding tables). when networks are built from different operational domains, access controls impose restrictions that cannot be overcome at the IP level, resulting in poor fault visibility. There is a need for hierarchical domains that can be monitored and maintained independently by each provider or operator. routing protocols choose a subset of the total network topology for forwarding, making it hard to detect faults in links and nodes that are not included in the active routing topology. This is made more complex when using some form of Traffic Engineering (TE) based routing. network and element discovery and cataloging is not clearly defined using IP troubleshooting tools.

802.1ag | 71

www.dell.com | support.dell.com

There is a need for Layer 2 equivalents to manage and troubleshoot native Layer 2 Ethernet networks. With these tools, you can identify, isolate, and repair faults quickly and easily, which reduces operational cost of running the network. OAM also increases availability and reduces mean time to recovery, which allows for tighter service level agreements, resulting in increased revenue for the service provider. In addition to providing end-to-end OAM in native Layer 2 Ethernet Service Provider/Metro networks, you can also use CFM to manage and troubleshoot any Layer 2 network including enterprise, datacenter, and cluster networks.

Maintenance Domains
Connectivity Fault Management (CFM) divides a network into hierarchical maintenance domains, as shown in Figure 5-1. A CFM maintenance domain is a management space on a network that is owned and operated by a single management entity. The network administrator assigns a unique maintenance level (0 to 7) to each domain to define the hierarchical relationship between domains. Domains can touch or nest but cannot overlap or intersect as that would require management by multiple entities.
Figure 5-1. OAM Domains
Service Provider Network
Customer Network Customer Network

Ethernet Access

MPLS Core
Customer Domain (7)

MPLS Access

Provider Domain (6)

Operator Domain (5)

Operator Domain (5)

MPLS Domain (4)

Operator Domain (5)

Maintenance Points
Domains are comprised of logical entities called Maintenance Points. A maintenance point is an interface demarcation that confines CFM frames to a domain. There are two types of maintenance points: Maintenance End Points (MEPs): a logical entity that marks the end-point of a domain Maintenance Intermediate Points (MIPs): a logical entity configured at a port of a switch that is an intermediate point of a Maintenance Entity (ME). An ME is a point-to-point relationship between two MEPs within a single domain. MIPs are internal to a domain, not at the boundary, and respond to CFM only when triggered by linktrace and loopback messages. MIPs can be configured to snoop Continuity Check Messages (CCMs) to build a MIP CCM database.

72

802.1ag

These roles define the relationships between all devices so that each device can monitor the layers under its responsibility. Maintenance points drop all lower-level frames and forward all higher-level frames.
Figure 5-2. Maintenance Points
Service Provider Network
Customer Network Customer Network

Ethernet Access

MPLS Core

MPLS Access

Customer Domain (7)

Provider Domain (6)

Operator Domain (5)
Operator Domain (5)
MPLS Domain (4)

Operator Domain (5)

MEP

MIP

Maintenance End Points

A Maintenance End Point (MEP) is a logical entity that marks the end-point of a domain. There are two types of MEPs defined in 802.1ag for an 802.1 bridge: Up-MEP: monitors the forwarding path internal to an bridge on the customer or provider edge; on Dell Networking systems the internal forwarding path is effectively the switch fabric and forwarding engine. Down-MEP: monitors the forwarding path external another bridge.

Configure Up- MEPs on ingress ports, ports that send traffic towards the bridge relay. Configure Down-MEPs on egress ports, ports that send traffic away from the bridge relay.
Figure 5-3. Up-MEP versus Down-MEP
Service Provider Ethernet Access

Customer Network

towards relay Up-MEP Down-MEP

away from relay

802.1ag | 73

www.dell.com | support.dell.com

Implementation Information
Since the S-Series has a single MAC address for all physical/LAG interfaces, only one MEP is allowed per MA (per VLAN or per MD level).

Configure CFM
Configuring CFM is a five-step process: 1. Configure the ecfmacl CAM region using the cam-acl command. See Configure Ingress Layer 2 ACL Sub-partitions. 2. Enable Ethernet CFM 3. Create a Maintenance Domain 4. Create a Maintenance Association 5. Create Maintenance Points 6. Use CFM tools: a b c Continuity Check Messages Loopback Message and Response Linktrace Message and Response

Related Configuration Tasks

Enable CFM SNMP Traps Display Ethernet CFM Statistics

74

802.1ag

Enable Ethernet CFM

Task Spawn the CFM process. No CFM configuration is allowed until the CFM process is spawned. Disable Ethernet CFM without stopping the CFM process. Command Syntax
ethernet cfm

Command Mode CONFIGURATION ETHERNET CFM

disable

Create a Maintenance Domain

Connectivity Fault Management (CFM) divides a network into hierarchical maintenance domains, as shown in Figure 5-1.
Step 1 2 Task Create maintenance domain. Display maintenance domain information. FTOS# show ethernet cfm domain Domain Name: customer Level: 7 Total Service: 1 Services MA-Name My_MA Domain Name: praveen Level: 6 Total Service: 1 Services MA-Name Your_MA Command Syntax
domain name md-level number

Command Mode ETHERNET CFM EXEC Privilege

Range: 0-7
show ethernet cfm domain [name | brief]

VLAN 200

CC-Int 10s

X-CHK Status enabled

VLAN 100

CC-Int 10s

X-CHK Status enabled

802.1ag | 75

www.dell.com | support.dell.com

Create a Maintenance Association

A Maintenance Association MA is a subdivision of an MD that contains all managed entities corresponding to a single end-to-end service, typically a VLAN. An MA is associated with a VLAN ID.
Task Create maintenance association. Command Syntax
service name vlan vlan-id

Command Mode ECFM DOMAIN

Create Maintenance Points

Domains are comprised of logical entities called Maintenance Points. A maintenance point is a interface demarcation that confines CFM frames to a domain. There are two types of maintenance points: Maintenance End Points (MEPs): a logical entity that marks the end-point of a domain Maintenance Intermediate Points (MIPs): a logical entity configured at a port of a switch that constitutes intermediate points of an Maintenance Entity (ME). An ME is a point-to-point relationship between two MEPs within a single domain.

These roles define the relationships between all devices so that each device can monitor the layers under its responsibility.

Create a Maintenance End Point

A Maintenance End Point (MEP) is a logical entity that marks the end-point of a domain. There are two types of MEPs defined in 802.1ag for an 802.1 bridge: Up-MEP: monitors the forwarding path internal to an bridge on the customer or provider edge; on Dell Networking systems the internal forwarding path is effectively the switch fabric and forwarding engine. Down-MEP: monitors the forwarding path external another bridge.

Configure Up- MEPs on ingress ports, ports that send traffic towards the bridge relay. Configure Down-MEPs on egress ports, ports that send traffic away from the bridge relay.
Task Create an MEP. Command Syntax
ethernet cfm mep {up-mep | down-mep} domain {name | level } ma-name name mepid mep-id

Command Mode INTERFACE

Range: 1-8191 Display configured MEPs and MIPs.

show ethernet cfm maintenance-points local [mep | mip]

EXEC Privilege

76

802.1ag

Task

Command Syntax

Command Mode

FTOS#show ethernet cfm maintenance-points local mep ------------------------------------------------------------------------------MPID Domain Name Level Type Port CCM-Status MA Name VLAN Dir ------------------------------------------------------------------------------100 200 300 cfm0 cfm1 cfm2 7 6 5 MEP MEP MEP Gi 4/10 Gi 4/10 Gi 4/10 Enabled Enabled Enabled test0 test1 test2 10 20 30 DOWN DOWN DOWN

MAC

00:01:e8:59:23:45 00:01:e8:59:23:45 00:01:e8:59:23:45

Create a Maintenance Intermediate Point

Maintenance Intermediate Point (MIP) is a logical entity configured at a port of a switch that constitutes intermediate points of an Maintenance Entity (ME). An ME is a point-to-point relationship between two MEPs within a single domain. An MIP is not associated with any MA or service instance, and it belongs to the entire MD.
Task Create an MIP. Display configured MEPs and MIPs. Command Syntax
ethernet cfm mip domain {name | level } ma-name name show ethernet cfm maintenance-points local [mep | mip]

Command Mode INTERFACE EXEC Privilege

FTOS#show ethernet cfm maintenance-points local mip ------------------------------------------------------------------------------MPID Domain Name Level Type Port CCM-Status MA Name VLAN Dir ------------------------------------------------------------------------------0 0 service1 service1 4 4 MIP MIP Gi 0/5 Gi 0/5 Disabled Disabled My_MA Your_MA 3333 3333 DOWN UP

MAC

00:01:e8:0b:c6:36 00:01:e8:0b:c6:36

MP Databases
CFM maintains two MP databases: MEP Database (MEP-DB): Every MEP must maintain a database of all other MEPs in the MA that have announced their presence via CCM. MIP Database (MIP-DB): Every MIP must maintain a database of all other MEPs in the MA that have announced their presence via CCM
Command Syntax
show ethernet cfm maintenance-points remote detail [active | domain {level | name} | expired | waiting]

Task Display the MEP Database.

Command Mode EXEC Privilege

802.1ag | 77

www.dell.com | support.dell.com

Task

Command Syntax

Command Mode

FTOS#show ethernet cfm maintenance-points remote detail MAC Address: 00:01:e8:58:68:78 Domain Name: cfm0 MA Name: test0 Level: 7 VLAN: 10 MP ID: 900 Sender Chassis ID: FTOS MEP Interface status: Up MEP Port status: Forwarding Receive RDI: FALSE MP Status: Active Display the MIP Database.
show ethernet cfm mipdb

EXEC Privilege

MP Database Persistence
Task Set the amount of time that data from a missing MEP is kept in the Continuity Check Database. Command Syntax
database hold-time minutes

Command Mode ECFM DOMAIN

Default: 100 minutes Range: 100-65535 minutes

78

802.1ag

Continuity Check Messages

Continuity Check Messages (CCM) are periodic hellos used to: discover MEPs and MIPs within a maintenance domain detect loss of connectivity between MEPs detect misconfiguration, such as VLAN ID mismatch between MEPs to detect unauthorized MEPs in a maintenance domain

Continuity Check Messages (CCM) are multicast Ethernet frames sent at regular intervals from each MEP. They have a destination address based on the MD level (01:80:C2:00:00:3X where X is the MD level of the transmitting MEP from 0 to 7). All MEPs must listen to these multicast MAC addresses and process these messages. MIPs may optionally processes the CCM messages originated by MEPs and construct a MIP CCM database. MEPs and MIPs filter CCMs from higher and lower domain levels as described in Table 5-1.
Table 5-1. Frames at Less than my level My level Continuity Check Message Processing Frames from Bridge-relay side or Wire side Bridge-relay side Wire side Greater than my level Bridge-relay side or Wire side UP-MEP Action Drop Consume Drop Forward Down-MEP Action MIP Action Drop Drop Consume Forward Drop Add to MIP-DB and forward Forward

All the remote MEPs in the maintenance domain are defined on each MEP. Each MEP then expects a periodic CCM from the configured list of MEPs. A connectivity failure is then defined as: 1. Loss of 3 consecutive CCMs from any of the remote MEP, which indicates a network failure 2. Reception of a CCM with an incorrect CCM transmission interval, which indicates a configuration error. 3. Reception of CCM with an incorrect MEP ID or MAID, which indicates a configuration or cross-connect error. This could happen when different VLANs are cross-connected due to a configuration error. 4. Reception of a CCM with an MD level lower than that of the receiving MEP, which indicates a configuration or cross-connect error. 5. Reception of a CCM containing a port status/interface status TLV, which indicates a failed bridge or aggregated port. The Continuity Check protocol sends fault notifications (Syslogs, and SNMP traps if enabled) whenever any of the above errors are encountered.

802.1ag | 79

www.dell.com | support.dell.com

Enable CCM
Step 1 2 Task Enable CCM. Configure the transmit interval (mandatory). The interval specified applies to all MEPs in the domain. Command Syntax
no ccm disable

Command Mode ECFM DOMAIN ECFM DOMAIN

Default: Disabled ccm transmit-interval seconds Default: 10 seconds

Enable Cross-checking
Task Enable cross-checking. Start the cross-check operation for an MEP. Configure the amount of time the system waits for a remote MEP to come up before the cross-check operation is started. Command Syntax
mep cross-check enable

Command Mode ETHERNET CFM ETHERNET CFM ETHERNET CFM

Default: Disabled mep cross-check mep-id

mep cross-check start-delay number

Loopback Message and Response

Loopback Message and Response (LBM, LBR), also called Layer 2 Ping, is an administrative echo transmitted by MEPs to verify reachability to another MEP or MIP within the maintenance domain. LBM and LBR are unicast frames.
Task Send a Loopback message. Command Syntax
ping ethernet domain name ma-name ma-name remote

Command Mode EXEC Privilege

{mep-id | mac-addr mac-address} source {mep-id | port interface}

Linktrace Message and Response

Linktrace Message and Response (LTM, LTR), also called Layer 2 Traceroute, is an administratively sent multicast frames transmitted by MEPs to track, hop-by-hop, the path to another MEP or MIP within the maintenance domain. All MEPs and MIPs in the same domain respond to an LTM with a unicast LTR. Intermediate MIPs forward the LTM toward the target MEP.

80

802.1ag

Figure 5-4.

Linktrace Message and Response

MPLS Core

MEP

MIP

MIP

MIP

Lin

ktra

c e m M essa

ge

L i n k t ra ce R e s p o n s e

Link trace messages carry a unicast target address (the MAC address of an MIP or MEP) inside a multicast frame. The destination group address is based on the MD level of the transmitting MEP (01:80:C2:00:00:3[8 to F]). The MPs on the path to the target MAC address reply to the LTM with an LTR, and relays the LTM towards the target MAC until the target MAC is reached or TTL equals 0.
Task Send a Linktrace message. Since the LTM is a Multicast message sent to the entire ME, there is no need to specify a destination. Command Syntax
traceroute ethernet domain

Command Mode EXEC Privilege

Link Trace Cache

After a Link Trace command is executed, the trace information can be cached so that you can view it later without retracing.
Task Enable Link Trace caching. Set the amount of time a trace result is cached. Command Syntax
traceroute cache traceroute cache hold-time minutes

Command Mode CONFIGURATION ETHERNET CFM

Default: 100 minutes Range: 10-65535 minutes Set the size of the Link Trace Cache.
traceroute cache size entries

ETHERNET CFM

Default: 100 Range: 1 - 4095 entries Display the Link Trace Cache.
show ethernet cfm traceroute-cache

EXEC Privilege

802.1ag | 81

www.dell.com | support.dell.com

Task
FTOS#show ethernet cfm traceroute-cache

Command Syntax

Command Mode

Traceroute to 00:01:e8:52:4a:f8 on Domain Customer2, Level 7, MA name Test2 with VLAN 2 -----------------------------------------------------------------------------Hops Host IngressMAC Ingr Action Relay Action Next Host -----------------------------------------------------------------------------4 00:00:00:01:e8:53:4a:f8 00:01:e8:52:4a:f8 IngOK RlyHit

Egress MAC Egress Action FWD Status

00:00:00:01:e8:52:4a:f8

Terminal MEP

Delete all Link Trace Cache entries.

clear ethernet cfm traceroute-cache

EXEC Privilege

Enable CFM SNMP Traps

Task Enable SNMP trap messages for Ethernet CFM. Command Syntax
snmp-server enable traps ecfm

Command Mode CONFIGURATION

A Trap is sent only when one of the five highest priority defects occur, as shown in Table 5-2.
Table 5-2. ECFM SNMP Traps

Cross-connect defect Error-CCM defect MAC Status defect Remote CCM defect RDI defect

%ECFM-5-ECFM_XCON_ALARM: Cross connect fault detected by MEP 1 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_ERROR_ALARM: Error CCM Defect detected by MEP 1 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_MAC_STATUS_ALARM: MAC Status Defect detected by MEP 1 in Domain provider at Level 4 VLAN 3000 %ECFM-5-ECFM_REMOTE_ALARM: Remote CCM Defect detected by MEP 3 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_RDI_ALARM: RDI Defect detected by MEP 3 in Domain customer1 at Level 7 VLAN 1000

82

802.1ag

Three values are giving within the trap messages: MD Index, MA Index, and MPID. You can reference these values against the output of show ethernet cfm domain and show ethernet cfm maintenance-points local mep.
FTOS#show ethernet cfm maintenance-points local mep ------------------------------------------------------------------------------MPID Domain Name Level Type Port CCM-Status MA Name VLAN Dir ------------------------------------------------------------------------------100 cfm0 7 MEP Gi 4/10 Enabled test0 10 DOWN

MAC

00:01:e8:59:23:45

FTOS(conf-if-gi-0/6)#do show ethernet cfm domain Domain Name: My_Name MD Index: 1 Level: 0 Total Service: 1 Services MA-Index MA-Name 1 test

VLAN 0

CC-Int 1s

X-CHK Status enabled

Domain Name: Your_Name MD Index: 2 Level: 2 Total Service: 1 Services MA-Index MA-Name 1 test

VLAN 100

CC-Int 1s

X-CHK Status enabled

Display Ethernet CFM Statistics

Task Display MEP CCM statistics. FTOS# Command Syntax
show ethernet cfm statistics [domain {name | level} vlan-id vlan-id mpid mpid

Command Mode EXEC Privilege

show ethernet cfm statistics

Domain Name: Customer Domain Level: 7 MA Name: My_MA MPID: 300 CCMs: Transmitted: LTRs: Unexpected Rcvd: LBRs: Received: Received Bad MSDU: Transmitted:

1503 0 0 0 0

RcvdSeqErrors:

Rcvd Out Of Order:

802.1ag | 83

www.dell.com | support.dell.com

Task Display CFM statistics by port.

Command Syntax
show ethernet cfm port-statistics [interface]

Command Mode EXEC Privilege

FTOS#show ethernet cfm port-statistics interface gigabitethernet 0/5 Port statistics for port: Gi 0/5 ================================== RX Statistics ============= Total CFM Pkts 75394 CCM Pkts 75394 LBM Pkts 0 LTM Pkts 0 LBR Pkts 0 LTR Pkts 0 Bad CFM Pkts 0 CFM Pkts Discarded 0 CFM Pkts forwarded 102417 TX Statistics ============= Total CFM Pkts 10303 CCM Pkts 0 LBM Pkts 0 LTM Pkts 3 LBR Pkts 0 LTR Pkts 0

84

802.1ag

6
Access Control Lists (ACL), Prefix Lists, and Route-maps
Access Control Lists, Prefix Lists, and Route-maps are supported on platforms:

ces

ces Egress IP and MAC ACLs are supported on platforms: e

Ingress IP and MAC ACLs are supported on platforms:

Overview
At their simplest, Access Control Lists (ACLs), Prefix lists, and Route-maps permit or deny traffic based on MAC and/or IP addresses. This chapter discusses implementing IP ACLs, IP Prefix lists and Route-maps. For MAC ACLS, refer to Chapter 20, Layer 2. An ACL is essentially a filter containing some criteria to match (examine IP, TCP, or UDP packets) and an action to take (permit or deny). ACLs are processed in sequence so that if a packet does not match the criterion in the first filter, the second filter (if configured) is applied. When a packet matches a filter, the switch drops or forwards the packet based on the filters specified action. If the packet does not match any of the filters in the ACL, the packet is dropped (implicit deny). The number of ACLs supported on a system depends on your CAM size. See CAM Profiling, CAM Allocation, and CAM Optimization in this chapter for more information. Refer to Chapter 9, Content Addressable Memory for complete CAM profiling information. This chapter covers the following topics: IP Access Control Lists (ACLs) CAM Profiling, CAM Allocation, and CAM Optimization Implementing ACLs on FTOS IP Fragment Handling Configure a standard IP ACL Configure an extended IP ACL Configuring Layer 2 and Layer 3 ACLs on an Interface Assign an IP ACL to an Interface

Access Control Lists (ACL), Prefix Lists, and Route-maps | 85

www.dell.com | support.dell.com

Configuring Ingress ACLs Configuring Egress ACLs Configuring ACLs to Loopback Applying an ACL on Loopback Interfaces IP Prefix Lists ACL Resequencing Route Maps

IP Access Control Lists (ACLs)

In the Dell Networking switch/routers, you can create two different types of IP ACLs: standard or extended. A standard ACL filters packets based on the source IP packet. An extended ACL filters traffic based on the following criteria (for more information on ACL supported options see the FTOS Command Reference): IP protocol number Source IP address Destination IP address Source TCP port number Destination TCP port number Source UDP port number Destination UDP port number

For extended ACL TCP and UDP filters, you can match criteria on specific or ranges of TCP or UDP ports. For extended ACL TCP filters, you can also match criteria on established TCP sessions. When creating an access list, the sequence of the filters is important. You have a choice of assigning sequence numbers to the filters as you enter them, or FTOS will assign numbers in the order the filters are created. The sequence numbers, whether configured or assigned by FTOS, are listed in the show config and show ip accounting access-list command display output. Ingress and egress Hot Lock ACLs allow you to append or delete new rules into an existing ACL (already written into CAM) without disrupting traffic flow. Existing entries in CAM are shuffled to accommodate the new entries. Hot Lock ACLs are enabled by default and support both standard and extended ACLs and on all platforms.
Note: Hot Lock ACLs are supported for Ingress ACLs only.

CAM Profiling, CAM Allocation, and CAM Optimization

CAM Profiling is supported on platform

e c and

User Configurable CAM Allocations are supported on platform

86

Access Control Lists (ACL), Prefix Lists, and Route-maps

CAM optimization is supported on platforms

cs et

CAM Profiling
CAM optimization is supported on platforms

The default CAM profile has 1K Layer 2 ingress ACL entries. If you need more memory for Layer 2 ingress ACLs, select the profile l2-ipv4-inacl. When budgeting your CAM allocations for ACLs and QoS configurations, remember that ACL and QoS rules might consume more than one CAM entry depending on complexity. For example, TCP and UDP rules with port range options might require more than one CAM entry. The Layer 2 ACL CAM partition has sub-partitions for several types of information. Table 6-1 lists the sub-partition and the percentage of the Layer 2 ACL CAM partition that FTOS allocates to each by default.
Table 6-1. Partition Sysflow L2ACL *PVST QoS L2PT FRRP Layer 2 ACL CAM Sub-partition Sizes % Allocated 6 14 50 12 13 5

You can re-configure the amount of space, in percentage, allocated to each sub-partition. As with the IPv4Flow partition, you can configure the Layer 2 ACL partition from EXEC Privilege mode or CONFIGURATION mode. The amount of space that you can distribute to the sub-partitions is equal to the amount of CAM space that the selected CAM profile allocates to the Layer 2 ACL partition. FTOS requires that you specify the amount of CAM space for all sub-partitions and that the sum of all sub-partitions is 100%. FTOS displays the following message if the total allocated space is not correct:
% Error: Sum of all regions does not total to 100%.

User Configurable CAM Allocation

User Configurable CAM Allocations are supported on platform

c and

Allocate space for IPV6 ACLs on the by using the cam-acl command in CONFIGURATION mode.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 87

www.dell.com | support.dell.com

The CAM space is allotted in FP blocks. The total space allocated must equal 13 FP blocks. Note that there are 16 FP blocks, but the System Flow requires 3 blocks that cannot be reallocated. The default CAM Allocation settings on a C-Series matching are: L3 ACL (ipv4acl): 6 L2 ACL(l2acl) : 5 IPv6 L3 ACL (ipv6acl): 0 L3 QoS (ipv4qos): 1 L2 QoS (l2qos): 1

The ipv6acl allocation must be entered as a factor of 2 (2, 4, 6, 8, 10). All other profile allocations can use either even or odd numbered ranges. You must save the new CAM settings to the startup-config (write-mem or copy run start) then reload the system for the new settings to take effect.

CAM optimization
CAM optimization is supported on platforms

cs

When this command is enabled, if a Policy Map containing classification rules (ACL and/or dscp/ ip-precedence rules) is applied to more than one physical interface on the same port-pipe, only a single copy of the policy is written (only 1 FP entry will be used). When the command is disabled, the system behaves as described in this chapter.

Test CAM Usage

The test cam-usage command is supported on platforms

ces

This command applies to both IPv4 and IPv6 CAM profiles, but is best used when verifying QoS optimization for IPv6 ACLs. Use this command to determine whether sufficient ACL CAM space is available to enable a service-policy. Create a Class Map with all required ACL rules, then execute the test cam-usage command in Privilege mode to verify the actual CAM space required. Figure 6-1 gives a sample of the output shown when executing the command. The status column indicates whether or not the policy can be enabled.
Figure 6-1. Command Example: test cam-usage (C-Series)

FTOS#test cam-usage service-policy input TestPolicy linecard all Linecard | Portpipe | CAM Partition | Available CAM | Estimated CAM per Port | Status -----------------------------------------------------------------------------------------2 | 1 | IPv4Flow | 232 | 0 | Allowed 2 | 1 | IPv6Flow | 0 | 0 | Allowed 4 | 0 | IPv4Flow | 232 | 0 | Allowed 4 | 0 | IPv6Flow | 0 | 0 | Allowed FTOS#

88

Access Control Lists (ACL), Prefix Lists, and Route-maps

Implementing ACLs on FTOS

One IP ACL can be assigned per interface with FTOS. If an IP ACL is not assigned to an interface, it is not used by the software in any other capacity. The number of entries allowed per ACL is hardware-dependent. Refer to your line card documentation for detailed specification on entries allowed per ACL. If counters are enabled on IP ACL rules that are already configured, those counters are reset when a new rule is inserted or prepended. If a rule is appended, the existing counters are not affected. This is applicable to the following features:
V

L2 Ingress Access list L2 Egress Access list L3 Ingress Access list L3 Egress Access list

Note: IP ACLs are supported over VLANs in Version 6.2.1.1 and higher.

ACLs and VLANs

There are some differences when assigning ACLs to a VLAN rather than a physical port. For example, when using a single port-pipe, if you apply an ACL to a VLAN, one copy of the ACL entries would get installed in the ACL CAM on the port-pipe. The entry would look for the incoming VLAN in the packet. Whereas if you apply an ACL on individual ports of a VLAN, separate copies of the ACL entries would be installed for each port belonging to a port-pipe. When you use the log keyword, CP processor will have to log details about the packets that match. Depending on how many packets match the log entry and at what rate, CP might become busy as it has to log these packets details. However the other processors (RP1 and RP2) should be unaffected. This option is typically useful when debugging some problem related to control traffic. We have used this option numerous times in the field and have not encountered any problems in such usage so far.

ACL Optimization
If an access list contains duplicate entries, FTOS deletes one entry to conserve CAM space. Standard and Extended ACLs take up the same amount of CAM space. A single ACL rule uses 2 CAM entries whether it is identified as a Standard or Extended ACL.

Determine the order in which ACLs are used to classify traffic

When you link class-maps to queues using the command service-queue, FTOS matches the class-maps according to queue priority (queue numbers closer to 0 have lower priorities). For example, in Figure 6-2, class-map cmap2 is matched against ingress packets before cmap1.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 89

www.dell.com | support.dell.com

ACLs acl1 and acl2 have overlapping rules because the address range 20.1.1.0/24 is within 20.0.0.0/8. Therefore, (without the keyword order) packets within the range 20.1.1.0/24 match positive against cmap1 and are buffered in queue 7, though you intended for these packets to match positive against cmap2 and be buffered in queue 4. In cases such as these, where class-maps with overlapping ACL rules are applied to different queues, use the order keyword to specify the order in which you want to apply ACL rules, as shown in Figure 6-2. The order can range from 0 to 254. FTOS writes to the CAM ACL rules with lower order numbers (order numbers closer to 0) before rules with higher order numbers so that packets are matched as you intended. By default, all ACL rules have an order of 254.
Figure 6-2. Using the Order Keyword in ACLs

FTOS(conf)#ip access-list standard acl1 FTOS(config-std-nacl)#permit 20.0.0.0/8 FTOS(config-std-nacl)#exit FTOS(conf)#ip access-list standard acl2 FTOS(config-std-nacl)#permit 20.1.1.0/24 order 0 FTOS(config-std-nacl)#exit FTOS(conf)#class-map match-all cmap1 FTOS(conf-class-map)#match ip access-group acl1 FTOS(conf-class-map)#exit FTOS(conf)#class-map match-all cmap2 FTOS(conf-class-map)#match ip access-group acl2 FTOS(conf-class-map)#exit FTOS(conf)#policy-map-input pmap FTOS(conf-policy-map-in)#service-queue 7 class-map cmap1 FTOS(conf-policy-map-in)#service-queue 4 class-map cmap2 FTOS(conf-policy-map-in)#exit FTOS(conf)#interface gig 1/0 FTOS(conf-if-gi-1/0)#service-policy input pmap

IP Fragment Handling
FTOS supports a configurable option to explicitly deny IP fragmented packets, particularly second and subsequent packets. It extends the existing ACL command syntax with the fragments keyword for all Layer 3 rules applicable to all Layer protocols (permit/deny ip/tcp/udp/icmp). Both standard and extended ACLs support IP fragments. Second and subsequent fragments are allowed because a Layer 4 rule cannot be applied to these fragments. If the packet is to be denied eventually, the first fragment would be denied and hence the packet as a whole cannot be reassembled. Implementing the required rules will use a significant number of CAM entries per TCP/UDP entry. For IP ACL, FTOS always applies implicit deny. You do not have to configure it. For IP ACL, FTOS applies implicit permit for second and subsequent fragment just prior to the implicit deny. If an explicit deny is configured, the second and subsequent fragments will not hit the implicit permit rule for fragments.

90

Access Control Lists (ACL), Prefix Lists, and Route-maps

Loopback interfaces do not support ACLs using the IP fragment option. If you configure an ACL with the fragments option and apply it to a loopback interface, the command is accepted, but the ACL entries are not actually installed the offending rule in CAM.

IP fragments ACL examples

The following configuration permits all packets (both fragmented & non-fragmented) with destination IP 10.1.1.1. The second rule does not get hit at all.
FTOS(conf)#ip access-list extended ABC FTOS(conf-ext-nacl)#permit ip any 10.1.1.1/32 FTOS(conf-ext-nacl)#deny ip any 10.1.1.1./32 fragments FTOS(conf-ext-nacl)

To deny second/subsequent fragments, use the same rules in a different order. These ACLs deny all second & subsequent fragments with destination IP 10.1.1.1 but permit the first fragment & non fragmented packets with destination IP 10.1.1.1 .
FTOS(conf)#ip access-list extended ABC FTOS(conf-ext-nacl)#deny ip any 10.1.1.1/32 fragments FTOS(conf-ext-nacl)#permit ip any 10.1.1.1/32 FTOS(conf-ext-nacl)

Layer 4 ACL rules examples

In the below scenario, first fragments non-fragmented TCP packets from 10.1.1.1 with TCP destination port equal to 24 are permitted. All other fragments are denied.
FTOS(conf)#ip access-list extended ABC FTOS(conf-ext-nacl)#permit tcp host 10.1.1.1 any eq 24 FTOS(conf-ext-nacl)#deny ip any any fragment FTOS(conf-ext-nacl)

In the following, TCP packets that are first fragments or non-fragmented from host 10.1.1.1 with TCP destination port equal to 24 are permitted. Additionally, all TCP non-first fragments from host 10.1.1.1 are permitted. All other IP packets that are non-first fragments are denied.
FTOS(conf)#ip access-list extended ABC FTOS(conf-ext-nacl)#permit tcp host 10.1.1.1 any eq 24 FTOS(conf-ext-nacl)#permit tcp host 10.1.1.1 any fragment FTOS(conf-ext-nacl)#deny ip any any fragment FTOS(conf-ext-nacl)

Access Control Lists (ACL), Prefix Lists, and Route-maps | 91

www.dell.com | support.dell.com

To log all the packets denied and to override the implicit deny rule and the implicit permit rule for TCP/ UDP fragments, use a configuration similar to the following.
FTOS(conf)#ip access-list extended ABC FTOS(conf-ext-nacl)#permit tcp any any fragment FTOS(conf-ext-nacl)#permit udp any any fragment FTOS(conf-ext-nacl)#deny ip any any log FTOS(conf-ext-nacl)

Note the following when configuring ACLs with the fragments keyword. When an ACL filters packets it looks at the Fragment Offset (FO) to determine whether or not it is a fragment. FO = 0 means it is either the first fragment or the packet is a non-fragment. FO > 0 means it is dealing with the fragments of the original packet. Permit ACL line with L3 information only, and the fragments keyword is present: If a packet's L3 information matches the L3 information in the ACL line, the packet's fragment offset (FO) is checked. If a packet's FO > 0, the packet is permitted. If a packet's FO = 0 , the next ACL entry is processed. Deny ACL line with L3 information only, and the fragments keyword is present: If a packet's L3 information does match the L3 information in the ACL line, the packet's fragment offset (FO) is checked. If a packet's FO > 0, the packet is denied. If a packet's FO = 0, the next ACL line is processed.

Configure a standard IP ACL

To configure an ACL, use commands in the IP ACCESS LIST mode and the INTERFACE mode. The following list includes the configuration tasks for IP ACLs: For a complete listing of all commands related to IP ACLs, refer to the FTOS Command Line Interface Reference document. Refer to Configure an extended IP ACL to set up extended ACLs. A standard IP ACL uses the source IP address as its match criterion. To configure a standard IP ACL, use these commands in the following sequence:
Step 1 Command Syntax
ip access-list standard access-listname

Command Mode CONFIGURATION

Purpose Enter IP ACCESS LIST mode by naming a standard IP access list.

92

Access Control Lists (ACL), Prefix Lists, and Route-maps

Step

Command Syntax
seq sequence-number {deny | permit} {source [mask] | any | host ip-address} [count [byte] | log ] [order] [monitor] [fragments]

Command Mode CONFIG-STD-NACL

Purpose Configure a drop or forward filter. The parameters are: log and monitor options are supported on E-Series only.

Note: When assigning sequence numbers to filters, keep in mind that you might need to insert a new filter. To prevent reconfiguring multiple filters, assign sequence numbers in multiples of five or another number.

When you use the log keyword, CP processor logs details about the packets that match. Depending on how many packets match the log entry and at what rate, the CP may become busy as it has to log these packets details. To view the rules of a particular ACL configured on a particular interface, use the show ip accounting access-list ACL-name interface interface command (Figure 226) in EXEC Privilege mode.
Figure 6-3. Command Example: show ip accounting access-list
FTOS#show ip accounting access ToOspf interface gig 1/6 Standard IP access list ToOspf seq 5 deny any seq 10 deny 10.2.0.0 /16 seq 15 deny 10.3.0.0 /16 seq 20 deny 10.4.0.0 /16 seq 25 deny 10.5.0.0 /16 seq 30 deny 10.6.0.0 /16 seq 35 deny 10.7.0.0 /16 seq 40 deny 10.8.0.0 /16 seq 45 deny 10.9.0.0 /16 seq 50 deny 10.10.0.0 /16 FTOS#

Figure 6-4 illustrates how the seq command orders the filters according to the sequence number assigned. In the example, filter 25 was configured before filter 15, but the show config command displays the filters in the correct order.
Figure 6-4. Command example: seq

FTOS(config-std-nacl)#seq 25 deny ip host 10.5.0.0 any log FTOS(config-std-nacl)#seq 15 permit tcp 10.3.0.0 /16 any FTOS(config-std-nacl)#show config ! ip access-list standard dilling seq 15 permit tcp 10.3.0.0/16 any seq 25 deny ip host 10.5.0.0 any log FTOS(config-std-nacl)#

To delete a filter, use the no seq sequence-number command in the IP ACCESS LIST mode. If you are creating a standard ACL with only one or two filters, you can let FTOS assign a sequence number based on the order in which the filters are configured. The software assigns filters in multiples of 5.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 93

www.dell.com | support.dell.com

To configure a filter without a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 2 Command Syntax
ip access-list standard
access-list-name

Command Mode CONFIGURATION CONFIG-STD-NACL

Purpose Create a standard IP ACL and assign it a unique name. Configure a drop or forward IP ACL filter. log and monitor options are supported on E-Series only.

{deny | permit} {source [mask] | any | host ip-address} [count [byte] | log ] [order] [monitor] [fragments]

When you use the log keyword, CP processor logs details about the packets that match. Depending on how many packets match the log entry and at what rate, the CP may become busy as it has to log these packets details. Figure 6-5 illustrates a standard IP ACL in which the sequence numbers were assigned by the FTOS. The filters were assigned sequence numbers based on the order in which they were configured (for example, the first filter was given the lowest sequence number). The show config command in the IP ACCESS LIST mode displays the two filters with the sequence numbers 5 and 10.
Figure 6-5. Standard IP ACL

FTOS(config-route-map)#ip access standard kigali FTOS(config-std-nacl)#permit 10.1.0.0/16 FTOS(config-std-nacl)#show config ! ip access-list standard kigali seq 5 permit 10.1.0.0/16 FTOS(config-std-nacl)#

To view all configured IP ACLs, use the show ip accounting access-list command (Figure 229) in the EXEC Privilege mode.
Figure 6-6. Command Example: show ip accounting access-list

FTOS#show ip accounting access example interface gig 4/12 Extended IP access list example seq 10 deny tcp any any eq 111 seq 15 deny udp any any eq 111 seq 20 deny udp any any eq 2049 seq 25 deny udp any any eq 31337 seq 30 deny tcp any any range 12345 12346 seq 35 permit udp host 10.21.126.225 10.4.5.0 /28 seq 40 permit udp host 10.21.126.226 10.4.5.0 /28 seq 45 permit udp 10.8.0.0 /16 10.50.188.118 /31 range 1812 1813 seq 50 permit tcp 10.8.0.0 /16 10.50.188.118 /31 eq 49 seq 55 permit udp 10.15.1.0 /24 10.50.188.118 /31 range 1812 1813

To delete a filter, enter the show config command in the IP ACCESS LIST mode and locate the sequence number of the filter you want to delete. Then use the no seq sequence-number command in the IP ACCESS LIST mode.

94

Access Control Lists (ACL), Prefix Lists, and Route-maps

Configure an extended IP ACL

Extended IP ACLs filter on source and destination IP addresses, IP host addresses, TCP addresses, TCP host addresses, UDP addresses, and UDP host addresses. Since traffic passes through the filter in the order of the filters sequence, you can configure the extended IP ACL by first entering the IP ACCESS LIST mode and then assigning a sequence number to the filter.
Note: On E-Series ExaScale systems, TCP ACL flags are not supported in an extended ACL with IPv6 microcode. An error message is shown if IPv6 microcode is configured and an ACL is entered with a TCP filter included.
FTOS(conf-ipv6-acl)#seq 8 permit tcp any any urg May 5 08:32:34: %E90MJ:0 %ACL_AGENT-2-ACL_AGENT_ENTRY_ERROR: Unable to write seq 8 of list test as individual TCP flags are not supported on linecard 0

Configure filters with sequence number

To create a filter for packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip access-list extended access-list-name seq sequence-number {deny | permit}

Command Mode CONFIGURATION CONFIG-EXT-NACL

Purpose Enter the IP ACCESS LIST mode by creating an extended IP ACL. Configure a drop or forward filter. log and monitor options are supported on E-Series only.

{ip-protocol-number | icmp | ip | tcp | udp} {source mask | any | host ip-address} {destination mask | any | host ip-address} [operator port [port]] [count [byte] | log ] [order] [monitor] [fragments]

When you use the log keyword, CP processor logs details about the packets that match. Depending on how many packets match the log entry and at what rate, the CP may become busy as it has to log these packets details. TCP packets: To create a filter for TCP packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip access-list extended
access-list-name

Command Mode CONFIGURATION

Purpose Create an extended IP ACL and assign it a unique name.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 95

www.dell.com | support.dell.com

Step

Command Syntax
seq

Command Mode CONFIG-EXT-NACL

Purpose Configure an extended IP ACL filter for TCP packets. log and monitor options are supported on E-Series only.

sequence-number {deny | permit} tcp {source mask | any | host ip-address}} [count [byte] | log ] [order] [monitor] [fragments]

When you use the log keyword, CP processor logs details about the packets that match. Depending on how many packets match the log entry and at what rate, the CP may become busy as it has to log these packets details. UDP packets: To create a filter for UDP packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip access-list extended
access-list-name

Command Mode CONFIGURATION CONFIG-EXT-NACL

Purpose Create a extended IP ACL and assign it a unique name. Configure an extended IP ACL filter for UDP packets. log and monitor options are supported on E-Series only.

seq

sequence-number {deny | permit} {ip-protocol-number udp} {source mask | any | host ip-address} {destination mask | any | host ip-address} [operator port [port]] [count [byte] | log ] [order] [monitor] [fragments]

When you create the filters with a specific sequence number, you can create the filters in any order and the filters are placed in the correct order.
Note: When assigning sequence numbers to filters, keep in mind that you might need to insert a new filter. To prevent reconfiguring multiple filters, assign sequence numbers in multiples of five or another number.

Figure 6-7 illustrates how the seq command orders the filters according to the sequence number assigned. In the example, filter 15 was configured before filter 5, but the show config command displays the filters in the correct order.

96

Access Control Lists (ACL), Prefix Lists, and Route-maps

Figure 6-7.

Command Example: seq

FTOS(config-ext-nacl)#seq 15 deny ip host 112.45.0.0 any log FTOS(config-ext-nacl)#seq 5 permit tcp 12.1.3.45 0.0.255.255 any FTOS(config-ext-nacl)#show confi ! ip access-list extended dilling seq 5 permit tcp 12.1.0.0 0.0.255.255 any seq 15 deny ip host 112.45.0.0 any log FTOS(config-ext-nacl)#

Configure filters without sequence number

If you are creating an extended ACL with only one or two filters, you can let FTOS assign a sequence number based on the order in which the filters are configured. FTOS assigns filters in multiples of 5. To configure a filter for an extended IP ACL without a specified sequence number, use any or all of the following commands in the IP ACCESS LIST mode:
Command Syntax Command Mode CONFIG-EXT-NACL Purpose Configure a deny or permit filter to examine IP packets. log and monitor options are supported on E-Series only. Configure a deny or permit filter to examine TCP packets. log and monitor options are supported on E-Series only. Configure a deny or permit filter to examine UDP packets. log and monitor options are supported on E-Series only.

{deny | permit} {source mask | any | host ip-address} [count [byte] | log ] [order] [monitor] [fragments] {deny | permit} tcp {source mask] | any | host ip-address}} [count [byte] | log ] [order] [monitor] [fragments] {deny | permit} udp {source mask | any | host ip-address}} [count [byte] | log ] [order] [monitor] [fragments]

CONFIG-EXT-NACL

CONFIG-EXT-NACL

When you use the log keyword, CP processor logs details about the packets that match. Depending on how many packets match the log entry and at what rate, the CP may become busy as it has to log these packets details. Figure 6-8 illustrates an extended IP ACL in which the sequence numbers were assigned by the software. The filters were assigned sequence numbers based on the order in which they were configured (for example, the first filter was given the lowest sequence number). The show config command in the IP ACCESS LIST mode displays the two filters with the sequence numbers 5 and 10.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 97

www.dell.com | support.dell.com

Figure 6-8.

Extended IP ACL

FTOS(config-ext-nacl)#deny tcp host 123.55.34.0 any FTOS(config-ext-nacl)#permit udp 154.44.123.34 0.0.255.255 host 34.6.0.0 FTOS(config-ext-nacl)#show config ! ip access-list extended nimule seq 5 deny tcp host 123.55.34.0 any seq 10 permit udp 154.44.0.0 0.0.255.255 host 34.6.0.0 FTOS(config-ext-nacl)#

To view all configured IP ACLs and the number of packets processed through the ACL, use the show ip accounting access-list command (Figure 232) in the EXEC Privilege mode.

Configuring Layer 2 and Layer 3 ACLs on an Interface

Both Layer 2 and Layer 3 ACLs may be configured on an interface in Layer 2 mode. If both L2 and L3 ACLs are applied to an interface, the following rules apply: The packets routed by FTOS are governed by the L3 ACL only, since they are not filtered against an L2 ACL. The packets switched by FTOS are first filtered by the L3 ACL, then by the L2 ACL. When packets are switched by FTOS, the egress L3 ACL does not filter the packet.

For the following features, if counters are enabled on rules that have already been configured and a new rule is either inserted or prepended, all the existing counters will be reset: L2 Ingress Access list L3 Egress Access list L2 Egress Access list

If a rule is simply appended, existing counters are not affected.

Table 6-2. L2 and L3 ACL Filtering on Switched Packets L3 ACL Behavior Deny Permit Deny Permit Decision on Targeted Traffic Denied by L3 ACL Permitted by L3 ACL Denied by L2 ACL Permitted by L2 ACL

L2 ACL Behavior Deny Deny Permit Permit

Note: If an interface is configured as a vlan-stack access port, the packets are filtered by an L2 ACL only. The L3 ACL applied to such a port does not affect traffic. That is, existing rules for other features (such as trace-list, PBR, and QoS) are applied accordingly to the permitted traffic.

98

Access Control Lists (ACL), Prefix Lists, and Route-maps

For information on MAC ACLs, refer to Chapter 20, Layer 2, on page 409.

Assign an IP ACL to an Interface

c s (S25/S50) Ingress and Egress IP ACL are supported on platforms: e

Ingress IP ACLs are supported on platforms: To pass traffic through a configured IP ACL, you must assign that ACL to a physical interface, a port channel interface, or a VLAN. The IP ACL is applied to all traffic entering a physical or port channel interface and the traffic is either forwarded or dropped depending on the criteria and actions specified in the ACL. The same ACL may be applied to different interfaces and that changes its functionality. For example, you can take ACL "ABCD", and apply it using the in keyword and it becomes an ingress access list. If you apply the same ACL using the out keyword, it becomes an egress access list. If you apply the same ACL to the loopback interface, it becomes a loopback access list. This chapter covers the following topics: Configuring Ingress ACLs Configuring Egress ACLs Configuring ACLs to Loopback

For more information on Layer-3 interfaces, refer to Chapter 15, Interfaces. To apply an IP ACL (standard or extended) to a physical or port channel interface, use these commands in the following sequence in the INTERFACE mode:
Step 1 2 Command Syntax
interface interface slot/port ip address ip-address ip access-group access-list-name {in | out} [implicit-permit] [vlan vlan-range]

Command Mode CONFIGURATION INTERFACE INTERFACE

Purpose Enter the interface number. Configure an IP address for the interface, placing it in Layer-3 mode. Apply an IP ACL to traffic entering or exiting an interface. out: configure the ACL to filter outgoing traffic. This keyword is supported only on E-Series. Note: The number of entries allowed per ACL is hardware-dependent. Refer to your line card documentation for detailed specification on entries allowed per ACL. Apply rules to the new ACL.

ip access-list [standard | extended] name

INTERFACE

Access Control Lists (ACL), Prefix Lists, and Route-maps | 99

www.dell.com | support.dell.com

To view which IP ACL is applied to an interface, use the show config command (Figure 232) in the INTERFACE mode or the show running-config command in the EXEC mode.
Figure 6-9. Command example: show config in the INTERFACE Mode
FTOS(conf-if)#show conf ! interface GigabitEthernet 0/0 ip address 10.2.1.100 255.255.255.0 ip access-group nimule in no shutdown FTOS(conf-if)#

Use only Standard ACLs in the access-class command to filter traffic on Telnet sessions.

Counting ACL Hits

You can view the number of packets matching the ACL by using the count option when creating ACL entries. E-Series supports packet and byte counts simultaneously. C-Series and S-Series support only one at any given time. To view the number of packets matching an ACL that is applied to an interface:
Step 1 2 3 Task Create an ACL that uses rules with the count option. See Configure a standard IP ACL Apply the ACL as an inbound or outbound ACL on an interface. See Assign an IP ACL to an Interface View the number of packets matching the ACL using the show ip accounting access-list from EXEC Privilege mode.

Configuring Ingress ACLs

Ingress ACLs are applied to interfaces and to traffic entering the system.These system-wide ACLs eliminate the need to apply ACLs onto each interface and achieves the same results. By localizing target traffic, it is a simpler implementation. To create an ingress ACLs, use the ip access-group command (Figure 233) in the EXEC Privilege mode. This example also shows applying the ACL, applying rules to the newly created access group, and viewing the access list:

100

Access Control Lists (ACL), Prefix Lists, and Route-maps

Figure 6-10. Creating an Ingress ACL

FTOS(conf)#interface gige 0/0 FTOS(conf-if-gige0/0)#ip access-group abcd in FTOS(conf-if-gige0/0)#show config ! gigethernet 0/0 no ip address ip access-group abcd in no shutdown FTOS(conf-if-gige0/0)#end FTOS#configure terminal FTOS(conf)#ip access-list extended abcd FTOS(config-ext-nacl)#permit tcp any any FTOS(config-ext-nacl)#deny icmp any any FTOS(config-ext-nacl)#permit 1.1.1.2 FTOS(config-ext-nacl)#end FTOS#show ip accounting access-list ! Extended Ingress IP access list abcd on gigethernet 0/0 seq 5 permit tcp any any seq 10 deny icmp any any permit 1.1.1.2

Use the in keyword to specify ingress.

Begin applying rules to the ACL named abcd.

View the access-list.

Configuring Egress ACLs

Egress ACLs are supported on platforms

e and the

Egress ACLs are applied to line cards and affect the traffic leaving the system. Configuring egress ACLs onto physical interfaces protects the system infrastructure from attackmalicious and incidentalby explicitly allowing only authorized traffic.These system-wide ACLs eliminate the need to apply ACLs onto each interface and achieves the same results. By localizing target traffic, it is a simpler implementation. An egress ACL is used when users would like to restrict egress traffic. For example, when a DOS attack traffic is isolated to one particular interface, you can apply an egress ACL to block that particular flow from exiting the box, thereby protecting downstream devices. To create an egress ACLs, use the ip access-group command (Figure 234) in the EXEC Privilege mode. This example also shows viewing the configuration, applying rules to the newly created access group, and viewing the access list:

Access Control Lists (ACL), Prefix Lists, and Route-maps | 101

www.dell.com | support.dell.com

Figure 6-11.

Creating an Egress ACL

FTOS(conf)#interface gige 0/0 FTOS(conf-if-gige0/0)#ip access-group abcd out FTOS(conf-if-gige0/0)#show config ! gigethernet 0/0 no ip address ip access-group abcd out no shutdown FTOS(conf-if-gige0/0)#end FTOS#configure terminal FTOS(conf)#ip access-list extended abcd FTOS(config-ext-nacl)#permit tcp any any FTOS(config-ext-nacl)#deny icmp any any FTOS(config-ext-nacl)#permit 1.1.1.2 FTOS(config-ext-nacl)#end FTOS#show ip accounting access-list ! Extended Ingress IP access list abcd on gigethernet 0/0 seq 5 permit tcp any any seq 10 deny icmp any any permit 1.1.1.2

Use the out keyword to specify egress.

Begin applying rules to the ACL named abcd. View the access-list.

Egress Layer 3 ACL Lookup for Control-plane IP Traffic

By default, packets originated from the system are not filtered by egress ACLs. If you initiate a ping session from the system, for example, and apply an egress ACL to block this type of traffic on the interface, the ACL does not affect that ping traffic. The Control Plane Egress Layer 3 ACL feature enhances IP reachability debugging by implementing control-plane ACLs for CPU-generated and CPU-forwarded traffic. Using permit rules with the count option, you can track on a per-flow basis whether CPU-generated and CPU-forwarded packets were transmitted successfully..
Task Apply Egress ACLs to IPv4 system traffic. Apply Egress ACLs to IPv6 system traffic. Create a Layer 3 ACL using permit rules with the count option to describe the desired CPU traffic Command Syntax
ip control-plane [egress filter] ipv6 control-plane [egress filter] permit ip {source mask | any | host ip-address} {destination mask | any | host ip-address} count

Command Mode CONFIGURATION CONFIGURATION CONFIG-NACL

Note: The ip control-plane [egress filter] and the ipv6 control-plane [egress filter] commands are not supported on S60 systems.

FTOS Behavior: VRRP hellos and IGMP packets are not affected when egress ACL filtering for CPU traffic is enabled. Packets sent by the CPU with the source address as the VRRP virtual IP address have the interface MAC address instead of VRRP virtual MAC address.

102

Access Control Lists (ACL), Prefix Lists, and Route-maps

Configuring ACLs to Loopback

ACLs can be supplied on Loopback interfaces supported on platform

Configuring ACLs onto the CPU in a loopback interface protects the system infrastructure from attack malicious and incidentalby explicitly allowing only authorized traffic. The ACLs on loopback interfaces are applied only to the CPU on the RPMthis eliminates the need to apply specific ACLs onto all ingress interfaces and achieves the same results. By localizing target traffic, it is a simpler implementation. The ACLs target and handle Layer 3 traffic destined to terminate on the system including routing protocols, remote access, SNMP, ICMP, and etc. Effective filtering of Layer 3 traffic from Layer 3 routers reduces the risk of attack.
Note: Loopback ACLs are supported only on ingress traffic.

Loopback interfaces do not support ACLs using the IP fragment option. If you configure an ACL with the fragments option and apply it to a loopback interface, the command is accepted, but the ACL entries are not actually installed the offending rule in CAM. See also Loopback Interfaces in the Interfaces chapter.

Applying an ACL on Loopback Interfaces

ACLs can be applied on Loopback interfaces supported on platform

To apply an ACL (standard or extended) for loopback, use these commands in the following sequence:
Step 1 Command Syntax
interface loopback 0 ip access-list [standard | extended] name ip access-group name in

Command Mode CONFIGURATION CONFIGURATION INTERFACE

Purpose Only loopback 0 is supported for the loopback ACL. Apply rules to the new ACL. Apply an ACL to traffic entering loopback. in: configure the ACL to filter incoming traffic Note: ACLs for loopback can only be applied to incoming traffic.

2 3

To apply ACLs on loopback, use the ip access-group command (Figure 235) in the INTERFACE mode. This example also shows the interface configuration status, adding rules to the access group, and displaying the list of rules in the ACL:

Access Control Lists (ACL), Prefix Lists, and Route-maps | 103

www.dell.com | support.dell.com

Figure 6-12. Applying an ACL to the Loopback Interface

FTOS(conf)#interface loopback 0 FTOS(conf-if-lo-0)#ip access-group abcd in FTOS(conf-if-lo-0)#show config ! interface Loopback 0 no ip address ip access-group abcd in no shutdown FTOS(conf-if-lo-0)#end FTOS#configure terminal FTOS(conf)#ip access-list extended abcd FTOS(config-ext-nacl)#permit tcp any any FTOS(config-ext-nacl)#deny icmp any any FTOS(config-ext-nacl)#permit 1.1.1.2 FTOS(config-ext-nacl)#end FTOS#show ip accounting access-list ! Extended Ingress IP access list abcd on Loopback 0 seq 5 permit tcp any any seq 10 deny icmp any any seq 10 deny icmp any any

Use the in keyword.

Add rules to the ACL named abcd. Display the ACL.

Note: See also the section VTY Line Local Authentication and Authorization.

IP Prefix Lists
Prefix Lists are supported on platforms:

ces

IP prefix lists control routing policy. An IP prefix list is a series of sequential filters that contain a matching criterion (examine IP route prefix) and an action (permit or deny) to process routes. The filters are processed in sequence so that if a route prefix does not match the criterion in the first filter, the second filter (if configured) is applied. When the route prefix matches a filter, FTOS drops or forwards the packet based on the filters designated action. If the route prefix does not match any of the filters in the prefix list, the route is dropped (that is, implicit deny). A route prefix is an IP address pattern that matches on bits within the IP address. The format of a route prefix is A.B.C.D/X where A.B.C.D is a dotted-decimal address and /X is the number of bits that should be matched of the dotted decimal address. For example, in 112.24.0.0/16, the first 16 bits of the address 112.24.0.0 match all addresses between 112.24.0.0 to 112.24.255.255. Below are some examples that permit or deny filters for specific routes using the le and ge parameters, where x.x.x.x/x represents a route prefix: To deny only /8 prefixes, enter deny x.x.x.x/x ge 8 le 8 To permit routes with the mask greater than /8 but less than /12, enter permit x.x.x.x/x ge 8 le 12 To deny routes with a mask less than /24, enter deny x.x.x.x/x le 24 To permit routes with a mask greater than /20, enter permit x.x.x.x/x ge 20

104

Access Control Lists (ACL), Prefix Lists, and Route-maps

The following rules apply to prefix lists: A prefix list without any permit or deny filters allows all routes. An implicit deny is assumed (that is, the route is dropped) for all route prefixes that do not match a permit or deny filter in a configured prefix list. Once a route matches a filter, the filters action is applied. No additional filters are applied to the route.

Implementation Information
In FTOS, prefix lists are used in processing routes for routing protocols (for example, RIP, OSPF, and BGP).
Note: The S-Series platform does not support all protocols. It is important to know which protocol you are supporting prior to implementing Prefix-Lists.

Configuration Task List for Prefix Lists

To configure a prefix list, you must use commands in the PREFIX LIST, the ROUTER RIP, ROUTER OSPF, and ROUTER BGP modes. Basically, you create the prefix list in the PREFIX LIST mode, and assign that list to commands in the ROUTER RIP, ROUTER OSPF and ROUTER BGP modes. The following list includes the configuration tasks for prefix lists: Configure a prefix list Use a prefix list for route redistribution

For a complete listing of all commands related to prefix lists, refer to the FTOS Command Line Interface Reference document.

Configure a prefix list

To configure a prefix list, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip prefix-list prefix-name

Command Mode CONFIGURATION

Purpose Create a prefix list and assign it a unique name. You are in the PREFIX LIST mode. Create a prefix list with a sequence number and a deny or permit action. The optional parameters are: ge min-prefix-length: is the minimum prefix length to be matched (0 to 32). le max-prefix-length: is the maximum prefix length to be matched (0 to 32).

seq sequence-number {deny | permit} ip-prefix [ge min-prefix-length] [le max-prefix-length]

CONFIG-NPREFIXL

Access Control Lists (ACL), Prefix Lists, and Route-maps | 105

www.dell.com | support.dell.com

If you want to forward all routes that do not match the prefix list criteria, you must configure a prefix list filter to permit all routes (permit 0.0.0.0/0 le 32). The permit all filter should be the last filter in your prefix list. To permit the default route only, enter permit 0.0.0.0/0. Figure 6-13 illustrates how the seq command orders the filters according to the sequence number assigned. In the example, filter 20 was configured before filter 15 and 12, but the show config command displays the filters in the correct order.
Figure 6-13. Command Example: seq
FTOS(conf-nprefixl)#seq 20 permit 0.0.0.0/0 le 32 FTOS(conf-nprefixl)#seq 12 deny 134.23.0.0 /16 FTOS(conf-nprefixl)#seq 15 deny 120.23.14.0 /8 le 16 FTOS(conf-nprefixl)#show config ! ip prefix-list juba seq 12 deny 134.23.0.0/16 seq 15 deny 120.0.0.0/8 le 16 seq 20 permit 0.0.0.0/0 le 32 FTOS(conf-nprefixl)#

Note the last line in the prefix list Juba contains a permit all statement. By including this line in a prefix list, you specify that all routes not matching any criteria in the prefix list are forwarded. To delete a filter, use the no seq sequence-number command in the PREFIX LIST mode. If you are creating a standard prefix list with only one or two filters, you can let FTOS assign a sequence number based on the order in which the filters are configured. The FTOS assigns filters in multiples of five. To configure a filter without a specified sequence number, use these commands in the following sequence starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip prefix-list prefix-name

Command Mode CONFIGURATION CONFIG-NPREFIXL

Purpose Create a prefix list and assign it a unique name. Create a prefix list filter with a deny or permit action. The optional parameters are: ge min-prefix-length: is the minimum prefix length to be matched (0 to 32). le max-prefix-length: is the maximum prefix length to be matched (0 to 32).

{deny | permit} ip-prefix [ge min-prefix-length] [le max-prefix-length]

Figure 6-14 illustrates a prefix list in which the sequence numbers were assigned by the software. The filters were assigned sequence numbers based on the order in which they were configured (for example, the first filter was given the lowest sequence number). The show config command in the PREFIX LIST mode displays the two filters with the sequence numbers 5 and 10.

106

Access Control Lists (ACL), Prefix Lists, and Route-maps

Figure 6-14. Prefix List

FTOS(conf-nprefixl)#permit 123.23.0.0 /16 FTOS(conf-nprefixl)#deny 133.24.56.0 /8 FTOS(conf-nprefixl)#show conf ! ip prefix-list awe seq 5 permit 123.23.0.0/16 seq 10 deny 133.0.0.0/8 FTOS(conf-nprefixl)#

To delete a filter, enter the show config command in the PREFIX LIST mode and locate the sequence number of the filter you want to delete; then use the no seq sequence-number command in the PREFIX LIST mode. To view all configured prefix lists, use either of the following commands in the EXEC mode:
Command Syntax
show ip prefix-list detail [prefix-name] show ip prefix-list summary

Command Mode EXEC Privilege EXEC Privilege

Purpose Show detailed information about configured Prefix lists. Show a table of summarized information about configured Prefix lists.

[prefix-name]

Figure 6-15. Command example: show ip prefix-list detail

FTOS>show ip prefix detail Prefix-list with the last deletion/insertion: filter_ospf ip prefix-list filter_in: count: 3, range entries: 3, sequences: 5 - 10 seq 5 deny 1.102.0.0/16 le 32 (hit count: 0) seq 6 deny 2.1.0.0/16 ge 23 (hit count: 0) seq 10 permit 0.0.0.0/0 le 32 (hit count: 0) ip prefix-list filter_ospf: count: 4, range entries: 1, sequences: 5 - 10 seq 5 deny 100.100.1.0/24 (hit count: 0) seq 6 deny 200.200.1.0/24 (hit count: 0) seq 7 deny 200.200.2.0/24 (hit count: 0) seq 10 permit 0.0.0.0/0 le 32 (hit count: 0) FTOS>

Figure 6-16. Command Example: show ip prefix-list summary

FTOS>show ip prefix summary Prefix-list with the last deletion/insertion: filter_ospf ip prefix-list filter_in: count: 3, range entries: 3, sequences: 5 - 10 ip prefix-list filter_ospf: count: 4, range entries: 1, sequences: 5 - 10 FTOS>

Access Control Lists (ACL), Prefix Lists, and Route-maps | 107

www.dell.com | support.dell.com

Use a prefix list for route redistribution

To pass traffic through a configured prefix list, you must use the prefix list in a route redistribution command. The prefix list is applied to all traffic redistributed into the routing process and the traffic is either forwarded or dropped depending on the criteria and actions specified in the prefix list. To apply a filter to routes in RIP (RIP is supported on C and E-Series.), use either of the following commands in the ROUTER RIP mode:
Command Syntax
router rip distribute-list prefix-list-name in

Command Mode CONFIGURATION CONFIG-ROUTER-RIP

Purpose Enter RIP mode Apply a configured prefix list to incoming routes. You can specify an interface. If you enter the name of a nonexistent prefix list, all routes are forwarded. Apply a configured prefix list to outgoing routes. You can specify an interface or type of route. If you enter the name of a non-existent prefix list, all routes are forwarded.

[interface]

distribute-list prefix-list-name out [interface | connected | static | ospf]

CONFIG-ROUTER-RIP

To view the configuration, use the show config command in the ROUTER RIP mode (Figure 240) or the show running-config rip command in the EXEC mode.
Figure 6-17. Command Example: show config in the ROUTER RIP Mode
FTOS(conf-router_rip)#show config ! router rip distribute-list prefix juba out network 10.0.0.0 FTOS(conf-router_rip)#router ospf 34

To apply a filter to routes in OSPF, use either of the following commands in the ROUTER OSPF mode:
Command Syntax
router ospf distribute-list prefix-list-name in [interface]

Command Mode CONFIGURATION CONFIG-ROUTER-OSPF

Purpose Enter OSPF mode Apply a configured prefix list to incoming routes. You can specify an interface. If you enter the name of a non-existent prefix list, all routes are forwarded. Apply a configured prefix list to incoming routes. You can specify which type of routes are affected. If you enter the name of a non-existent prefix list, all routes are forwarded.

distribute-list prefix-list-name out [connected | rip | static]

CONFIG-ROUTER-OSPF

108

Access Control Lists (ACL), Prefix Lists, and Route-maps

To view the configuration, use the show config command in the ROUTER OSPF mode (Figure 241) or the show running-config ospf command in the EXEC mode.
Figure 6-18. Command Example: show config in ROUTER OSPF Mode
FTOS(conf-router_ospf)#show config ! router ospf 34 network 10.2.1.1 255.255.255.255 area 0.0.0.1 distribute-list prefix awe in FTOS(conf-router_ospf)#

ACL Resequencing
ACL Resequencing allows you to re-number the rules and remarks in an access or prefix list. The placement of rules within the list is critical because packets are matched against rules in sequential order. Use Resequencing whenever there is no longer an opportunity to order new rules as desired using current numbering scheme. For example, Table 6-3 contains some rules that are numbered in increments of 1. No new rules can be placed between these, so apply resequencing to create numbering space, as shown in Table 6-4. In the same example, apply resequencing if more than two rules must be placed between rules 7 and 10. IPv4 and IPv6 ACLs and prefixes and MAC ACLs can be resequenced. No CAM writes happen as a result of resequencing, so there is no packet loss; the behavior is like Hot-lock ACLs.
Note: ACL Resequencing does not affect the rules or remarks or the order in which they are applied. It merely renumbers them so that new rules can be placed within the list as desired.

Table 6-3.

ACL Resequencing Example (Insert New Rules)

seq 5 permit any host 1.1.1.1 seq 6 permit any host 1.1.1.2 seq 7 permit any host 1.1.1.3 seq 10 permit any host 1.1.1.4

Table 6-4.

ACL Resequencing Example (Resequenced)

seq 5 permit any host 1.1.1.1 seq 10 permit any host 1.1.1.2 seq 15 permit any host 1.1.1.3 seq 20 permit any host 1.1.1.4

Access Control Lists (ACL), Prefix Lists, and Route-maps | 109

www.dell.com | support.dell.com

Resequencing an ACL or Prefix List

Resequencing is available for IPv4 and IPv6 ACLs and prefix lists and MAC ACLs. To resequence an ACL or prefix list use the appropriate command in Table 6-5. You must specify the list name, starting number, and increment when using these commands.
Table 6-5. List IPv4, IPv6, or MAC ACL IPv4 or IPv6 prefix-list Resequencing ACLs and Prefix Lists Command
resequence access-list {ipv4 | ipv6 | mac} {access-list-name StartingSeqNum Step-to-Increment}

Command Mode Exec

resequence prefix-list {ipv4 | ipv6} {prefix-list-name StartingSeqNum Exec Step-to-Increment}

Figure 6-19 shows the resequencing of an IPv4 access-list beginning with the number 2 and incrementing by 2.
Figure 6-19. Resequencing ACLs
FTOS(config-ext-nacl)# show config ! ip access-list extended test remark 4 XYZ remark 5 this remark corresponds to permit any host 1.1.1.1 seq 5 permit ip any host 1.1.1.1 remark 9 ABC remark 10 this remark corresponds to permit ip any host 1.1.1.2 seq 10 permit ip any host 1.1.1.2 seq 15 permit ip any host 1.1.1.3 seq 20 permit ip any host 1.1.1.4 FTOS# end FTOS# resequence access-list ipv4 test 2 2 FTOS# show running-config acl ! ip access-list extended test remark 2 XYZ remark 4 this remark corresponds to permit any host 1.1.1.1 seq 4 permit ip any host 1.1.1.1 remark 6 this remark has no corresponding rule remark 8 this remark corresponds to permit ip any host 1.1.1.2 seq 8 permit ip any host 1.1.1.2 seq 10 permit ip any host 1.1.1.3 seq 12 permit ip any host 1.1.1.4

Remarks and rules that originally have the same sequence number have the same sequence number after the resequence command is applied. Remarks that do not have a corresponding rule will be incremented as as a rule. These two mechanisms allow remarks to retain their original position in the list. For example, in Figure 6-20, remark 10 corresponds to rule 10 and as such they have the same number before and after the command is entered. Remark 4 is incremented as a rule, and all rules have retained their original positions.

110

Access Control Lists (ACL), Prefix Lists, and Route-maps

Figure 6-20. Resequencing Remarks

FTOS(config-ext-nacl)# show config ! ip access-list extended test remark 4 XYZ remark 5 this remark corresponds to permit any host 1.1.1.1 seq 5 permit ip any host 1.1.1.1 remark 9 ABC remark 10 this remark corresponds to permit ip any host 1.1.1.2 seq 10 permit ip any host 1.1.1.2 seq 15 permit ip any host 1.1.1.3 seq 20 permit ip any host 1.1.1.4 FTOS# end FTOS# resequence access-list ipv4 test 2 2 FTOS# show running-config acl ! ip access-list extended test remark 2 XYZ remark 4 this remark corresponds to permit any host 1.1.1.1 seq 4 permit ip any host 1.1.1.1 remark 6 this remark has no corresponding rule remark 8 this remark corresponds to permit ip any host 1.1.1.2 seq 8 permit ip any host 1.1.1.2 seq 10 permit ip any host 1.1.1.3 seq 12 permit ip any host 1.1.1.4

Route Maps
Route-maps are supported on platforms:

ces

Like ACLs and prefix lists, route maps are composed of a series of commands that contain a matching criterion and an action, yet route maps can change the packets meeting the criterion. ACLs and prefix lists can only drop or forward the packet or traffic. Route maps process routes for route redistribution. For example, a route map can be called to filter only specific routes and to add a metric. Route maps also have an implicit deny. Unlike ACLs and prefix lists, however, where the packet or traffic is dropped, in route maps, if a route does not match any of the route map conditions, the route is not redistributed.

Implementation Information
The FTOS implementation of route maps allows route maps with no match command or no set command. When there is no match command, all traffic matches the route map and the set command applies.

Important Points to Remember

For route-maps with more than one match clause:

Access Control Lists (ACL), Prefix Lists, and Route-maps | 111

www.dell.com | support.dell.com

Two or more match clauses within the same route-map sequence have the same match commands (though the values are different), matching a packet against these clauses is a logical OR operation. Two or more match clauses within the same route-map sequence have different match commands, matching a packet against these clauses is a logical AND operation. If no match is found in a route-map sequence, the process moves to the next route-map sequence until a match is found, or there are no more sequences. When a match is found, the packet is forwarded; no more route-map sequences are processed. If a continue clause is included in the route-map sequence, the next or a specified route-map sequence is processed after a match is found.

Configuration Task List for Route Maps

You configure route maps in the ROUTE-MAP mode and apply them in various commands in the ROUTER RIP and ROUTER OSPF modes. The following list includes the configuration tasks for route maps: Create a route map (mandatory) Configure route map filters (optional) Configure a route map for route redistribution (optional) Configure a route map for route tagging (optional)

Create a route map

Route maps, ACLs, and prefix lists are similar in composition because all three contain filters, but route map filters are do not contain the permit and deny actions found in ACLs and prefix lists. Route map filters match certain routes and set or specify values. To create a route map and enter the ROUTE-MAP mode, use the following command in the CONFIGURATION mode:
Command Syntax
route-map map-name [permit | deny]

Command Mode CONFIGURATION

Purpose Create a route map and assign it a unique name. The optional permit and deny keywords are the action of the route map. The default is permit. The optional parameter seq allows you to assign a sequence number to the route map instance.

[sequence-number]

The default action is permit and the default sequence number starts at 10. When the keyword deny is used in configuring a route map, routes that meet the match filters are not redistributed. To view the configuration, use the show config command in the ROUTE-MAP mode (Figure 244).

112

Access Control Lists (ACL), Prefix Lists, and Route-maps

Figure 6-21. Command Example: show config in the ROUTE-MAP Mode

FTOS(config-route-map)#show config ! route-map dilling permit 10 FTOS(config-route-map)#

You can create multiple instances of this route map by using the sequence number option to place the route maps in the correct order. FTOS processes the route maps with the lowest sequence number first. When a configured route map is applied to a command, like redistribute, traffic passes through all instances of that route map until a match is found. Figure 6-22 shows an example with two instances of a route map.
Figure 6-22. Command Example: show route-map with Multiple Instances of a Route Map
FTOS#show route-map route-map zakho, permit, sequence 10 Match clauses: Set clauses: route-map zakho, permit, sequence 20 Match clauses: interface GigabitEthernet 0/1 Set clauses: tag 35 level stub-area FTOS#

Route map zakho has two instances

To delete all instances of that route map, use the no route-map map-name command. To delete just one instance, add the sequence number to the command syntax (Figure 246).
Figure 6-23. Deleting One Instance of a Route Map
FTOS(conf)#no route-map zakho 10 FTOS(conf)#end FTOS#show route-map route-map zakho, permit, sequence 20 Match clauses: interface GigabitEthernet 0/1 Set clauses: tag 35 level stub-area FTOS#

Figure 6-24 shows an example of a route map with multiple instances. The show config command displays only the configuration of the current route map instance. To view all instances of a specific route map, use the show route-map command.

Access Control Lists (ACL), Prefix Lists, and Route-maps | 113

www.dell.com | support.dell.com

Figure 6-24. Command Example: show route-map

FTOS#show route-map dilling route-map dilling, permit, sequence 10 Match clauses: Set clauses: route-map dilling, permit, sequence 15 Match clauses: interface Loopback 23 Set clauses: tag 3444 FTOS#

To delete a route map, use the no route-map map-name command in the CONFIGURATION mode.

Configure route map filters

Within the ROUTE-MAP mode, there are match and set commands. Basically, match commands search for a certain criterion in the routes and the set commands change the characteristics of those routes, either adding something or specifying a level. When there are multiple match commands of the same parameter under one instance of route-map, then FTOS does a match between either of those match commands. If there are multiple match commands of different parameter, then FTOS does a match ONLY if there is a match among ALL match commands. The following example explains better:

Example 1
FTOS(conf)#route-map force permit 10 FTOS(config-route-map)#match tag 1000 FTOS(config-route-map)#match tag 2000 FTOS(config-route-map)#match tag 3000

In the above route-map, if a route has any of the tag value specified in the match commands, then there is a match.

Example 2
FTOS(conf)#route-map force permit 10 FTOS(config-route-map)#match tag 1000 FTOS(config-route-map)#match metric 2000

In the above route-map, only if a route has both the characteristics mentioned in the route-map, it is matched. Explaining further, the route must have a tag value of 1000 and a metric value of 2000. Only then is there a match.

114

Access Control Lists (ACL), Prefix Lists, and Route-maps

Also, if there are different instances of the same route-map, then its sufficient if a permit match happens in any instance of that route-map. As an example:
FTOS(conf)#route-map force permit 10 FTOS(config-route-map)#match tag 1000 FTOS(conf)#route-map force deny 20 FTOS(config-route-map)#match tag 1000 FTOS(conf)#route-map force deny 30 FTOS(config-route-map)#match tag 1000

In the above route-map, instance 10 permits the route having a tag value of 1000 and instances 20 & 30 denies the route having a tag value of 1000. In the above scenario, FTOS scans all the instances of the route-map for any permit statement. If there is a match anywhere, the route is permitted, though other instances of the route-map denies it. To configure match criterion for a route map, use any or all of the following commands in the ROUTE-MAP mode:
Command Syntax
match as-path as-path-name match community
community-list-name [exact]

Command Mode CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

Purpose Match routes with the same AS-PATH numbers. Match routes with COMMUNITY list attributes in their path. Match routes whose next hop is a specific interface. The parameters are: For a Fast Ethernet interface, enter the keyword FastEthernet followed by the slot/ port information. For a 1-Gigabit Ethernet interface, enter the keyword gigabitEthernet followed by the slot/port information. For a loopback interface, enter the keyword loopback followed by a number between zero (0) and 16383. For a port channel interface, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword tengigabitEthernet followed by the slot/port information. For a VLAN, enter the keyword vlan followed by a number from 1 to 4094. E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

match interface interface

Access Control Lists (ACL), Prefix Lists, and Route-maps | 115

www.dell.com | support.dell.com

Command Syntax
match ip address prefix-list-name match ipv6 address prefix-list-name match ip next-hop {access-list-name | prefix-list prefix-list-name} match ipv6 next-hop {access-list-name | prefix-list prefix-list-name} match ip route-source {access-list-name | prefix-list prefix-list-name} match ipv6 route-source {access-list-name | prefix-list prefix-list-name} match metric metric-value match origin {egp | igp | incomplete} match route-type {external [type-1 | type-2] | internal | level-1 | level-2 | local } match tag tag-value

Command Mode CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

Purpose Match destination routes specified in a prefix list (IPv4). Match destination routes specified in a prefix list (IPv6). Match next-hop routes specified in a prefix list (IPv4). Match next-hop routes specified in a prefix list (IPv6). Match source routes specified in a prefix list (IPv4). Match source routes specified in a prefix list (IPv6). Match routes with a specific value. Match BGP routes based on the ORIGIN attribute. Match routes specified as internal or external to OSPF, ISIS level-1, ISIS level-2, or locally generated. Match routes with a specific tag.

CONFIG-ROUTE-MAP

CONFIG-ROUTE-MAP

CONFIG-ROUTE-MAP

CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

CONFIG-ROUTE-MAP

To configure a set condition, use any or all of the following commands in the ROUTE-MAP mode:
Command Syntax
set as-path prepend as-number [... as-number] set automatic-tag set level {backbone | level-1 | level-1-2 | level-2 | stub-area } set local-preference value set metric {+ | - | metric-value} set metric-type {external | internal | type-1 | type-2} set next-hop ip-address

Command Mode CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

Purpose Add an AS-PATH number to the beginning of the AS-PATH Generate a tag to be added to redistributed routes. Specify an OSPF area or ISIS level for redistributed routes. Specify a value for the BGP routes LOCAL_PREF attribute. Specify a value for redistributed routes. Specify an OSPF or ISIS type for redistributed routes. Assign an IP address as the routes next hop.

116

Access Control Lists (ACL), Prefix Lists, and Route-maps

Command Syntax
set ipv6 next-hop ip-address set origin {egp | igp | incomplete} set tag tag-value set weight value

Command Mode CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

Purpose Assign an IPv6 address as the routes next hop. Assign an ORIGIN attribute. Specify a tag for the redistributed routes. Specify a value as the routes weight.

Use these commands to create route map instances. There is no limit to the number of set and match commands per route map, but the convention is to keep the number of match and set filters in a route map low. Set commands do not require a corresponding match command.

Configure a route map for route redistribution

Route maps on their own cannot affect traffic and must be included in different commands to affect routing traffic. To apply a route map to traffic on the E-Series, you must call or include that route map in a command such as the redistribute or default-information originate commands in OSPF, ISIS, and BGP. Route redistribution occurs when FTOS learns the advertising routes from static or directly connected routes or another routing protocol. Different protocols assign different values to redistributed routes to identify either the routes and their origins. The metric value is the most common attribute that is changed to properly redistribute other routes into a routing protocol. Other attributes that can be changed include the metric type (for example, external and internal route types in OSPF) and route tag. Use the redistribute command in OSPF, RIP, ISIS, and BGP to set some of these attributes for routes that are redistributed into those protocols. Route maps add to that redistribution capability by allowing you to match specific routes and set or change more attributes when redistributing those routes. In Figure 6-25, the redistribute command calls the route map static ospf to redistribute only certain static routes into OSPF. According to the route map static ospf, only routes that have a next hop of Gigabitethernet interface 0/0 and that have a metric of 255 will be redistributed into the OSPF backbone area.
Note: When re-distributing routes using route-maps, the user must take care to create the route-map defined in the redistribute command under the routing protocol. If no route-map is created, then NO routes are redistributed. Figure 6-25. Route Redistribution into OSPF
router ospf 34 default-information originate metric-type 1 redistribute static metric 20 metric-type 2 tag 0 route-map staticospf ! route-map staticospf permit 10 match interface GigabitEthernet 0/0 match metric 255 set level backbone

Access Control Lists (ACL), Prefix Lists, and Route-maps | 117

www.dell.com | support.dell.com

Configure a route map for route tagging

One method for identifying routes from different routing protocols is to assign a tag to routes from that protocol. As the route enters a different routing domain, it is tagged and that tag is passed along with the route as it passes through different routing protocols. This tag can then be used when the route leaves a routing domain to redistribute those routes again. In Figure 6-26, the redistribute ospf command with a route map is used in the ROUTER RIP mode to apply a tag of 34 to all internal OSPF routes that are redistributed into RIP.
Figure 6-26. Tagging OSPF Routes Entering a RIP Routing Domain
! router rip redistribute ospf 34 metric 1 route-map torip ! route-map torip permit 10 match route-type internal set tag 34 !

Continue clause
Normally, when a match is found, set clauses are executed, and the packet is then forwarded; no more route-map modules are processed. If the continue command is configured at the end of a module, the next module (or a specified module) is processed even after a match is found. Figure 6-27 shows a continue clause at the end of a route-map module. In this example, if a match is found in the route-map test module 10, module 30 will be processed.
Note: If the continue clause is configured without specifying a module, the next sequential module is processed.

Figure 6-27. Command Example: continue

! route-map test permit 10 match commu comm-list1 set community 1:1 1:2 1:3 set as-path prepend 1 2 3 4 5 continue 30!

118

Access Control Lists (ACL), Prefix Lists, and Route-maps

7
Border Gateway Protocol IPv4 (BGPv4)
Border Gateway Protocol IPv4 (BGPv4) version 4 (BGPv4) is supported on platforms: Platforms support BGP according to the following table:
FTOS version 8.1.1.0 7.8.1.0 7.7.1.0. pre-7.7.1.0 Platform support E-Series ExaScale S-Series C-Series E-Series TeraScale

ces

ex s c et

This chapter is intended to provide a general description of Border Gateway Protocol version 4 (BGPv4) as it is supported in the Dell Networking operating system (FTOS). This chapter includes the following topics: Protocol Overview Autonomous Systems (AS) Sessions and Peers Route Reflectors Confederations BGP Attributes Best Path Selection Criteria Weight Local Preference Multi-Exit Discriminators (MEDs) AS Path Next Hop

Border Gateway Protocol IPv4 (BGPv4) | 119

www.dell.com | support.dell.com

Multiprotocol BGP Implementing BGP with FTOS Advertise IGP cost as MED for redistributed routes Ignore Router-ID for some best-path calculations 4-Byte AS Numbers AS4 Number Representation AS Number Migration BGP4 Management Information Base (MIB) Important Points to Remember Configuration Information Configuration Task List for BGP MBGP Configuration Storing Last and Bad PDUs Capturing PDUs PDU Counters Sample Configurations

BGP protocol standards are listed in the Appendix , Standards Compliance chapter.

Protocol Overview
Border Gateway Protocol (BGP) is an external gateway protocol that transmits interdomain routing information within and between Autonomous Systems (AS). Its primary function is to exchange network reachability information with other BGP systems. BGP generally operates with an Internal Gateway Protocol (IGP) such as OSPF or RIP, allowing you to communicate to external ASs smoothly. BGP adds reliability to network connections having multiple paths from one router to another.

Autonomous Systems (AS)

BGP Autonomous Systems (ASs) are a collection of nodes under common administration, with common network routing policies. Each AS has a number, already assigned by an internet authority. You do not assign the BGP number. AS Numbers (ASNs) are important because the ASN uniquely identifies each network on the Internet. The IANA has reserved AS numbers 64512 through 65534 to be used for private purposes. The ASNs 0 and 65535 are reserved by the IANA and should not be used in a live environment. Autonomous Systems can be grouped into three categories, defined by their connections and operation. A multihomed AS is one that maintains connections to more than one other AS. This allows the AS to remain connected to the internet in the event of a complete failure of one of their connections. However, this type of AS does not allow traffic from one AS to pass through on its way to another AS. A simple example of this is seen in Figure 7-1.

120

Border Gateway Protocol IPv4 (BGPv4)

A stub AS is one that is connected to only one other AS. A transit AS is one that provides connections through itself to separate networks. For example as seen in Figure 7-1, Router 1 can use Router 2 (the transit AS) to connect to Router 4. ISPs are always transit ASs, because they provide connections from one network to another. The ISP is considered to be selling transit service to the customer network, so thus the term Transit AS.

When BGP operates inside an Autonomous System (AS1 or AS2 as seen in Figure 7-1), it is referred to as Internal BGP (IBGP Interior Border Gateway Protocol). When BGP operates between Autonomous Systems (AS1 and AS2), it is called External BGP (EBGP Exterior Border Gateway Protocol). IBGP provides routers inside the AS with the knowledge to reach routers external to
the AS. EBGP routers exchange information with other EBGP routers as well as IBGP routers to maintain connectivity and accessibility.
Figure 7-1.
lpbgp1111

BGP Autonomous Zones

Router 3

Router 5

Router 1

Router 2

Router 4

Router 6

Exterior BGP (EBGP)

Router 7

AS 1
Interior BGP (IBGP)

AS 2
Interior BGP (IBGP)

BGP version 4 (BGPv4) supports classless interdomain routing and aggregate routes and AS paths. BGP is a path vector protocol - a computer network in which BGP maintains the path that update

information takes as it diffuses through the network. Updates traveling through the network and returning to the same node are easily detected and discarded.
BGP does not use traditional Interior Gateway Protocol (IGP) matrix, but makes routing decisions based on path, network policies and/or rulesets. Unlike most protocols, BGP uses TCP as its transport protocol. Since each BGP routers talking to another router is a session, a BGP network needs to be in full mesh. This is a topology that has every router directly connected to every other router. For example, as seen in Figure 7-2, four routers connected in a full mesh have three peers each, six routers have 5 peers each, and eight routers in full mesh will have seven peers each.

Border Gateway Protocol IPv4 (BGPv4) | 121

www.dell.com | support.dell.com

Figure 7-2.

Full Mesh Examples

4 Routers 6 Routers

8 Routers
The number of BGP speakers each BGP peer must maintain increases exponentially. Network management quickly becomes impossible.

Sessions and Peers

When two routers communicate using the BGP protocol, a BGP session is started. The two end-points of that session are Peers. A Peer is also called a Neighbor.

Establishing a session
Information exchange between peers is driven by events and timers. The focus in BGP is on the traffic routing policies.

122

Border Gateway Protocol IPv4 (BGPv4)

In order to make decisions in its operations with other BGP peers, a BGP peer uses a simple finite state machine that consists of six states: Idle, Connect, Active, OpenSent, OpenConfirm, and Established. For each peer-to-peer session, a BGP implementation tracks which of these six states the session is in. The BGP protocol defines the messages that each peer should exchange in order to change the session from one state to another. The first state is the Idle mode. BGP initializes all resources, refuses all inbound BGP connection attempts, and initiates a TCP connection to the peer. The next state is Connect. In this state the router waits for the TCP connection to complete, transitioning to the OpenSent state if successful. If that transition is not successful, BGP resets the ConnectRetry timer and transitions to the Active state when the timer expires. In the Active state, the router resets the ConnectRetry timer to zero, and returns to the Connect state. Upon successful OpenSent transition, the router sends an Open message and waits for one in return. Keepalive messages are exchanged next, and upon successful receipt, the router is placed in the Established state. Keepalive messages continue to be sent at regular periods (established by the Keepalive timer) to verify connections. Once established, the router can now send/receive Keepalive, Update, and Notification messages to/from its peer.

Peer Groups
Peer Groups are neighbors grouped according to common routing policies. They enable easier system configuration and management by allowing groups of routers to share and inherit policies. Peer groups also aid in convergence speed. When a BGP process needs to send the same information to a large number of peers, it needs to set up a long output queue to get that information to all the proper peers. If they are members of a peer group, however, the information can be sent to one place then passed onto the peers within the group.

Route Reflectors
Route Reflectors reorganize the iBGP core into a hierarchy and allows some route advertisement rules. Route reflection divides iBGP peers into two groups: client peers and nonclient peers. A route reflector and its client peers form a route reflection cluster. Since BGP speakers announce only the best route for a given prefix, route reflector rules are applied after the router makes its best path decision. If a route was received from a nonclient peer, reflect the route to all client peers. If the route was received from a client peer, reflect the route to all nonclient and all client peers.

Border Gateway Protocol IPv4 (BGPv4) | 123

www.dell.com | support.dell.com

To illustrate how these rules affect routing, see Figure 7-3 and the following steps. Routers B, C, D, E, and G are members of the same AS - AS100. These routers are also in the same Route Reflection Cluster, where Router D is the Route Reflector. Router E and H are client peers of Router D; Routers B and C and nonclient peers of Router D.
Figure 7-3.
Router A

Route Reflection Example

Router B iBGP Routes

eBGP Route

Router E

eBGP Route Router F

iBGP Route

Route Reflector Router D

Route Reflector Client Peers

Router C iBGP Routes

Router G

Router H

eBGP Route

1. Router B receives an advertisement from Router A through eBGP. Since the route is learned through eBGP, Router B advertises it to all its iBGP peers: Routers C and D. 2. Router C receives the advertisement but does not advertise it to any peer because its only other peer is Router D, an iBGP peer, and Router D has already learned it through iBGP from Router B. 3. Router D does not advertise the route to Router C because Router C is a nonclient peer and the route advertisement came from Router B who is also a non-client peer. 4. Router D does reflect the advertisement to Routers E and G because they are client peers of Router D. 5. Routers E and G then advertise this iBGP learned route to their eBGP peers Routers F and H.

Confederations
Communities
BGP communities are sets of routes with one or more common attributes. This is a way to assign common attributes to multiple routes at the same time.

124

Border Gateway Protocol IPv4 (BGPv4)

BGP Attributes
Routes learned via BGP have associated properties that are used to determine the best route to a destination when multiple paths exist to a particular destination. These properties are referred to as BGP attributes, and an understanding of how BGP attributes influence route selection is required for the design of robust networks. This section describes the attributes that BGP uses in the route selection process: Weight Local Preference Multi-Exit Discriminators (MEDs) Origin AS Path Next Hop

Best Path Selection Criteria

Paths for active routes are grouped in ascending order according to their neighboring external AS number (BGP best path selection is deterministic by default, which means the bgp non-deterministic-med command is NOT applied). The best path in each group is selected based on specific criteria. Only one best path is selected at a time. If any of the criteria results in more than one path, BGP moves on to the next option in the list. For example, two paths may have the same weights, but different local preferences. BGP sees that the Weight criteria results in two potential best paths and moves to local preference to reduce the options. If a number of best paths is determined, this selection criteria is applied to groups best to determine the ultimate best path. In non-deterministic mode (the bgp non-deterministic-med command is applied), paths are compared in the order in which they arrive. This method can lead to FTOS choosing different best paths from a set of paths, depending on the order in which they were received from the neighbors, since MED may or may not get compared between adjacent paths. In deterministic mode, FTOS compares MED between adjacent paths within an AS group since all paths in the AS group are from the same AS. Figure 7-4 illustrates the decisions BGP goes through to select the best path. The list following the illustration details the path selection criteria.

Border Gateway Protocol IPv4 (BGPv4) | 125

www.dell.com | support.dell.com

Figure 7-4.

BGP Best Path Selection

No, or Not Resulting in a Single Route

Largest Weight

Highest Local Pref

Locally Originated Path

Shortest AS Path

Lowest Origin Code

Lowest MED

Learned via EBGP

Lowest NEXT-HOP Cost

Tie Breakers
Short Cluster List from Lowest BGP ID

Lowest Peering Addr

A Single Route is Selected and Installed in the Forwarding Table

Best Path selection details

1. Prefer the path with the largest WEIGHT attribute. 2. Prefer the path with the largest LOCAL_PREF attribute. 3. Prefer the path that was locally Originated via a network command, redistribute command or aggregate-address command. Routes originated with the network or redistribute commands are preferred over routes originated with the aggregate-address command.

4. Prefer the path with the shortest AS_PATH (unless the bgp bestpath as-path ignore command is configured, then AS_PATH is not considered). The following criteria apply: An AS_SET has a path length of 1, no matter how many ASs are in the set. A path with no AS_PATH configured has a path length of 0. AS_CONFED_SET is not included in the AS_PATH length.

126

Border Gateway Protocol IPv4 (BGPv4)

AS_CONFED_SEQUENCE has a path length of 1, no matter how many ASs are in the AS_CONFED_SEQUENCE.

5. Prefer the path with the lowest ORIGIN type (IGP is lower than EGP, and EGP is lower than INCOMPLETE). 6. Prefer the path with the lowest Multi-Exit Discriminator (MED) attribute. The following criteria apply: This comparison is only done if the first (neighboring) AS is the same in the two paths; the MEDs are compared only if the first AS in the AS_SEQUENCE is the same for both paths. If the bgp always-compare-med command is entered, MEDs are compared for all paths. Paths with no MED are treated as worst and assigned a MED of 4294967295.

7. Prefer external (EBGP) to internal (IBGP) paths or confederation EBGP paths. 8. Prefer the path with the lowest IGP metric to the BGP next-hop. 9. FTOS deems the paths as equal and does not perform steps 9 through 11 listed below, if the following criteria is met: the IBGP multipath or EBGP multipath are configured (maximum-path command) the paths being compared were received from the same AS with the same number of ASs in the AS Path but with different NextHops the paths were received from IBGP or EBGP neighbor respectively

10. If the bgp bestpath router-id ignore command is enabled and: If the Router-ID is the same for multiple paths (because the routes were received from the same route) skip this step. If the Router-ID is NOT the same for multiple paths, Prefer the path that was first received as the Best Path. The path selection algorithm should return without performing any of the checks outlined below.

11. Prefer the path originated from the BGP router with the lowest router ID. For paths containing a Route Reflector (RR) attribute, the originator ID is substituted for the router ID. 12. If two paths have the same router ID, prefer the path with the lowest cluster ID length. Paths without a cluster ID length are set to a 0 cluster ID length. 13. Prefer the path originated from the neighbor with the lowest address. (The neighbor address is used in the BGP neighbor configuration, and corresponds to the remote peer used in the TCP connection with the local router.) After a number of best paths is determined, this selection criteria is applied to groups best to determine the ultimate best path. In non-deterministic mode (the bgp non-deterministic-med command is applied), paths are compared in the order in which they arrive. This method can lead to FTOS choosing different best paths from a set of paths, depending on the order in which they were received from the neighbors since MED may or may not get compared between adjacent paths. In deterministic mode, FTOS compares MED between adjacent paths within an AS group since all paths in the AS group are from the same AS.

Border Gateway Protocol IPv4 (BGPv4) | 127

www.dell.com | support.dell.com

Weight
The Weight attribute is local to the router and is not advertised to neighboring routers. If the router learns about more than one route to the same destination, the route with the highest weight will be preferred. The route with the highest weight is installed in the IP routing table.

Local Preference
Local Preference (LOCAL_PREF) represents the degree of preference within the entire AS. The higher the number, the greater the preference for the route. The Local Preference (LOCAL_PREF) is one of the criteria used to determine the best path, so keep in mind that other criteria may impact selection, as shown in Figure 7-4. For this example, assume that LOCAL_PREF is the only attribute applied. In Figure 7-5, AS100 has two possible paths to AS 200. Although the path through the Router A is shorter (one hop instead of two) the LOCAL_PREF settings have the preferred path go through Router B and AS300. This is advertised to all routers within AS100 causing all BGP speakers to prefer the path through Router B.
Figure 7-5. LOCAL_PREF Example

Set Local Preference to 100

AS 100

Router A T1 Link Router C

Router B

AS 200
Router E

Set Local Preference to 200 Router E

OC3 Link Router D

AS 300

Router F

Multi-Exit Discriminators (MEDs)

If two Autonomous Systems (AS) connect in more than one place, a Multi-Exit Discriminator (MED) can be used to assign a preference to a preferred path. The MED is one of the criteria used to determine the best path, so keep in mind that other criteria may impact selection, as shown in Figure 7-4.

128

Border Gateway Protocol IPv4 (BGPv4)

One AS assigns the MED a value and the other AS uses that value to decide the preferred path. For this example, assume the MED is the only attribute applied. In Figure 7-6, AS100 and AS200 connect in two places. Each connection is a BGP session. AS200 sets the MED for its T1 exit point to 100 and the MED for its OC3 exit point to 50. This sets up a path preference through the OC3 link. The MEDs are advertised to AS100 routers so they know which is the preferred path. An MED is a non-transitive attribute. If AS100 sends an MED to AS200, AS200 does not pass it on to AS300 or AS400. The MED is a locally relevant attribute to the two participating Autonomous Systems (AS100 and AS200). Note that the MEDs are advertised across both links, so that if a link goes down AS 1 still has connectivity to AS300 and AS400.
Figure 7-6. MED Route Example

AS 100

Router A

Set MED to 100 T1 Link

Router C

Router B

AS 200
Router E

OC3 Link

Router D Set MED to 50

Note: With FTOS Release 8.3.1.0, configuring the set metric-type internal command in a route-map advertises the IGP cost as MED to outbound EBGP peers when redistributing routes. The configured set metric value overwrites the default IGP cost.

Origin
The Origin indicates the origin of the prefix, or how the prefix came into BGP. There are three Origin codes: IGP, EGP, INCOMPLETE. IGP indicated the prefix originated from information learned through an interior gateway protocol. EGP indicated the prefix originated from information learned from an EGP protocol, which NGP replaced. INCOMPLETE indicates that the prefix originated from an unknown source.

Border Gateway Protocol IPv4 (BGPv4) | 129

www.dell.com | support.dell.com

Generally, an IGP indicator means that the route was derived inside the originating AS. EGP generally means that a route was learned from an external gateway protocol. An INCOMPLETE origin code generally results from aggregation, redistribution or other indirect ways of installing routes into BGP. In FTOS, these origin codes appear as shown in Figure 7-7. The question mark (?) indicates an Origin code of INCOMPLETE. The lower case letter (i) indicates an Origin code of IGP.
Figure 7-7. Origin attribute reported

FTOS#show ip bgp BGP table version is 0, local router ID is 10.101.15.13 Status codes: s suppressed, d damped, h history, * valid, > best Path source: I - internal, a - aggregate, c - confed-external, r - redistributed, n - network Origin codes: i - IGP, e - EGP, ? - incomplete Network 7.0.0.0/29 7.0.0.0/30 9.2.0.0/16 Next Hop 10.114.8.33 10.114.8.33 10.114.8.33 Metric 0 0 10 LocPrf 0 0 0 Weight 18508 18508 18508 Path ? ? 701 i

*> *> *>

AS Path
The AS Path is the list of all Autonomous Systems that all the prefixes listed in the update have passed through. The local AS number is added by the BGP speaker when advertising to a eBGP neighbor. In FTOS the AS Path is shown in Figure 7-8. Note that the Origin attribute is shown following the AS Path information.
Figure 7-8. AS Path attribute reported

FTOS#show ip bgp paths Total 30655 Paths Address Hash Refcount 0x4014154 0 3 0x4013914 0 3 0x5166d6c 0 3 0x5e62df4 0 2 0x3a1814c 0 26 0x567ea9c 0 75 0x6cc1294 0 2 0x6cc18d4 0 1 0x5982e44 0 162 0x67d4a14 0 2 0x559972c 0 31 0x59cd3b4 0 2 0x7128114 0 10 0x536a914 0 3 0x2ffe884 0 1

Metric 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508 18508

Path 701 3549 19421 i 701 7018 14990 i 209 4637 1221 9249 9249 i 701 17302 i 209 22291 i 209 3356 2529 i 209 1239 19265 i 701 2914 4713 17935 i 209 i 701 19878 ? 209 18756 i 209 7018 15227 i 209 3356 13845 i 209 701 6347 7781 i 701 3561 9116 21350 i

130

Border Gateway Protocol IPv4 (BGPv4)

Next Hop
The Next Hop is the IP address used to reach the advertising router. For EBGP neighbors, the Next-Hop address is the IP address of the connection between the neighbors. For IBGP, the EBGP Next-Hop address is carried into the local AS. A Next Hop attribute is set when a BGP speaker advertises itself to another BGP speaker outside its local AS. It can also be set when advertising routes within an AS. The Next Hop attribute also serves as a way to direct traffic to another BGP speaker, rather than waiting for a speaker to advertise. FTOS allows you to set the Next Hop attribute in the CLI. Setting the Next Hop attribute lets you determine a router as the next hop for a BGP neighbor.

Multiprotocol BGP

ec MBGP for IPv4 Multicast is supported on platform c e s

MBGP for IPv6 unicast is supported on platforms Multiprotocol Extensions for BGP (MBGP) is defined in IETF RFC 2858. MBGP allows different types of address families to be distributed in parallel. This allows information about the topology of IP Multicast-capable routers to be exchanged separately from the topology of normal IPv4 and IPv6 unicast routers. It allows a multicast routing topology different from the unicast routing topology.
Note: It is possible to configure BGP peers that exchange both unicast and multicast network layer reachability information (NLRI), but you cannot connect Multiprotocol BGP with BGP. Therefore, You cannot redistribute Multiprotocol BGP routes into BGP.

Implementing BGP with FTOS

Advertise IGP cost as MED for redistributed routes
When using multipath connectivity to an external AS, you can advertise the MED value selectively to each peer for redistributed routes. For some peers you can set the internal/IGP cost as the MED while setting others to a constant pre-defined metric as MED value. FTOS 8.3.1.0 and later support configuring the set metric-type internal command in a route-map to advertise the IGP cost as the MED to outbound EBGP peers when redistributing routes. The configured set metric value overwrites the default IGP cost. By using the redistribute command in conjunction with the route-map command, you can specify whether a peer advertises the standard MED or uses the IGP cost as the MED. Note the following when configuring this functionality:

Border Gateway Protocol IPv4 (BGPv4) | 131

www.dell.com | support.dell.com

If the redistribute command does not have any metric configured and BGP Peer out-bound route-map does have metric-type internal configured, BGP advertises the IGP cost as MED. If the redistribute command has metric configured (route-map set metric or redistribute route-type metric ) and the BGP Peer out-bound route-map has metric-type internal configured, BGP advertises the metric configured in the redistribute command as MED. If BGP peer out-bound route-map has metric configured, then all other metrics are overwritten by this.
Note: When redistributing static, connected or OSPF routes, there is no metric option. Simply assign the appropriate route-map to the redistributed route.

Table 7-1gives some examples of these rules.

Table 7-1. Example MED advertisement BGP Local Routing Information Base

Command Settings

MED Advertised to Peer WITH route-map WITHOUT route-map metric-type internal metric-type internal

redistribute isis (IGP cost = 20) redistribute isis route-map set metric 50 redistribute isis metric 100

MED: IGP cost 20 MED: IGP cost 50 MED: IGP cost 100

MED = 20 MED: 50 MED: 100

MED = 0 MED: 50 MED: 100

Ignore Router-ID for some best-path calculations

FTOS 8.3.1.0 and later allow you to avoid unnecessary BGP best-path transitions between external paths under certain conditions. The bgp bestpath router-id ignore command reduces network disruption caused by routing and forwarding plane changes and allows for faster convergence.

4-Byte AS Numbers
FTOS Version 7.7.1 and later support 4-Byte (32-bit) format when configuring Autonomous System Numbers (ASNs). The 4-Byte support is advertised as a new BGP capability (4-BYTE-AS) in the OPEN message. If a 4-Byte BGP speaker has sent and received this capability from another speaker, all the messages will be 4-octet. The behavior of a 4-Byte BGP speaker will be different with the peer depending on whether the peer is 4-Byte or 2-Byte BGP speaker.

132

Border Gateway Protocol IPv4 (BGPv4)

Where the 2-Byte format is 1-65535, the 4-Byte format is 1-4294967295. Enter AS Numbers using the traditional format. If the ASN is greater than 65535, the dot format is shown when using the show ip bgp commands. For example, an ASN entered as 3183856184 will appear in the show commands as 48581.51768; an ASN of 65123 is shown as 65123. To calculate the comparable dot format for an ASN from a traditional format, use ASN/65536. ASN%65536.
Table 7-2. 4-Byte ASN Dot Format Examples Dot Format Is The Same As 0.65501 1.0 1.34464 65535.65535

Traditional Format 65001 65536 100000 4294967295

When creating Confederations, all the routers in a Confederation must be either 4-Byte or 2-Byte identified routers. You cannot mix them. Configure the 4-byte AS numbers with the four-octet-support command.

AS4 Number Representation

FTOS version 8.2.1.0 supports multiple representations of an 4-byte AS Numbers: asplain, asdot+, and asdot.
Note: The ASDOT and ASDOT+ representations are supported only in conjunction with the 4-Byte AS Numbers feature. If 4-Byte AS Numbers are not implemented, only ASPLAIN representation is supported.

ASPLAIN is the method FTOS has used for all previous FTOS versions.It remains the default method with FTOS 8.2.1.0 and later. With the ASPLAIN notation, a 32 bit binary AS number is translated into a decimal value. All AS Numbers between 0-65535 are represented as a decimal number when entered in the CLI as well as when displayed in the show command outputs. AS Numbers larger than 65535 are represented using ASPLAIN notation as well. 65546 is

represented as 65546.
ASDOT+ representation splits the full binary 4-byte AS number into two words of 16 bits separated by a decimal point (.): <high-order 16 bit value>.<low-order 16 bit value>. Some examples are shown in Table 7-2. All AS Numbers between 0-65535 are represented as a decimal number, when entered in the CLI as well as when displayed in the show command outputs. AS Numbers larger than 65535 is represented using ASDOT notation as <higher 2 bytes in decimal>.<lower 2 bytes in decimal>. For example: AS 65546 is represented as 1.10.

Border Gateway Protocol IPv4 (BGPv4) | 133

www.dell.com | support.dell.com

ASDOT representation combines the ASPLAIN and ASDOT+ representations. AS Numbers less than 65536 appear in integer format (asplain); AS Numbers equal to or greater than 65536 appear using the decimal method (asdot+). For example, the AS Number 65526 appears as 65526, and the AS Number 65546 appears as 1.10.

Dynamic AS Number Notation application

FTOS 8.3.1.0 applies the ASN Notation type change dynamically to the running-config statements. When you apply or change an asnotation, the type selected is reflected immediately in the running-configuration and the show commands (Figure 7-9 and Figure 7-10).
Figure 7-9. ASDOT
FTOS(conf-router_bgp)#bgp asnotation asdot FTOS(conf-router_bgp)#show conf ! router bgp 100 bgp asnotation asdot bgp four-octet-as-support neighbor 172.30.1.250 local-as 65057 <output truncated> FTOS(conf-router_bgp)#do show ip bgp BGP table version is 24901, local router ID is 172.30.1.57 <output truncated>

Dynamic changes of the bgp asnotation command in the show running config

ASDOT+
FTOS(conf-router_bgp)#bgp asnotation asdot+ FTOS(conf-router_bgp)#show conf ! router bgp 100 bgp asnotation asdot+ bgp four-octet-as-support neighbor 172.30.1.250 local-as 65057 <output truncated> FTOS(conf-router_bgp)#do show ip bgp BGP table version is 31571, local router ID is 172.30.1.57 <output truncated>

AS-PLAIN
FTOS(conf-router_bgp)#bgp asnotation asplain FTOS(conf-router_bgp)#sho conf ! router bgp 100 bgp four-octet-as-support neighbor 172.30.1.250 local-as 65057 <output truncated> FTOS(conf-router_bgp)#do sho ip bgp BGP table version is 34558, local router ID is 172.30.1.57 <output truncated>

134

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-10. Dynamic changes when bgp asnotation command is disabled in the show running config AS NOTATION DISABLED
FTOS(conf-router_bgp)#no bgp asnotation FTOS(conf-router_bgp)#sho conf ! router bgp 100 bgp four-octet-as-support neighbor 172.30.1.250 local-as 65057 <output truncated> FTOS(conf-router_bgp)#do sho ip bgp BGP table version is 28093, local router ID is 172.30.1.57

AS4 SUPPORT DISABLED

FTOS(conf-router_bgp)#no bgp four-octet-as-support FTOS(conf-router_bgp)#sho conf ! router bgp 100 neighbor 172.30.1.250 local-as 65057 FTOS(conf-router_bgp)#do show ip bgp BGP table version is 28093, local router ID is 172.30.1.57

AS Number Migration
When migrating one AS to another, perhaps combining ASs, an eBGP network may lose its routing to an iBGP if the ASN changes. Migration can be difficult as all the iBGP and eBGP peers of the migrating network need to be updated to maintain network reachability. With this feature you can transparently change the AS number of entire BGP network and ensure that the routes are propagated throughout the network while the migration is in progress. Essentially, Local-AS provides a capability to the BGP speaker to operate as if it belongs to "virtual" AS network besides its physical AS network. Figure 7-11 shows a scenario where Router A, Router B and Router C belong to AS 100, 200, 300 respectively. Router A acquired Router B; Router B has Router C as its customer. When Router B is migrating to Router A, it needs to maintain the connection with Router C without immediately updating Router C's configuration. Local-AS allows this to happen by allowing Router B to appear as if it still belongs to Router B's old network (AS 200) as far as communicating with Router C is concerned.

Border Gateway Protocol IPv4 (BGPv4) | 135

www.dell.com | support.dell.com

Figure 7-11.

Local-AS Scenario

Router A AS 100

Router C AS 300

Router B AS 200

Before Migration

Router A

AS 100 AS 100

Router C AS 300

Router B Local AS 200

After Migration, with Local-AS enabled

When you complete your migration, and you have reconfigured your network with the new information you must disable this feature. If the no prepend option is used, the local-as will not be prepended to the updates received from the eBGP peer. If no prepend is not selected (the default), the local-as is added to the first AS segment in the AS-PATH. If an inbound route-map is used to prepend the as-path to the update from the peer, the local-as is added first. For example, consider the topology described in Figure 7-11. If Router B has an inbound route-map applied on Router C to prepend 65001 65002" to the as-path, the following events will take place on Router B
1. 2. 3. Receive and validate the update Prepend local-as 200 to as-path Prepend "65001 65002" to as-path

136

Border Gateway Protocol IPv4 (BGPv4)

Local-as is prepended before the route-map to give an impression that update passed thru a router in AS 200 before it reached Router B.

BGP4 Management Information Base (MIB)

The FORCE10-BGP4-V2-MIB enhances FTOS BGP Management Information Base (MIB) support with many new SNMP objects and notifications (traps) defined in the draft-ietf-idr-bgp4-mibv2-05. To see these enhancements, download the MIB from the Dell Networking website, www.force10networks.com.
Note: See the Dell Networking iSupport webpage for the Force10-BGP4-V2-MIB and other MIB documentation.

Important Points to Remember

In f10BgpM2AsPathTableEntry table, f10BgpM2AsPathSegmentIndex, and f10BgpM2AsPathElementIndex are used to retrieve a particular ASN from the AS path. These indices are assigned to the AS segments and individual ASN in each segment starting from 0. For example, an AS path list of {200 300 400} 500 consists of two segments: {200 300 400} with segment index 0 and 500 with segment index 1. ASN 200, 300, and 400 are be assigned 0, 1, and 2 element indices in that order. Unknown optional transitive attributes within a given path attribute (PA) are assigned indices in order. These indices correspond to f10BgpM2PathAttrUnknownIndex field in the f10BgpM2PathAttrUnknownEntry table. Negotiation of multiple instances of the same capability is not supported. F10BgpM2PeerCapAnnouncedIndex and f10BgpM2PeerCapReceivedIndex are ignored in the peer capability lookup. Inbound BGP soft-reconfiguration must be configured on a peer for f10BgpM2PrefixInPrefixesRejected to display the number of prefixes filtered due to a policy. If BGP soft-reconfig is not enabled, the denied prefixes are not accounted for. F10BgpM2AdjRibsOutRoute stores the pointer to the NLRI in the peer's Adj-Rib-Out. PA Index (f10BgpM2PathAttrIndex field in various tables) is used to retrieve specific attributes from the PA table. The Next-Hop, RR Cluster-list, Originator ID attributes are not stored in the PA Table and cannot be retrieved using the index passed in. These fields are not populated in f10BgpM2PathAttrEntry, f10BgpM2PathAttrClusterEntry, f10BgpM2PathAttrOriginatorIdEntry. F10BgpM2PathAttrUnknownEntry contains the optional-transitive attribute details. Query for f10BgpM2LinkLocalNextHopEntry returns default value for Link-local Next-hop. RFC 2545 and the f10BgpM2Rfc2545Group are not supported. An SNMP query will display up to 89 AS paths. A query for a larger AS path count will display as "" at the end of the output. SNMP set for BGP is not supported. For all peer configuration tables (f10BgpM2PeerConfigurationGroup, f10BgpM2PeerRouteReflectorCfgGroup, and f10BgpM2PeerAsConfederationCfgGroup), an SNMP set operation will return an error. Only SNMP queries are supported. In addition, the f10BgpM2CfgPeerError, f10BgpM2CfgPeerBgpPeerEntry, and f10BgpM2CfgPeerRowEntryStatus fields are to hold the SNMP set status and are ignored in SNMP query.

Border Gateway Protocol IPv4 (BGPv4) | 137

www.dell.com | support.dell.com

The AFI/SAFI is not used as an index to the f10BgpM2PeerCountersEntry table. The BGP peer's AFI/ SAFI (IPv4 Unicast or IPv6 Multicast) is used for various outbound counters. Counters corresponding to IPv4 Multicast cannot be queried. The f10BgpM2[Cfg]PeerReflectorClient field is populated based on the assumption that route-reflector clients are not in a full mesh if BGP client-2-client reflection is enabled and that the BGP speaker acting as reflector will advertise routes learned from one client to another client. If disabled, it is assumed that clients are in a full mesh, and there is no need to advertise prefixes to the other clients. High CPU utilization may be observed during an SNMP walk of a large BGP Loc-RIB. To avoid SNMP timeouts with a large-scale configuration (large number of BGP neighbors and a large BGP Loc-RIB), Dell Networking recommends setting the timeout and retry count values to a relatively higher number. e.g. t = 60 or r = 5. To return all values on an snmpwalk for the f10BgpM2Peer sub-OID, use the -C c option, such as snmpwalk -v 2c -C c -c public <IP_address> <OID>. An SNMP walk may terminate pre-maturely if the index does not increment lexicographically. Dell Networking recommends using options to ignore such errors. Multiple BPG process instances are not supported. Thus, the F10BgpM2PeerInstance field in various tables is not used to locate a peer. Multiple instances of the same NLRI in the BGP RIB are not supported and are set to zero in the SNMP query response. F10BgpM2NlriIndex and f10BgpM2AdjRibsOutIndex fields are not used. Carrying MPLS labels in BGP is not supported. F10BgpM2NlriOpaqueType and f10BgpM2NlriOpaquePointer fields are set to zero.

4-byte ASN is supported. f10BgpM2AsPath4byteEntry table contains 4-byte ASN-related parameters based on the configuration.

Traps (notifications) specified in the BGP4 MIB draft <draft-ietf-idr-bgp4-mibv2-05.txt> are not supported. Such traps (bgpM2Established and bgpM2BackwardTransition) are supported as part of RFC 1657.

Configuration Information
The software supports BGPv4 as well as the following: deterministic multi-exit discriminator (MED) (default) a path with a missing MED is treated as worst path and assigned an MED value of (0xffffffff) the community format follows RFC 1998 delayed configuration (the software at system boot reads the entire configuration file prior to sending messages to start BGP peer sessions)

The following are not yet supported: auto-summarization (the default is no auto-summary) synchronization (the default is no synchronization)

138

Border Gateway Protocol IPv4 (BGPv4)

BGP Configuration
To enable the BGP process and begin exchanging information, you must assign an AS number and use commands in the ROUTER BGP mode to configure a BGP neighbor.

Defaults
By default, BGP is disabled. By default, FTOS compares the MED attribute on different paths from within the same AS (the bgp always-compare-med command is not enabled).
Note: In FTOS, all newly configured neighbors and peer groups are disabled. You must enter the neighbor {ip-address | peer-group-name} no shutdown command to enable a neighbor or peer group.

Table 7-3 displays the default values for BGP on FTOS.

Table 7-3. Item BGP Neighbor Adjacency changes Fast External Fallover feature graceful restart feature Local preference MED Route Flap Damping Parameters FTOS BGP Defaults Default All BGP neighbor changes are logged. Enabled Disabled 100 0 half-life = 15 minutes reuse = 750 suppress = 2000 max-suppress-time = 60 minutes external distance = 20 internal distance = 200 local distance = 200 keepalive = 60 seconds holdtime = 180 seconds

Distance

Timers

Configuration Task List for BGP

The following list includes the configuration tasks for BGP: Enable BGP Configure AS4 Number Representations Configure Peer Groups BGP fast fall-over

Border Gateway Protocol IPv4 (BGPv4) | 139

Configure passive peering Maintain existing AS numbers during an AS migration Allow an AS number to appear in its own AS path Enable graceful restart Filter on an AS-Path attribute Configure IP community lists Manipulate the COMMUNITY attribute Change MED attribute Change LOCAL_PREFERENCE attribute Change NEXT_HOP attribute Change WEIGHT attribute Enable multipath Filter BGP routes Redistribute routes Configure BGP route reflectors Aggregate routes Configure BGP confederations Enable route flap dampening Change BGP timers BGP neighbor soft-reconfiguration Route map continue

www.dell.com | support.dell.com

Enable BGP
By default, BGP is not enabled on the system. FTOS supports one Autonomous System (AS) and you must assign the AS Number (ASN). To establish BGP sessions and route traffic, you must configure at least one BGP neighbor or peer. In BGP, routers with an established TCP connection are called neighbors or peers. Once a connection is established, the neighbors exchange full BGP routing tables with incremental updates afterwards. In addition, neighbors exchange KEEPALIVE messages to maintain the connection. In BGP, neighbor routers or peers can be classified as internal or external. External BGP peers must be connected physically to one another (unless you enable the EBGP multihop feature), while internal BGP peers do not need to be directly connected. The IP address of an EBGP neighbor is usually the IP address of the interface directly connected to the router. First, the BGP process determines if all internal BGP peers are reachable, and then it determines which peers outside the AS are reachable. Note: Sample Configurations for enabling BGP routers are found at the end of this chapter.

140

Border Gateway Protocol IPv4 (BGPv4)

Use these commands in the following sequence, starting in the CONFIGURATION mode to establish BGP sessions on the router.
Step 1 Command Syntax
router bgp as-number

Command Mode CONFIGURATION

Purpose Assign an AS number and enter the ROUTER BGP mode. AS Number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte) or 0.1-65535.65535 (Dotted format)

Only one AS is supported per system Note: If you enter a 4-Byte AS Number, 4-Byte AS Support is enabled automatically. 1a bgp four-octet-as-support CONFIG-ROUTER-B GP Enable 4-Byte support for the BGP process. Note: This is an OPTIONAL command. Enable if you want to use 4-Byte AS numbers or if you support AS4 Number Representation.

Note: Use it only if you support 4-Byte AS Numbers or if you support AS4 Number Representation. If you are supporting 4-Byte ASNs, this command must be enabled first. Disable 4-Byte support and return to the default 2-Byte format by using the no bgp four-octet-as-support command. You cannot disable 4-Byte support if you currently have a 4-Byte ASN configured. FTOS Behavior: Disabling 4-Byte AS Numbers also disables ASDOT and ASDOT+ number representation. All AS Numbers will be displayed in ASPLAIN format. 1b address-family [ipv4 | ipv6} CONFIG-ROUTER-B GP CONFIG-ROUTER-B GP Enable IPv4 multicast or IPv6 mode. Use this command to enter BGP for IPv6 mode (CONF-ROUTER_BGPv6_AF). Add a neighbor as a remote AS. Formats: IP Address A.B.C.D Peer-Group Name: 16 characters AS-number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte) or 0.1-65535.65535 (Dotted format)

neighbor {ip-address | peer-group name} remote-as as-number

You must Configure Peer Groups before assigning it a remote AS.

neighbor {ip-address | peer-group-name} no shutdown

CONFIG-ROUTER-B GP

Enable the BGP neighbor.

Note: When you change the configuration of a BGP neighbor, always reset it by entering the clear ip bgp command in EXEC Privilege mode.

Border Gateway Protocol IPv4 (BGPv4) | 141

www.dell.com | support.dell.com

Enter show config in CONFIGURATION ROUTER BGP mode to view the BGP configuration. Use the show ip bgp summary command in EXEC Privilege mode to view the BGP status. Figure 7-12 shows the summary with a 2-Byte AS Number displayed; Figure 7-13 shows the summary with a 4-Byte AS Number displayed.
Figure 7-12. Command example: show ip bgp summary (2-Byte AS Number displayed)
R2#show ip bgp summary BGP router identifier 192.168.10.2, local AS number 65123 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 1 paths using 72 bytes of memory BGP-RIB over all using 73 bytes of memory 1 BGP path attribute entrie(s) using 72 bytes of memory 1 BGP AS-PATH entrie(s) using 47 bytes of memory 5 neighbor(s) using 23520 bytes of memory Neighbor 10.10.21.1 10.10.32.3 100.10.92.9 192.168.10.1 192.168.12.2 R2# AS 65123 65123 65192 65123 65123 MsgRcvd 0 0 0 0 0 MsgSent 0 0 0 0 0 TblVer 0 0 0 0 0 InQ 0 0 0 0 0

2-Byte AS Number

OutQ Up/Down 0 0 0 0 0 never never never never never

State/Pfx Active Active Active Active Active

Figure 7-13. Command example: show ip bgp summary (4-Byte AS Number displayed)
R2#show ip bgp summary BGP router identifier 192.168.10.2, local AS number 48735.59224 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 1 paths using 72 bytes of memory BGP-RIB over all using 73 bytes of memory 1 BGP path attribute entrie(s) using 72 bytes of memory 1 BGP AS-PATH entrie(s) using 47 bytes of memory 5 neighbor(s) using 23520 bytes of memory Neighbor 10.10.21.1 10.10.32.3 100.10.92.9 192.168.10.1 192.168.12.2 R2# AS 65123 65123 65192 65123 65123 MsgRcvd 0 0 0 0 0 MsgSent 0 0 0 0 0 TblVer 0 0 0 0 0 InQ 0 0 0 0 0

4-Byte AS Number

OutQ Up/Down 0 0 0 0 0 never never never never never

State/Pfx Active Active Active Active Active

For the routers identifier, FTOS uses the highest IP address of the Loopback interfaces configured. Since Loopback interfaces are virtual, they cannot go down, thus preventing changes in the router ID. If no Loopback interfaces are configured, the highest IP address of any interface is used as the router ID. To view the status of BGP neighbors, use the show ip bgp neighbors (Figure 7-14) command in EXEC Privilege mode. For BGP neighbor configuration information, use the show running-config bgp command in EXEC Privilege mode (Figure 7-15). Note that the showconfig command in CONFIGURATION ROUTER BGP mode gives the same information as the show running-config bgp.

142

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-14 displays two neighbors, one is an external and the second one is an internal BGP neighbor. The first line of the output for each neighbor displays the AS number and states whether the link is an external or internal. The third line of the show ip bgp neighbors output contains the BGP State. If anything other than ESTABLISHED is listed, the neighbor is not exchanging information and routes. For more details on using the show ip bgp neighbors command, refer to the FTOS Command Line Interface Reference.
Figure 7-14. Command example: show ip bgp neighbors
FTOS#show ip bgp neighbors

BGP neighbor is 10.114.8.60, remote AS 18508, external link External BGP version 4, remote router ID 10.20.20.20 BGP state ESTABLISHED, in this state for 00:01:58 Last read 00:00:14, hold time is 90, keepalive interval is 30 seconds Received 18552 messages, 0 notifications, 0 in queue Sent 11568 messages, 0 notifications, 0 in queue Received 18549 updates, Sent 11562 updates Minimum time between advertisement runs is 30 seconds

BGP neighbor

For address family: IPv4 Unicast BGP table version 216613, neighbor version 201190 130195 accepted prefixes consume 520780 bytes Prefix advertised 49304, rejected 0, withdrawn 36143 Connections established 1; dropped 0 Last reset never Local host: 10.114.8.39, Local port: 1037 Foreign host: 10.114.8.60, Foreign port: 179

BGP neighbor is 10.1.1.1, remote AS 65535, internal link Internal Administratively shut down BGP version 4, remote router ID 10.0.0.0 BGP state IDLE, in this state for 17:12:40 Last read 17:12:40, hold time is 180, keepalive interval is 60 seconds Received 0 messages, 0 notifications, 0 in queue Sent 0 messages, 0 notifications, 0 in queue Received 0 updates, Sent 0 updates Minimum time between advertisement runs is 5 seconds

BGP neighbor

For address family: IPv4 Unicast BGP table version 0, neighbor version 0 0 accepted prefixes consume 0 bytes Prefix advertised 0, rejected 0, withdrawn 0 Connections established 0; dropped 0 Last reset never No active TCP connection FTOS#

Border Gateway Protocol IPv4 (BGPv4) | 143

www.dell.com | support.dell.com

Figure 7-15. Command example: show running-config bgp

R2#show running-config bgp ! router bgp 65123 bgp router-id 192.168.10.2 network 10.10.21.0/24 network 10.10.32.0/24 network 100.10.92.0/24 network 192.168.10.0/24 bgp four-octet-as-support neighbor 10.10.21.1 remote-as 65123 neighbor 10.10.21.1 filter-list ISP1in neighbor 10.10.21.1 no shutdown neighbor 10.10.32.3 remote-as 65123 neighbor 10.10.32.3 no shutdown neighbor 100.10.92.9 remote-as 65192 neighbor 100.10.92.9 no shutdown neighbor 192.168.10.1 remote-as 65123 neighbor 192.168.10.1 update-source Loopback 0 neighbor 192.168.10.1 no shutdown neighbor 192.168.12.2 remote-as 65123 neighbor 192.168.12.2 update-source Loopback 0 neighbor 192.168.12.2 no shutdown R2#

Configure AS4 Number Representations

Enable one type of AS Number Representation: ASPLAIN, ASDOT+, or ASDOT. ASPLAIN is the method FTOS has used for all previous FTOS versions. It remains the default method with FTOS 8.2.1.0 and later. With the ASPLAIN notation, a 32 bit binary AS number is translated into a decimal value. ASDOT+ representation splits the full binary 4-byte AS number into two words of 16 bits separated by a decimal point (.): <high-order 16 bit value>.<low-order 16 bit value>. ASDOT representation combines the ASPLAIN and ASDOT+ representations. AS Numbers less than 65536 appear in integer format (asplain); AS Numbers equal to or greater than 65536 appear using the decimal method (asdot+). For example, the AS Number 65526 appears as 65526, and the AS Number 65546 appears as 1.10.
Note: The ASDOT and ASDOT+ representations are supported only in conjunction with the 4-Byte AS Numbers feature. If 4-Byte AS Numbers are not implemented, only ASPLAIN representation is supported.

Only one form of AS Number Representation is supported at a time. You cannot combine the types of representations within an AS.
Task Enable ASPLAIN AS Number representation. Figure 7-16 Command Syntax
bgp asnotation asplain

Command Mode CONFIG-ROUTER-BGP

Note: ASPLAIN is the default method FTOS uses and does not appear in the configuration display.

144

Border Gateway Protocol IPv4 (BGPv4)

Task Enable ASDOT AS Number representation. Figure 7-17 Enable ASDOT+ AS Number representation.Figure 7-18

Command Syntax
bgp asnotation asdot

Command Mode CONFIG-ROUTER-BGP CONFIG-ROUTER-BGP

bgp asnotation asdot+

Figure 7-16. Command example and output: bgp asnotation asplain

FTOS(conf-router_bgp)#bgp asnotation asplain FTOS(conf-router_bgp)#sho conf ! router bgp 100 bgp four-octet-as-support neighbor 172.30.1.250 remote-as 18508 neighbor 172.30.1.250 local-as 65057 neighbor 172.30.1.250 route-map rmap1 in neighbor 172.30.1.250 password 7 5ab3eb9a15ed02ff4f0dfd4500d6017873cfd9a267c04957 neighbor 172.30.1.250 no shutdown 5332332 9911991 65057 18508 12182 7018 46164 i

Figure 7-17. Command example and output: bgp asnotation asdot

FTOS(conf-router_bgp)#bgp asnotation asdot FTOS(conf-router_bgp)#sho conf ! router bgp 100 bgp asnotation asdot bgp four-octet-as-support neighbor 172.30.1.250 remote-as 18508 neighbor 172.30.1.250 local-as 65057 neighbor 172.30.1.250 route-map rmap1 in neighbor 172.30.1.250 password 7 5ab3eb9a15ed02ff4f0dfd4500d6017873cfd9a267c04957 neighbor 172.30.1.250 no shutdown 5332332 9911991 65057 18508 12182 7018 46164 i

Figure 7-18. Command example and output: bgp asnotation asdot+

FTOS(conf-router_bgp)#bgp asnotation asdot+ FTOS(conf-router_bgp)#sho conf ! router bgp 100 bgp asnotation asdot+ bgp four-octet-as-support neighbor 172.30.1.250 remote-as 18508 neighbor 172.30.1.250 local-as 65057 neighbor 172.30.1.250 route-map rmap1 in neighbor 172.30.1.250 password 7 5ab3eb9a15ed02ff4f0dfd4500d6017873cfd9a267c04957 neighbor 172.30.1.250 no shutdown 5332332 9911991 65057 18508 12182 7018 46164 i

Border Gateway Protocol IPv4 (BGPv4) | 145

www.dell.com | support.dell.com

Configure Peer Groups

To configure multiple BGP neighbors at one time, create and populate a BGP peer group. Another advantage of peer groups is that members of a peer groups inherit the configuration properties of the group and share same update policy. A maximum of 256 Peer Groups are allowed on the system. You create a peer group by assigning it a name, then adding members to the peer group. Once a peer group is created, you can configure route policies for it. Refer to Filter BGP routes for information on configuring route policies for a peer group. Note: Sample Configurations for enabling Peer Groups are found at the end of this chapter. Use these commands in the following sequence starting in the CONFIGURATION ROUTER BGP mode to create a peer group
Step 1 Command Syntax
neighbor peer-group-name peer-group neighbor peer-group-name no shutdown neighbor ip-address remote-as
as-number

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Create a peer group by assigning a name to it. Enable the peer group. By default, all peer groups are disabled Create a BGP neighbor. Enable the neighbor. Add an enabled neighbor to the peer group. Add a neighbor as a remote AS. Formats: IP Address A.B.C.D Peer-Group Name16 characters AS-number: 0-65535 (2-Byte) or 1-4294967295 | 0.1- 65535.65535 (4-Byte) or 0.1-65535.65535 (Dotted format)

2 3 4 5 6

neighbor ip-address no shutdown neighbor ip-address peer-group

peer-group-name

neighbor {ip-address | peer-group name} remote-as as-number

To add an external BGP (EBGP) neighbor, configure the as-number parameter with a number different from the BGP as-number configured in the router bgp as-number command. To add an internal BGP (IBGP neighbor, configure the as-number parameter with the same BGP as-number configured in the router bgp as-number command.

146

Border Gateway Protocol IPv4 (BGPv4)

After you create a peer group, you can use any of the commands beginning with the keyword neighbor to configure that peer group. When you add a peer to a peer group, it inherits all the peer groups configured parameters. A neighbor cannot become part of a peer group if it has any of the following commands are configured: neighbor advertisement-interval neighbor distribute-list out neighbor filter-list out neighbor next-hop-self neighbor route-map out neighbor route-reflector-client neighbor send-community

A neighbor may keep its configuration after it was added to a peer group if the neighbors configuration is more specific than the peer groups, and the neighbors configuration does not affect outgoing updates.
Note: When you configure a new set of BGP policies for a peer group, always reset the peer group by entering the clear ip bgp peer-group peer-group-name command in EXEC Privilege mode.

Use the show config command in the CONFIGURATION ROUTER BGP mode to view the configuration. When you create a peer group, it is disabled (shutdown). Figure 7-19 shows the creation of a peer group (zanzibar).
Figure 7-19. Command example: show config (creating peer-group)
FTOS(conf-router_bgp)#neighbor zanzibar peer-group FTOS(conf-router_bgp)#show conf ! router bgp 45 bgp fast-external-fallover bgp log-neighbor-changes neighbor zanzibar peer-group neighbor zanzibar shutdown neighbor 10.1.1.1 remote-as 65535 neighbor 10.1.1.1 shutdown neighbor 10.14.8.60 remote-as 18505 neighbor 10.14.8.60 no shutdown FTOS(conf-router_bgp)#

Configuring neighbor zanzibar

Use the neighbor peer-group-name no shutdown command in the CONFIGURATION ROUTER BGP mode to enable a peer group.

Border Gateway Protocol IPv4 (BGPv4) | 147

www.dell.com | support.dell.com

Figure 7-20. Command example: show config (peer-group enabled

FTOS(conf-router_bgp)#neighbor zanzibar no shutdown FTOS(conf-router_bgp)#show config ! router bgp 45 bgp fast-external-fallover bgp log-neighbor-changes neighbor zanzibar peer-group neighbor zanzibar no shutdown neighbor 10.1.1.1 remote-as 65535 neighbor 10.1.1.1 shutdown neighbor 10.14.8.60 remote-as 18505 neighbor 10.14.8.60 no shutdown FTOS(conf-router_bgp)#

Enabling neighbor zanzibar

To disable a peer group, use the neighbor peer-group-name shutdown command in the CONFIGURATION ROUTER BGP mode. The configuration of the peer group is maintained, but it is not applied to the peer group members. When you disable a peer group, all the peers within the peer group that are in ESTABLISHED state are moved to IDLE state. Use the show ip bgp peer-group command in EXEC Privilege mode (Figure 7-21) to view the status of peer groups.

148

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-21. Command example: show ip bgp peer-group

FTOS>show ip bgp peer-group

Peer-group zanzibar, remote AS 65535 BGP version 4 Minimum time between advertisement runs is 5 seconds For address family: IPv4 Unicast BGP neighbor is zanzibar, peer-group internal, Number of peers in this group 26 Peer-group members (* - outbound optimized): 10.68.160.1 10.68.161.1 10.68.162.1 10.68.163.1 10.68.164.1 10.68.165.1 10.68.166.1 10.68.167.1 10.68.168.1 10.68.169.1 10.68.170.1 10.68.171.1 10.68.172.1 10.68.173.1 10.68.174.1 10.68.175.1 10.68.176.1 10.68.177.1 10.68.178.1 10.68.179.1 10.68.180.1 10.68.181.1 10.68.182.1 10.68.183.1 10.68.184.1 10.68.185.1 FTOS>

BGP fast fall-over

By default, a BGP session is governed by the hold time. BGP routers typically carry large routing tables, so frequent session resets are not desirable. The BGP fast fall-over feature reduces the convergence time while maintaining stability. The connection to a BGP peer is immediately reset if a link to a directly connected external peer fails. When fall-over is enabled, BGP tracks IP reachability to the peer remote address and the peer local address. Whenever either address becomes unreachable (for example, no active route exists in the routing table for peer IPv6 destinations/local address), BGP brings down the session with the peer.

Border Gateway Protocol IPv4 (BGPv4) | 149

www.dell.com | support.dell.com

The BGP fast fall-over feature is configured on a per-neighbor or peer-group basis and is disabled by default.
Command Syntax
neighbor {ip-address | peer-group-name} fall-over

Command Mode CONFIG-ROUTER-BGP

Purpose Enable BGP Fast Fall-Over

To disable Fast Fall-Over, use the [no] neighbor [neighbor | peer-group] fall-over command in CONFIGURATION ROUTER BGP mode Use the show ip bgp neighbors command as shown in Figure 7-22 to verify that fast fall-over is enabled on a particular BGP neighbor. Note that since Fast Fall-Over is disabled by default, it will appear only if it has been enabled

150

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-22. Command example: show ip bgp neighbors

FTOS#sh ip bgp neighbors BGP neighbor is 100.100.100.100, remote AS 65517, internal link Member of peer-group test for session parameters BGP version 4, remote router ID 30.30.30.5 BGP state ESTABLISHED, in this state for 00:19:15 Last read 00:00:15, last write 00:00:06 Hold time is 180, keepalive interval is 60 seconds Received 52 messages, 0 notifications, 0 in queue Sent 45 messages, 5 notifications, 0 in queue Received 6 updates, Sent 0 updates Route refresh request: received 0, sent 0 Minimum time between advertisement runs is 5 seconds Minimum time before advertisements start is 0 seconds Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Fall-over enabled Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 52, neighbor version 52 4 accepted prefixes consume 16 bytes Prefix advertised 0, denied 0, withdrawn 0 Connections established 6; dropped 5 Last reset 00:19:37, due to Reset by peer Notification History 'Connection Reset' Sent : 5

Fast Fall-Over Indicator

Recv: 0

Local host: 200.200.200.200, Local port: 65519 Foreign host: 100.100.100.100, Foreign port: 179 FTOS#

Use the show ip bgp peer-group command to verify that fast fall-over is enabled on a peer-group.

Border Gateway Protocol IPv4 (BGPv4) | 151

www.dell.com | support.dell.com

Figure 7-23. Command example: show ip bgp peer-group

FTOS#sh ip bgp peer-group Peer-group test Fall-over enabled BGP version 4 Minimum time between advertisement runs is 5 seconds For address family: IPv4 Unicast BGP neighbor is test Number of peers in this group 1 Peer-group members (* - outbound optimized): 100.100.100.100* FTOS# router bgp 65517 neighbor test peer-group Fast Fall-Over neighbor test fall-over neighbor test no shutdown neighbor 100.100.100.100 remote-as 65517 neighbor 100.100.100.100 fall-over neighbor 100.100.100.100 update-source Loopback 0 neighbor 100.100.100.100 no shutdown FTOS#

Indicator

Configure passive peering

When you enable a peer-group, the software sends an OPEN message to initiate a TCP connection. If you enable passive peering for the peer group, the software does not send an OPEN message, but it will respond to an OPEN message. When a BGP neighbor connection with authentication configured is rejected by a passive peer-group, FTOS does not allow another passive peer-group on the same subnet to connect with the BGP neighbor. To work around this, change the BGP configuration or change the order of the peer group configuration. Use these commands in the following sequence, starting in the CONFIGURATION ROUTER BGP mode to configure passive peering.
Step 1 Command Syntax
neighbor peer-group-name peer-group passive neighbor peer-group-name subnet subnet-number mask

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Configure a peer group that does not initiate TCP connections with other peers. Assign a subnet to the peer group. The peer group will respond to OPEN messages sent on this subnet. Enable the peer group. Create and specify a remote peer for BGP neighbor.

3 4

neighbor peer-group-name no shutdown neighbor peer-group-name remote-as as-number

152

Border Gateway Protocol IPv4 (BGPv4)

Only after the peer group responds to an OPEN message sent on the subnet does its BGP state change to ESTABLISHED. Once the peer group is ESTABLISHED, the peer group is the same as any other peer group. For more information on peer groups, refer to Configure Peer Groups.

Maintain existing AS numbers during an AS migration

The local-as feature smooths out the BGP network migration operation and allows you to maintain existing ASNs during a BGP network migration. When you complete your migration, be sure to reconfigure your routers with the new information and disable this feature.
Command Syntax
neighbor {IP address | peer-group-name local-as as

Command Mode CONFIG-ROUTERBGP

Purpose Allow external routes from this neighbor. Format: IP Address: A.B.C.D Peer Group Name: 16 characters AS-number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte) or 0.1-65535.65535 (Dotted format) No Prepend specifies that local AS values are not prepended to announcements from the neighbor.

number [no prepend]

You must Configure Peer Groups before assigning it to an AS. This feature is not supported on passive peer groups.

Disable this feature, using the no neighbor local-as command in CONFIGURATION ROUTER BGP mode.

Border Gateway Protocol IPv4 (BGPv4) | 153

www.dell.com | support.dell.com

Figure 7-24. Local-as information shown

R2(conf-router_bgp)#show conf ! router bgp 65123 bgp router-id 192.168.10.2 network 10.10.21.0/24 network 10.10.32.0/24 network 100.10.92.0/24 network 192.168.10.0/24 bgp four-octet-as-support neighbor 10.10.21.1 remote-as 65123 neighbor 10.10.21.1 filter-list Laura in neighbor 10.10.21.1 no shutdown neighbor 10.10.32.3 remote-as 65123 neighbor 10.10.32.3 no shutdown neighbor 100.10.92.9 remote-as 65192 neighbor 100.10.92.9 local-as 6500 neighbor 100.10.92.9 no shutdown neighbor 192.168.10.1 remote-as 65123 neighbor 192.168.10.1 update-source Loopback 0 neighbor 192.168.10.1 no shutdown neighbor 192.168.12.2 remote-as 65123 neighbor 192.168.12.2 update-source Loopback 0 neighbor 192.168.12.2 no shutdown R2(conf-router_bgp)#

Actual AS Number

Local-AS Number 6500 Maintained During Migration

Allow an AS number to appear in its own AS path

This command allows you to set the number of times a particular AS number can occur in the AS path. The allow-as feature permits a BGP speaker to allow the ASN to be present for specified number of times in the update received from the peer, even if that ASN matches its own. The AS-PATH loop is detected if the local ASN is present more than the specified number of times in the command.
Command Syntax
neighbor {IP address | peer-group-name} allowas-in

Command Mode CONFIG-ROUTERBGP

Purpose Allow this neighbor ID to use the AS path the specified number of times. Format: IP Address: A.B.C.D Peer Group Name: 16 characters Number: 1-10

number

You must Configure Peer Groups before assigning it to an AS.

To disable this feature, use the no neighbor allow-as in number command in the CONFIGURATION ROUTER BGP mode.

154

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-25. Allowas-in information shown

R2(conf-router_bgp)#show conf ! router bgp 65123 bgp router-id 192.168.10.2 network 10.10.21.0/24 network 10.10.32.0/24 network 100.10.92.0/24 network 192.168.10.0/24 bgp four-octet-as-support neighbor 10.10.21.1 remote-as 65123 neighbor 10.10.21.1 filter-list Laura in neighbor 10.10.21.1 no shutdown neighbor 10.10.32.3 remote-as 65123 neighbor 10.10.32.3 no shutdown neighbor 100.10.92.9 remote-as 65192 neighbor 100.10.92.9 local-as 6500 neighbor 100.10.92.9 no shutdown neighbor 192.168.10.1 remote-as 65123 neighbor 192.168.10.1 update-source Loopback 0 neighbor 192.168.10.1 no shutdown neighbor 192.168.12.2 remote-as 65123 neighbor 192.168.12.2 allowas-in 9 neighbor 192.168.12.2 update-source Loopback 0 neighbor 192.168.12.2 no shutdown R2(conf-router_bgp)#R2(conf-router_bgp)#

Number of Times ASN 65123 Can Appear in AS PATH

Enable graceful restart

Use this feature to lessen the negative effects of a BGP restart. FTOS advertises support for this feature to BGP neighbors through a capability advertisement. You can enable graceful restart by router and/or by peer or peer group. Note: By default, BGP graceful restart is disabled. The default role for BGP on is as a receiving or restarting peer. If you enable BGP, when a peer that supports graceful restart resumes operating, FTOS performs the following tasks: Continues saving routes received from the peer if the peer advertised it had graceful restart capability. Continues forwarding traffic to the peer. Flags routes from the peer as Stale and sets a timer to delete them if the peer does not perform a graceful restart. Deletes all routes from the peer if forwarding state information is not saved. Speeds convergence by advertising a special update packet known as an end-of-RIB marker. This marker indicates the peer has been updated with all routes in the local RIB.

If you configure your system to do so, FTOS can perform the following actions during a hot failover: Save all FIB and CAM entries on the line card and continue forwarding traffic while the secondary RPM is coming online.

Border Gateway Protocol IPv4 (BGPv4) | 155

www.dell.com | support.dell.com

Advertise to all BGP neighbors and peer-groups that the forwarding state of all routes has been saved. This prompts all peers to continue saving the routes they receive from your E-Series and to continue forwarding traffic. Bring the secondary RPM online as the primary and re-open sessions with all peers operating in no shutdown mode. Defer best path selection for a certain amount of time. This helps optimize path selection and results in fewer updates being sent out.

Enable graceful restart using the configure router bgp graceful-restart command. The table below shows the command and its available options:
Command Syntax
bgp graceful-restart bgp graceful-restart [restart-time time-in-seconds] bgp graceful-restart [stale-path-time time-in-seconds] bgp graceful-restart [role receiver-only]

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Usage Enable graceful restart for the BGP node. Set maximum restart time for all peers. Default is 120 seconds. Set maximum time to retain the restarting peers stale paths. Default is 360 seconds. Local router supports graceful restart as a receiver only.

BGP graceful restart is active only when the neighbor becomes established. Otherwise it is disabled. Graceful-restart applies to all neighbors with established adjacency. With the graceful restart feature, FTOS enables the receiving/restarting mode by default. In receiver-only mode, graceful restart saves the advertised routes of peers that support this capability when they restart. However, the E-Series does not advertise that it saves these forwarding states when it restarts. This option provides support for remote peers for their graceful restart without supporting the feature itself. You can implement BGP graceful restart either by neighbor or by BGP peer-group. For more information, please see the following table or the FTOS Command Line Interface Reference.
Command Syntax
neighbor {ip-address | peer-group-name} graceful-restart neighbor {ip-address | peer-group-name} graceful-restart [restart-time time-in-seconds] neighbor {ip-address | peer-group-name} graceful-restart [role receiver-only] neighbor {ip-address | peer-group-name} graceful-restart [stale-path-time time-in-seconds]

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Add graceful restart to a BGP neighbor or peer-group. Set maximum restart time for the neighbor or peer-group. Default is 120 seconds. Local router supports graceful restart for this neighbor or peer-group as a receiver only. Set maximum time to retain the restarting neighbors or peer-groups stale paths. Default is 360 seconds.

156

Border Gateway Protocol IPv4 (BGPv4)

Filter on an AS-Path attribute

The BGP attribute, AS_PATH, can be used to manipulate routing policies. The AS_PATH attribute contains a sequence of AS numbers representing the routes path. As the route traverses an Autonomous System, the AS number is prepended to the route. You can manipulate routes based on their AS_PATH to affect interdomain routing. By identifying certain AS numbers in the AS_PATH, you can permit or deny routes based on the number in its AS_PATH. To view all BGP path attributes in the BGP database, use the show ip bgp paths command in EXEC Privilege mode (Figure 7-26).
Figure 7-26. Command example: show ip bgp paths
FTOS#show ip bgp paths Total 30655 Paths Address Hash Refcount Metric Path 0x4014154 0 3 18508 701 3549 19421 i 0x4013914 0 3 18508 701 7018 14990 i 0x5166d6c 0 3 18508 209 4637 1221 9249 9249 i 0x5e62df4 0 2 18508 701 17302 i 0x3a1814c 0 26 18508 209 22291 i 0x567ea9c 0 75 18508 209 3356 2529 i 0x6cc1294 0 2 18508 209 1239 19265 i 0x6cc18d4 0 1 18508 701 2914 4713 17935 i 0x5982e44 0 162 18508 209 i 0x67d4a14 0 2 18508 701 19878 ? 0x559972c 0 31 18508 209 18756 i 0x59cd3b4 0 2 18508 209 7018 15227 i 0x7128114 0 10 18508 209 3356 13845 i 0x536a914 0 3 18508 209 701 6347 7781 i 0x2ffe884 0 1 18508 701 3561 9116 21350 i 0x2ff7284 0 99 18508 701 1239 577 855 ? 0x2ff7ec4 0 4 18508 209 3561 4755 17426 i 0x2ff8544 0 3 18508 701 5743 2648 i 0x736c144 0 1 18508 701 209 568 721 1494 i 0x3b8d224 0 10 18508 209 701 2019 i 0x5eb1e44 0 1 18508 701 8584 16158 i 0x5cd891c 0 9 18508 209 6453 4759 i --More--

AS-PATH ACLs use regular expressions to search AS_PATH values. AS-PATH ACLs have an implicit deny. This means that routes that do not meet a deny or match filter are dropped. Use these commands in the following sequence, starting in the CONFIGURATION mode to configure an AS-PATH ACL to filter a specific AS_PATH value.
Step 1 Command Syntax
ip as-path access-list
as-path-name

Command Mode CONFIGURATION

Purpose Assign a name to a AS-PATH ACL and enter AS-PATH ACL mode.

Border Gateway Protocol IPv4 (BGPv4) | 157

www.dell.com | support.dell.com

Step

Command Syntax {deny | permit} filter

parameter

Command Mode CONFIG-AS-PATH

Purpose Enter the parameter to match BGP AS-PATH for filtering. This is the filter that will be used to match the AS-path. The entries can be any format, letters, numbers, or regular expressions. This command can be entered multiple times if multiple filters are desired. See Table 7-4 for accepted expressions. Return to CONFIGURATION mode Enter ROUTER BGP mode. Use a configured AS-PATH ACL for route filtering and manipulation. If you assign an non-existent or empty AS-PATH ACL, the software allows all routes.

3 4 5

exit router bgp as-number neighbor {ip-address |

peer-group-name} filter-list as-path-name {in

AS-PATH ACL CONFIGURATION CONFIG-ROUTER-B GP

| out}

Regular Expressions as filters

Regular expressions are used to filter AS paths or community lists. A regular expression is a special character used to define a pattern that is then compared with an input string. For an AS-path access list as shown in the commands above, if the AS path matches the regular expression in the access list, then the route matches the access list. Figure 7-27 applies access list Eagle to routes inbound from BGP peer 10.5.5.2. Access list Eagle uses a regular expression to deny routes originating in AS 32.

158

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-27. Filtering with Regular Expression

FTOS(config)#router bgp 99 FTOS(conf-router_bgp)#neigh AAA peer-group FTOS(conf-router_bgp)#neigh AAA no shut FTOS(conf-router_bgp)#show conf ! router bgp 99 neighbor AAA peer-group neighbor AAA no shutdown neighbor 10.155.15.2 remote-as 32 neighbor 10.155.15.2 shutdown FTOS(conf-router_bgp)#neigh 10.155.15.2 filter-list 1 in FTOS(conf-router_bgp)#ex FTOS(conf)#ip as-path access-list Eagle FTOS(config-as-path)#deny 32$ FTOS(config-as-path)#ex FTOS(conf)#router bgp 99 FTOS(conf-router_bgp)#neighbor AAA filter-list Eagle in FTOS(conf-router_bgp)#show conf ! router bgp 99 neighbor AAA peer-group neighbor AAA filter-list Eaglein neighbor AAA no shutdown neighbor 10.155.15.2 remote-as 32 neighbor 10.155.15.2 filter-list 1 in neighbor 10.155.15.2 shutdown FTOS(conf-router_bgp)#ex FTOS(conf)#ex FTOS#show ip as-path-access-lists ip as-path access-list Eagle deny 32$ FTOS#

Create the Access List and Filter

Regular Expression shown as part of Access List filter

Table 7-4 lists the Regular Expressions accepted in FTOS.

Table 7-4. Regular Expressions

Regular Expression ^ (carrot)

Definition Matches the beginning of the input string. Alternatively, when used as the first character within brackets [^ ] matches any number except the ones specified within the brackets.

$ (dollar) . (period) * (asterisk) + (plus) ? (question)

Matches the end of the input string. Matches any single character, including white space. Matches 0 or more sequences of the immediately previous character or pattern. Matches 1 or more sequences of the immediately previous character or pattern. Matches 0 or 1 sequence of the immediately previous character or pattern.

Border Gateway Protocol IPv4 (BGPv4) | 159

www.dell.com | support.dell.com

Table 7-4.

Regular Expressions

Regular Expression ( ) (parenthesis) [ ] (brackets) - (hyphen) _ (underscore)

Definition Specifies patterns for multiple use when followed by one of the multiplier metacharacters: asterisk *, plus sign +, or question mark ? Matches any enclosed character; specifies a range of single characters Used within brackets to specify a range of AS or community numbers. Matches a ^, a $, a comma, a space, a {, or a }. Placed on either side of a string to specify a literal and disallow substring matching. Numerals enclosed by underscores can be preceded or followed by any of the characters listed above. Matches characters on either side of the metacharacter; logical OR.

| (pipe)

As seen in Figure 7-27, the expressions are displayed when using the show commands. Use the show config command in the CONFIGURATION AS-PATH ACL mode and the show ip as-path-access-list command in EXEC Privilege mode to view the AS-PATH ACL configuration. For more information on this command and route filtering, refer to Filter BGP routes.

Redistribute routes
In addition to filtering routes, you can add routes from other routing instances or protocols to the BGP process. With the redistribute command syntax, you can include ISIS, OSPF, static, or directly connected routes in the BGP process. Use any of the following commands in ROUTER BGP mode to add routes from other routing instances or protocols.
Command Syntax
redistribute {connected | static} [route-map map-name]

Command Mode ROUTER BGP or CONF-ROUTER_BGPv6_ AF

Purpose Include, directly connected or user-configured (static) routes in BGP. Configure the following parameters: map-name: name of a configured route map. Include specific ISIS routes in BGP. Configure the following parameters: level-1, level-1-2, or level-2: Assign all redistributed routes to a level. Default is level-2. metric range: 0 to 16777215. Default is 0. map-name: name of a configured route map.

redistribute isis [level-1 | level-1-2 | level-2] [metric value] [route-map map-name]

ROUTER BGP or CONF-ROUTER_BGPv6_ AF

160

Border Gateway Protocol IPv4 (BGPv4)

Command Syntax
redistribute ospf process-id [match external {1 | 2} | match internal] [metric-type {external | internal}] [route-map map-name]

Command Mode ROUTER BGP or CONF-ROUTER_BGPv6_ AF

Purpose Include specific OSPF routes in IS-IS. Configure the following parameters: process-id range: 1 to 65535 match external range: 1 or 2 match internal metric-type: external or internal. map-name: name of a configured route map.

Configure IP community lists

Within FTOS, you have multiple methods of manipulating routing attributes. One attribute you can manipulate is the COMMUNITY attribute. This attribute is an optional attribute that is defined for a group of destinations. In FTOS, you can assign a COMMUNITY attribute to BGP routers by using an IP Community list. After you create an IP Community list, you can apply routing decisions to all routers meeting the criteria in the IP Community list. IETF RFC 1997 defines the COMMUNITY attribute and the pre-defined communities of INTERNET, NO_EXPORT_SUBCONFED, NO_ADVERTISE, and NO_EXPORT. All BGP routes belong to the INTERNET community. In the RFC, the other communities are defined as follows: All routes with the NO_EXPORT_SUBCONFED (0xFFFFFF03) community attribute are not sent to CONFED-EBGP or EBGP peers, but are sent to IBGP peers within CONFED-SUB-AS. All routes with the NO_ADVERTISE (0xFFFFFF02) community attribute must not be advertised. All routes with the NO_EXPORT (0xFFFFFF01) community attribute must not be advertised outside a BGP confederation boundary, but are sent to CONFED-EBGP and IBGP peers.

FTOS also supports BGP Extended Communities as described in RFC 4360BGP Extended Communities Attribute.

Border Gateway Protocol IPv4 (BGPv4) | 161

www.dell.com | support.dell.com

Use these commands in the following sequence, starting in the CONFIGURATION mode to configure an IP community list.
Step 1 Command Syntax
ip community-list
community-list-name

Command Mode CONFIGURATION CONFIG-COMMUNITYLIST

Purpose Create a Community list and enter the COMMUNITY-LIST mode. Configure a Community list by denying or permitting specific community numbers or types of community community-number: use AA:NN format where AA is the AS number (2 or 4 Bytes) and NN is a value specific to that autonomous system. local-AS: routes with the COMMUNITY attribute of NO_EXPORT_SUBCONFED. no-advertise: routes with the COMMUNITY attribute of NO_ADVERTISE. no-export: routes with the COMMUNITY attribute of NO_EXPORT. quote-regexp: followed by any number of regular expressions. The software applies all regular expressions in the list. regexp: followed by a regular expression.

{deny | permit} {community-number | local-AS | no-advertise | no-export | quote-regexp regular-expression-list | regexp regular-expression}

Use these commands in the following sequence, starting in the CONFIGURATION mode to configure an IP extended community list.
Step 1 Command Syntax
ip extcommunity-list
extcommunity-list-name

Command Mode CONFIGURATION CONFIG-COMMUNITYLIST

Purpose Create a extended community list and enter the EXTCOMMUNITY-LIST mode. Two types of extended communities are supported. Filter routes based on the type of extended communities they carry using one of the following keywords: rt: Route Target soo: Route Origin or Site-of-Origin. Support for matching extended communities against regular expression is also supported. Match against a regular expression using the following keyword: regexp: regular expression

{permit | deny} {{rt | soo} {ASN:NN | IPADDR:N} | regex REGEX-LINE}

To set or modify an extended community attribute, use the set extcommunity {rt | soo} {ASN:NN | IPADDR:NN} command. To view the configuration, use the show config command in the CONFIGURATION COMMUNITY-LIST or CONFIGURATION EXTCOMMUNITY LIST mode or the show ip {community-lists | extcommunity-list} command in EXEC Privilege mode (Figure 7-28).

162

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-28. Command example: show ip community-lists

FTOS#show ip community-lists ip community-list standard 1 deny 701:20 deny 702:20 deny 703:20 deny 704:20 deny 705:20 deny 14551:20 deny 701:112 deny 702:112 deny 703:112 deny 704:112 deny 705:112 deny 14551:112 deny 701:667 deny 702:667 deny 703:667

Use these commands in the following sequence, starting in the CONFIGURATION mode, To use an IP Community list or Extended Community List to filter routes, you must apply a match community filter to a route map and then apply that route map to a BGP neighbor or peer group.
Step 1 Command Syntax
route-map map-name [permit | deny] [sequence-number] match {community community-list-name [exact] | extcommunity extcommunity-list-name [exact]} exit router bgp as-number

Command Mode CONFIGURATION CONFIG-ROUTE-MAP

Purpose Enter the ROUTE-MAP mode and assign a name to a route map. Configure a match filter for all routes meeting the criteria in the IP Community or Extended Community list. Return to the CONFIGURATION mode. Enter the ROUTER BGP mode. AS-number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte) or 0.1-65535.65535 (Dotted format) Apply the route map to the neighbor or peer groups incoming or outgoing routes.

3 4

CONFIG-ROUTE-MAP CONFIGURATION

neighbor {ip-address | peer-group-name} route-map map-name {in | out}

CONFIG-ROUTER-BGP

To view the BGP configuration, use the show config command in CONFIGURATION ROUTER BGP mode. To view a route map configuration, use the show route-map command in EXEC Privilege mode. To view which BGP routes meet an IP Community or Extended Community lists criteria, use the show ip
bgp {community-list | extcommunity-list} command in EXEC Privilege mode.

Border Gateway Protocol IPv4 (BGPv4) | 163

www.dell.com | support.dell.com

Manipulate the COMMUNITY attribute

In addition to permitting or denying routes based on the values of the COMMUNITY attributes, you can manipulate the COMMUNITY attribute value and send the COMMUNITY attribute with the route information. By default, FTOS does not send the COMMUNITY attribute. Use the following command in the CONFIGURATION ROUTER BGP mode to send the COMMUNITY attribute to BGP neighbors.
Command Syntax
neighbor {ip-address | peer-group-name} send-community

Command Mode CONFIG-ROUTERBGP

Purpose Enable the software to send the routers COMMUNITY attribute to the BGP neighbor or peer group specified.

To view the BGP configuration, use the show config command in the CONFIGURATION ROUTER BGP mode. If you want to remove or add a specific COMMUNITY number from a BGP path, you must create a route map with one or both of the following statements in the route map. Then apply that route map to a BGP neighbor or peer group. Use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
route-map map-name [permit | deny] [sequence-number] set comm-list
community-list-name delete

Command Mode CONFIGURATION

Purpose Enter the ROUTE-MAP mode and assign a name to a route map. Configure a set filter to delete all COMMUNITY numbers in the IP Community list. Configure a Community list by denying or permitting specific community numbers or types of community community-number: use AA:NN format where AA is the AS number (2 or 4 Bytes) and NN is a value specific to that autonomous system. local-AS: routes with the COMMUNITY attribute of NO_EXPORT_SUBCONFED and are not sent to EBGP peers. no-advertise: routes with the COMMUNITY attribute of NO_ADVERTISE and are not advertised. no-export: routes with the COMMUNITY attribute of NO_EXPORT. none: remove the COMMUNITY attribute. additive: add the communities to already existing communities.

CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

set community {community-number | local-as | no-advertise | no-export | none}

164

Border Gateway Protocol IPv4 (BGPv4)

Step

Command Syntax
exit router bgp as-number neighbor {ip-address | peer-group-name} route-map map-name {in | out}

Command Mode CONFIG-ROUTE-MAP CONFIGURATION CONFIG-ROUTER-BGP

Purpose Return to the CONFIGURATION mode. Enter the ROUTER BGP mode. Apply the route map to the neighbor or peer groups incoming or outgoing routes.

3 4 5

To view the BGP configuration, use the show config command in CONFIGURATION ROUTER BGP mode. To view a route map configuration, use the show route-map command in EXEC Privilege mode. Use the show ip bgp community command in EXEC Privilege mode (Figure 7-29) to view BGP routes matching a certain community number or pre-defined BGP community.
Figure 7-29. Command example: show ip bgp community (Partial)
FTOS>show ip bgp community BGP table version is 3762622, local router ID is 10.114.8.48 Status codes: s suppressed, d damped, h history, * valid, > best, i - internal Origin codes: i - IGP, e - EGP, ? - incomplete Network 3.0.0.0/8 4.2.49.12/30 4.21.132.0/23 4.24.118.16/30 4.24.145.0/30 4.24.187.12/30 4.24.202.0/30 4.25.88.0/30 6.1.0.0/16 6.2.0.0/22 6.3.0.0/18 6.4.0.0/16 6.5.0.0/19 6.8.0.0/20 6.9.0.0/20 6.10.0.0/15 Next Hop 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 195.171.0.16 Metric LocPrf 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Weight 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Path 209 701 80 i 209 i 209 6461 16422 i 209 i 209 i 209 i 209 i 209 3561 3908 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i 209 7170 1455 i

* i *>i * i *>i *>i *>i *>i *>i *>i *>i *>i *>i *>i *>i *>i *>i

Change MED attribute

By default, FTOS uses the MULTI_EXIT_DISC or MED attribute when comparing EBGP paths from the same AS.

Border Gateway Protocol IPv4 (BGPv4) | 165

www.dell.com | support.dell.com

Use any or all of the following commands in the CONFIGURATION ROUTER BGP mode to change how the MED attribute is used.
Command Syntax
bgp always-compare-med

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Enable MED comparison in the paths from neighbors with different ASs. By default, this comparison is not performed. Change the bestpath MED selection to one of the following: confed: Chooses the bestpath MED comparison of paths learned from BGP confederations. missing-as-best: Treat a path missing an MED as the most preferred one

bgp bestpath med {confed | missing-as-best}

Use the show config command in the CONFIGURATION ROUTER BGP mode to view the nondefault values.

Change LOCAL_PREFERENCE attribute

In FTOS, you can change the value of the LOCAL_PREFERENCE attribute. Use the following command in the CONFIGURATION ROUTER BGP mode to change the default values of this attribute for all routes received by the router.
Command Syntax
bgp default local-preference value

Command Mode CONFIG-ROUTERBGP

Purpose Change the LOCAL_PREF value. value range: 0 to 4294967295 Default is 100.

Use the show config command in CONFIGURATION ROUTER BGP mode or the show running-config bgp command in EXEC Privilege mode to view BGP configuration. A more flexible method for manipulating the LOCAL_PREF attribute value is to use a route map. Use these commands in the following sequence, starting CONFIGURATION mode to change the default value of the LOCAL_PREF attribute for specific routes.
Step 1 Command Syntax
route-map map-name [permit | deny] [sequence-number] set local-preference value exit

Command Mode CONFIGURATION CONFIG-ROUTE-MAP CONFIG-ROUTE-MAP

Purpose Enter the ROUTE-MAP mode and assign a name to a route map. Change LOCAL_PREF value for routes meeting the criteria of this route map. Return to the CONFIGURATION mode.

2 3

166

Border Gateway Protocol IPv4 (BGPv4)

Step

Command Syntax
router bgp as-number neighbor {ip-address | peer-group-name} route-map map-name {in | out}

Command Mode CONFIGURATION CONFIG-ROUTER-BGP

Purpose Enter the ROUTER BGP mode. Apply the route map to the neighbor or peer groups incoming or outgoing routes.

4 5

To view the BGP configuration, use the show config command in the CONFIGURATION ROUTER BGP mode. To view a route map configuration, use the show route-map command in EXEC Privilege mode.

Change NEXT_HOP attribute

You can change how the NEXT_HOP attribute is used. Use the following command in the CONFIGURATION ROUTER BGP mode to change the how the NEXT_HOP attribute is used.
Command Syntax
neighbor {ip-address | peer-group-name} next-hop-self

Command Mode CONFIG-ROUTER-B GP

Purpose Disable next hop processing and configure the router as the next hop for a BGP neighbor.

Use the show config command in CONFIGURATION ROUTER BGP mode or the show running-config bgp command in EXEC Privilege mode to view BGP configuration. You can also use route maps to change this and other BGP attributes. For example, you can include the following command in a route map to specify the next hop address:
Command Syntax
set next-hop ip-address

Command Mode CONFIG-ROUTE-M AP

Purpose Sets the next hop address.

Change WEIGHT attribute

Use the following command in CONFIGURATION ROUTER BGP mode to change the how the WEIGHT attribute is used.
Command Syntax
neighbor {ip-address | peer-group-name} weight weight

Command Mode CONFIG-ROUTERBGP

Purpose Assign a weight to the neighbor connection. weight range: 0 to 65535 Default is 0

Use the show config command in CONFIGURATION ROUTER BGP mode or the show running-config bgp command in EXEC Privilege mode to view BGP configuration.

Border Gateway Protocol IPv4 (BGPv4) | 167

www.dell.com | support.dell.com

You can also use route maps to change this and other BGP attributes. For example, you can include the following command in a route map to specify the next hop address:
Command Syntax
set weight weight

Command Mode CONFIG-ROUTE-MAP

Purpose Sets weight for the route. weight range: 0 to 65535

Enable multipath
By default, the software allows one path to a destination. You can enable multipath to allow up to 16 parallel paths to a destination. Use the following command in the CONFIGURATION ROUTER BGP mode to allow more than one path.
Command Syntax
maximum-paths {ebgp | ibgp} number

Command Mode CONFIG-ROUTERBGP

Purpose Enable multiple parallel paths. number range: 1 to 16 Default is 1

The show ip bgp network command includes multipath information for that network.

Filter BGP routes

Filtering routes allows you to implement BGP policies. You can use either IP prefix lists, route maps, AS-PATH ACLs or IP Community lists (via a route map) to control which routes are accepted and advertised by the BGP neighbor or peer group. Prefix lists filter routes based on route and prefix length, while AS-Path ACLs filter routes based on the Autonomous System number. Route maps can filter and set conditions, change attributes, and assign update policies.
Note: FTOS supports up to 255 characters in a set community statement inside a route map.

Note: With FTOS, you can create inbound and outbound policies. Each of the commands used for filtering, has in and out parameters that must be applied. In FTOS, the order of preference varies depending on whether the attributes are applied for inbound updates or outbound updates.

For inbound and outbound updates the order of preference is: prefix lists (using neighbor distribute-list command) AS-PATH ACLs (using neighbor filter-list command) route maps (using neighbor route-map command)

Prior to filtering BGP routes, you must create the prefix list, AS-PATH ACL, or route map to be used.

168

Border Gateway Protocol IPv4 (BGPv4)

Refer to Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps, on page 85 for configuration information on prefix lists, AS-PATH ACLs, and route maps.
Note: When you configure a new set of BGP policies, always reset the neighbor or peer group by entering the clear ip bgp command in EXEC Privilege mode.

Use these commands in the following sequence, starting in the CONFIGURATION mode to filter routes using prefix lists.
Step 1 Command Syntax
ip prefix-list prefix-name seq sequence-number {deny | permit} {any | ip-prefix [ge | le] }

Command Mode CONFIGURATION CONFIG-PREFIX LIST

Purpose Create a prefix list and assign it a name. Create multiple prefix list filters with a deny or permit action. ge: Minimum prefix length to be matched le: maximum prefix length to me matched Refer to Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps, on page 85 for information on configuring prefix lists. Return to the CONFIGURATION mode. Enter ROUTER BGP mode. Filter routes based on the criteria in the configured prefix list. Configure the following parameters: ip-address or peer-group-name: enter the neighbors IP address or the peer groups name. prefix-list-name: enter the name of a configured prefix list. in: apply the prefix list to inbound routes. out: apply the prefix list to outbound routes.

3 4 5

exit router bgp as-number neighbor {ip-address | peer-group-name} distribute-list prefix-list-name {in | out}

CONFIG-PREFIX LIST CONFIGURATION CONFIG-ROUTERBGP

As a reminder, below are some rules concerning prefix lists: If the prefix list contains no filters, all routes are permitted. If none of the routes match any of the filters in the prefix list, the route is denied. This action is called an implicit deny. (If you want to forward all routes that do not match the prefix list criteria, you must configure a prefix list filter to permit all routes. For example, you could have the following filter as the last filter in your prefix list permit 0.0.0.0/0 le 32). Once a route matches a filter, the filters action is applied. No additional filters are applied to the route.

To view the BGP configuration, use the show config command in the ROUTER BGP mode. To view a prefix list configuration, use the show ip prefix-list detail or show ip prefix-list summary commands in EXEC Privilege mode.

Border Gateway Protocol IPv4 (BGPv4) | 169

www.dell.com | support.dell.com

Use these commands in the following sequence, starting in the CONFIGURATION mode to filter routes using a route map.
Step 1 Command Syntax
route-map map-name [permit | deny] [sequence-number]

Command Mode CONFIGURATION CONFIG-ROUTE-MAP

Purpose Create a route map and assign it a name. Create multiple route map filters with a match or set action. Refer to Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps, on page 85 for information on configuring route maps. Return to the CONFIGURATION mode. Enter ROUTER BGP mode. Filter routes based on the criteria in the configured route map. Configure the following parameters: ip-address or peer-group-name: enter the neighbors IP address or the peer groups name. map-name: enter the name of a configured route map. in: apply the route map to inbound routes. out: apply the route map to outbound routes.

{match | set}

3 4

exit router bgp as-number neighbor {ip-address | peer-group-name} route-map map-name {in | out}

CONFIG-ROUTE-MAP CONFIGURATION CONFIG-ROUTER-BGP

Use the show config command in CONFIGURATION ROUTER BGP mode to view the BGP configuration. Use the show route-map command in EXEC Privilege mode to view a route map configuration. Use these commands in the following sequence, beginning in the CONFIGURATION mode to filter routes based on AS-PATH information.
Step 1 Command Syntax
ip as-path access-list
as-path-name

Command Mode CONFIGURATION AS-PATH ACL AS-PATH ACL CONFIGURATION

Purpose Create a AS-PATH ACL and assign it a name. Create a AS-PATH ACL filter with a deny or permit action. Return to the CONFIGURATION mode. Enter ROUTER BGP mode.

2 3 4

{deny | permit}
as-regular-expression

exit router bgp as-number

170

Border Gateway Protocol IPv4 (BGPv4)

Step

Command Syntax
neighbor {ip-address | peer-group-name} filter-list as-path-name {in | out}

Command Mode CONFIG-ROUTER-B GP

Purpose Filter routes based on the criteria in the configured route map. Configure the following parameters: ip-address or peer-group-name: enter the neighbors IP address or the peer groups name. as-path-name: enter the name of a configured AS-PATH ACL. in: apply the AS-PATH ACL map to inbound routes. out: apply the AS-PATH ACL to outbound routes.

Use the show config command in CONFIGURATION ROUTER BGP mode and show ip as-path-access-list command in EXEC Privilege mode to view which commands are configured. Include this filter permit .* in your AS-PATH ACL to forward all routes not meeting the AS-PATH ACL criteria.

Configure BGP route reflectors

BGP route reflectors are intended for Autonomous Systems with a large mesh and they reduce the amount of BGP control traffic. With route reflection configured properly, IBGP routers are not fully meshed within a cluster but all receive routing information. Configure clusters of routers where one router is a concentration router and others are clients who receive their updates from the concentration router. Use the following commands in the CONFIGURATION ROUTER BGP mode to configure a route reflector.
Command Syntax
bgp cluster-id cluster-id neighbor {ip-address | peer-group-name} route-reflector-client

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Assign an ID to a router reflector cluster. You can have multiple clusters in an AS. Configure the local router as a route reflector and the neighbor or peer group identified is the route reflector client.

To view a route reflector configuration, use the show config command in the CONFIGURATION ROUTER BGP mode or show running-config bgp in EXEC Privilege mode. When you enable a route reflector, FTOS automatically enables route reflection to all clients. To disable route reflection between all clients in this reflector, use the no bgp client-to-client reflection command in CONFIGURATION ROUTER BGP mode. All clients should be fully meshed before you disable route reflection.

Border Gateway Protocol IPv4 (BGPv4) | 171

www.dell.com | support.dell.com

Aggregate routes
FTOS provides multiple ways to aggregate routes in the BGP routing table. At least one specific route of the aggregate must be in the routing table for the configured aggregate to become active. Use the following command in the CONFIGURATION ROUTER BGP mode to aggregate routes.
Command Syntax
aggregate-address ip-address mask [advertise-map map-name] [as-set] [attribute-map map-name] [summary-only] [suppress-map map-name]

Command Mode CONFIG-ROUTERBGP

Purpose Assign the IP address and mask of the prefix to be aggregated. Optional parameters are: advertise-map map-name: set filters for advertising an aggregate route as-set: generate path attribute information and include it in the aggregate. attribute-map map-name: modify attributes of the aggregate, except for the AS_PATH and NEXT_HOP attributes summary-only: advertise only the aggregate address. Specific routes will not be advertised suppress-map map-name: identify which more-specific routes in the aggregate are suppressed

AS_SET includes AS_PATH and community information from the routes included in the aggregated route. In the show ip bgp command, aggregates contain an a in the first column and routes suppressed by the aggregate contain an s in the first column.
Figure 7-30. Command Example: show ip bgp
FTOS#show ip bgp BGP table version is 0, local router ID is 10.101.15.13 Status codes: s suppressed, d damped, h history, * valid, > best Path source: I - internal, a - aggregate, c - confed-external, r - redistributed, n - network Origin codes: i - IGP, e - EGP, ? - incomplete Network *> 7.0.0.0/29 *> 7.0.0.0/30 *>a 9.0.0.0/8 *> 9.2.0.0/16 *> 9.141.128.0/24 FTOS# Next Hop 10.114.8.33 10.114.8.33 192.0.0.0 10.114.8.33 10.114.8.33 Metric 0 0 LocPrf Weight 0 0 32768 Route 0 0 Path 18508 18508 18508 18508 18508

Aggregate Indicators

? ? 701 {7018 2686 3786} ? 701 i 701 7018 2686 ?

Configure BGP confederations

Another way to organize routers within an AS and reduce the mesh for IBGP peers is to configure BGP confederations. As with route reflectors, BGP confederations are recommended only for IBGP peering involving a large number of IBGP peering sessions per router. Basically, when you configure BGP confederations, you break the AS into smaller sub-AS, and to those outside your network, the confederations appear as one AS. Within the confederation sub-AS, the IBGP neighbors are fully meshed and the MED, NEXT_HOP, and LOCAL_PREF attributes are maintained between confederations.

172

Border Gateway Protocol IPv4 (BGPv4)

Use the following commands in the CONFIGURATION ROUTER BGP mode to configure BGP confederations.
Command Syntax
bgp confederation identifier
as-number

Command Mode CONFIG-ROUTERBGP CONFIG-ROUTERBGP

Purpose Specifies the confederation ID. AS-number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte) Specifies which confederation sub-AS are peers. AS-number: 0-65535 (2-Byte) or 1-4294967295 (4-Byte)

bgp confederation peers as-number [... as-number]

All Confederation routers must be either 4-Byte or 2-Byte. You cannot have a mix of router ASN support,

Use the show config command in the CONFIGURATION ROUTER BGP mode to view the configuration.

Enable route flap dampening

When EBGP routes become unavailable, they flap and the router issues both WITHDRAWN and UPDATE notices. A flap is when a route is withdrawn is readvertised after being withdrawn has an attribute change

The constant router reaction to the WITHDRAWN and UPDATE notices causes instability in the BGP process. To minimize this instability, you may configure penalties, a numeric value, for routes that flap. When that penalty value reaches a configured limit, the route is not advertised, even if the route is up. In FTOS, that penalty value is 1024. As time passes and the route does not flap, the penalty value decrements or is decayed. However, if the route flaps again, it is assigned another penalty. The penalty value is cumulative and penalty is added under following cases: Withdraw Readvertise Attribute change

When dampening is applied to a route, its path is described by one of the following terms: history entryan entry that stores information on a downed route dampened patha path that is no longer advertised penalized patha path that is assigned a penalty

The CLI example below shows configuring values to start reusing or restarting a route, as well as their default values.

Border Gateway Protocol IPv4 (BGPv4) | 173

www.dell.com | support.dell.com

Figure 7-31. Setting Reuse and Restart Route Values

FTOS(conf-router_bgp)#bgp dampening ? Set time before <1-45> Half-life time for the penalty (default = 15) route-map Route-map to specify criteria for dampening value decrements <cr> FTOS(conf-router_bgp)#bgp dampening 2 ? Set readvertise value <1-20000> Value to start reusing a route (default = 750) FTOS(conf-router_bgp)#bgp dampening 2 2000 ? Set suppress value <1-20000> Value to start suppressing a route (default = 2000) Set time to suppress FTOS(conf-router_bgp)#bgp dampening 2 2000 3000 ? a route <1-255> Maximum duration to suppress a stable route (default = 60) FTOS(conf-router_bgp)#bgp dampening 2 2000 3000 10 ? route-map Route-map to specify criteria for dampening <cr>

Use the following command in the CONFIGURATION ROUTER BGP mode to configure route flap dampening parameters.
Command Syntax
bgp dampening [half-life | reuse | suppress max-suppress-time] [route-map map-name]

Command Mode CONFIG-ROUTER-B GP

Purpose Enable route dampening. Enter the following optional parameters to configure route dampening parameters: half-life range: 1 to 45. Number of minutes after which the Penalty is decreased. After the router assigns a Penalty of 1024 to a route, the Penalty is decreased by half after the half-life period expires. (Default: 15 minutes) reuse range: 1 to 20000. This number is compared to the flapping routes Penalty value. If the Penalty value is less than the reuse value, the flapping route is once again advertised (or no longer suppressed). Withdrawn routes are removed from history state. (Default: 750) suppress range: 1 to 20000. This number is compared to the flapping routes Penalty value. If the Penalty value is greater than the suppress value, the flapping route is no longer advertised (that is, it is suppressed). (Default: 2000.) max-suppress-time range: 1 to 255. The maximum number of minutes a route can be suppressed. The default is four times the half-life value. (Default: 60 minutes.) route-map map-name: name of a configured route map. Only match commands in the configured route map are supported. Use this parameter to apply route dampening to selective routes.

To view the BGP configuration, use show config in the CONFIGURATION ROUTER BGP mode or show running-config bgp in EXEC Privilege mode.

174

Border Gateway Protocol IPv4 (BGPv4)

To set dampening parameters via a route map, use the following command in CONFIGURATION ROUTE-MAP mode:

Command Syntax
set dampening half-life reuse
suppress max-suppress-time

Command Mode CONFIG-ROUTE-MAP

Purpose Enter the following optional parameters to configure route dampening parameters: half-life range: 1 to 45. Number of minutes after which the Penalty is decreased. After the router assigns a Penalty of 1024 to a route, the Penalty is decreased by half after the half-life period expires. (Default: 15 minutes) reuse range: 1 to 20000. This number is compared to the flapping routes Penalty value. If the Penalty value is less than the reuse value, the flapping route is once again advertised (or no longer suppressed). (Default: 750) suppress range: 1 to 20000. This number is compared to the flapping routes Penalty value. If the Penalty value is greater than the suppress value, the flapping route is no longer advertised (that is, it is suppressed). (Default: 2000.) max-suppress-time range: 1 to 255. The maximum number of minutes a route can be suppressed. The default is four times the half-life value. (Default: 60 minutes.)

To view a count of dampened routes, history routes and penalized routes when route dampening is enabled, look at the seventh line of the show ip bgp summary command output (Figure 7-32).
Figure 7-32. Command example: show ip bgp summary
FTOS>show ip bgp summary BGP router identifier 10.114.8.131, local AS number 65515 BGP table version is 855562, main routing table version 780266 122836 network entrie(s) and 221664 paths using 29697640 bytes of memory 34298 BGP path attribute entrie(s) using 1920688 bytes of memory 29577 BGP AS-PATH entrie(s) using 1384403 bytes of memory 184 BGP community entrie(s) using 7616 bytes of memory Dampening enabled. 0 history paths, 0 dampened paths, 0 penalized paths Neighbor 10.114.8.34 10.114.8.33 FTOS> AS 18508 18508 MsgRcvd MsgSent 82883 117265 79977 25069 TblVer 780266 780266 InQ 0 0 OutQ Up/Down 2 00:38:51 20 00:38:50

Dampening Information

State/PfxRcd 118904 102759

To view which routes are dampened (non-active), use the show ip bgp dampened-routes command in EXEC Privilege mode.

Border Gateway Protocol IPv4 (BGPv4) | 175

www.dell.com | support.dell.com

Use the following command in EXEC Privilege mode to clear information on route dampening and return suppressed routes to active state.
Command Syntax
clear ip bgp dampening [ip-address mask]

Command Mode EXEC Privilege

Purpose Clear all information or only information on a specific route.

Use the following command in EXEC and EXEC Privilege mode to view statistics on route flapping.
Command Syntax
show ip bgp flap-statistics [ip-address [mask]] [filter-list as-path-name] [regexp regular-expression]

Command Mode EXEC EXEC Privilege

Purpose View all flap statistics or for specific routes meeting the following criteria: ip-address [mask]: enter the IP address and mask filter-list as-path-name: enter the name of an AS-PATH ACL. regexp regular-expression: enter a regular express to match on.

By default, the path selection in FTOS is deterministic, that is, paths are compared irrespective of the order of their arrival. You can change the path selection method to non-deterministic, that is, paths are compared in the order in which they arrived (starting with the most recent). Furthermore, in non-deterministic mode, the software may not compare MED attributes though the paths are from the same AS. Use the following command in CONFIGURATION ROUTER BGP mode to change the path selection from the default mode (deterministic) to non-deterministic.
Command Syntax
bgp non-deterministic-med

Command Mode CONFIG-ROUTERBGP

Purpose Change the best path selection method to non-deterministic.

Note: When you change the best path selection method, path selection for existing paths remains unchanged until you reset it by entering the clear ip bgp command in EXEC Privilege mode.

176

Border Gateway Protocol IPv4 (BGPv4)

Change BGP timers

Use either or both of the following commands in the CONFIGURATION ROUTER BGP mode to configure BGP timers.
Command Syntax
neighbors {ip-address | peer-group-name} timers keepalive
holdtime

Command Mode CONFIG-ROUTERBGP

Purpose Configure timer values for a BGP neighbor or peer group. keepalive range: 1 to 65535. Time interval, in seconds, between keepalive messages sent to the neighbor routers. (Default: 60 seconds) holdtime range: 3 to 65536. Time interval, in seconds, between the last keepalive message and declaring the router dead. (Default: 180 seconds) Configure timer values for all neighbors. keepalive range: 1 to 65535. Time interval, in seconds, between keepalive messages sent to the neighbor routers. (Default: 60 seconds) holdtime range: 3 to 65536. Time interval, in seconds, between the last keepalive message and declaring the router dead. (Default: 180 seconds)

timers bgp keepalive holdtime

CONFIG-ROUTERBGP

Use the show config command in CONFIGURATION ROUTER BGP mode or the show running-config bgp command in EXEC Privilege mode to view non-default values. Timer values configured with the neighbor timers command override the timer values configured with the timers bgp command. When two neighbors, configured with different keepalive and holdtime values, negotiate for new values, the resulting values will be as follows: the lower of the holdtime values is the new holdtime value, and whichever is the lower value; one-third of the new holdtime value, or the configured keepalive value is the new keepalive value.

BGP neighbor soft-reconfiguration

Changing routing policies typically requires a reset of BGP sessions (the TCP connection) for the policies to take effect. Such resets cause undue interruption to traffic due to hard reset of the BGP cache and the time it takes to re-establish the session. BGP soft reconfig allows for policies to be applied to a session without clearing the BGP Session. Soft-reconfig can be done on a per-neighbor basis and can either be inbound or outbound. BGP Soft Reconfiguration clears the policies without resetting the TCP connection.

Border Gateway Protocol IPv4 (BGPv4) | 177

www.dell.com | support.dell.com

Use the clear ip bgp command in EXEC Privilege mode at the system prompt to reset a BGP connection using BGP soft reconfiguration.
Command Syntax
clear ip bgp {* | neighbor-address | AS Numbers | ipv4 | peer-group-name} [soft [in | out]]

Command Mode EXEC Privilege

Purpose Clear all information or only specific details. *: Clear all peers neighbor-address: Clear the neighbor with this IP address AS Numbers: Peers AS numbers to be cleared ipv4: Clear information for IPv4 Address family peer-group-name: Clear all members of the specified peer group Enable soft-reconfiguration for the BGP neighbor specified. BGP stores all the updates received by the neighbor but does not reset the peer-session.

neighbor {ip-address | peer-group-name} soft-reconfiguration inbound

CONFIG-ROUTER-B GP

Entering this command starts the storage of updates, which is required to do inbound soft reconfiguration. Outbound BGP soft reconfiguration does not require inbound soft reconfiguration to be enabled.

When soft-reconfiguration is enabled for a neighbor and the clear ip bgp soft in command is executed, the update database stored in the router is replayed and updates are reevaluated. With this command, the replay and update process is triggered only if route-refresh request is not negotiated with the peer. If the request is indeed negotiated (upon execution of clear ip bgp soft in), then BGP sends a route-refresh request to the neighbor and receives all of the peers updates. To use soft reconfiguration, or soft reset, without preconfiguration, both BGP peers must support the soft route refresh capability, which is advertised in the open message sent when the peers establish a TCP session. To determine whether a BGP router supports this capability, use the show ip bgp neighbors command. If a router supports the route refresh capability, the following message should be displayed:
Received route refresh capability from peer.

If you specify a BGP peer group by using the peer-group-name argument, all members of the peer group inherit the characteristic configured with this command. The following (Figure 7-33) enables inbound soft reconfiguration for the neighbor 10.108.1.1. All updates received from this neighbor are stored unmodified, regardless of the inbound policy. When inbound soft reconfiguration is done later, the stored information is used to generate a new set of inbound updates.
Figure 7-33. Command example: router bgp
FTOS>router bgp 100 neighbor 10.108.1.1 remote-as 200 neighbor 10.108.1.1 soft-reconfiguration inbound

178

Border Gateway Protocol IPv4 (BGPv4)

Route map continue

The BGP route map continue feature (in ROUTE-MAP mode) allows movement from one route-map entry to a specific route-map entry (the sequence number). If the sequence number is not specified, the continue feature moves to the next sequence number (also known as an implied continue). If a match clause exists, the continue feature executes only after a successful match occurs. If there are no successful matches, continue is ignored.
continue [sequence-number]

Match Clause with a Continue Clause

The continue feature can exist without a match clause. Without a match clause, the continue clause executes and jumps to the specified route-map entry. With a match clause and a continue clause, the match clause executes first and the continue clause next in a specified route map entry. The continue clause launches only after a successful match. The behavior is: A successful match with a continue clausethe route map executes the set clauses and then goes to the specified route map entry upon execution of the continue clause. If the next route map entry contains a continue clause, the route map executes the continue clause if a successful match occurs. If the next route map entry does not contain a continue clause, the route map evaluates normally. If a match does not occur, the route map does not continue and falls-through to the next sequence number, if one exists.

Set Clause with a Continue Clause

If the route-map entry contains sets with the continue clause, then the set actions operation is performed first followed by the continue clause jump to the specified route map entry. If a set actions operation occurs in the first route map entry and then the same set action occurs with a different value in a subsequent route map entry, the last set of actions overrides the previous set of actions with the same set command. If the set community additive and set as-path prepend commands are configured, the communities and AS numbers are prepended.

Border Gateway Protocol IPv4 (BGPv4) | 179

www.dell.com | support.dell.com

MBGP Configuration

et c MBGP for IPv4 Multicast is supported on platform c et s MBGP is not supported on the E-Series ExaScale ex platform.
MBGP for IPv6 unicast is supported on platforms Multiprotocol BGP (MBGP) is an enhanced BGP that carries IP multicast routes. BGP carries two sets of routes: one set for unicast routing and one set for multicast routing. The routes associated with multicast routing are used by the Protocol Independent Multicast (PIM) to build data distribution trees. FTOS MBGP is implemented as per RFC 1858. The MBGP feature can be enabled per router and/or per peer/peer-group. Default is IPv4 Unicast routes.
Command Syntax
address family ipv4 multicast neighbor [ip-address | peer-group-name] activate

Command Mode CONFIG-ROUTER-BGP CONFIG-ROUTER-BGP-AF (Address Family)

Purpose Enables support for the IPv4 Multicast family on the BGP node Enable IPv4 Multicast support on a BGP neighbor/peer group

When a peer is configured to support IPv4 Multicast, FTOS takes the following actions: Send a capacity advertisement to the peer in the BGP Open message specifying IPv4 Multicast as a supported AFI/SAFI (Subsequent Address Family Identifier). If the corresponding capability is received in the peers Open message, BGP will mark the peer as supporting the AFI/SAFI. When exchanging updates with the peer, BGP sends and receives IPv4 Multicast routes if the peer is marked as supporting that AFI/SAFI. Exchange of IPv4 Multicast route information occurs through the use of two new attributes called MP_REACH_NLRI and MP_UNREACH_NLRI, for feasible and withdrawn routes, respectively. If the peer has not been activated in any AFI/SAFI, the peer remains in Idle state.

Most FTOS BGP IPv4 Unicast commands are extended to support the IPv4 Multicast RIB using extra options to the command. See the FTOS Command Line Interface Reference for a detailed description of the MBGP commands.

180

Border Gateway Protocol IPv4 (BGPv4)

BGP Regular Expression Optimization

BGP policies that contain regular expressions to match against as-paths and communities might take a lot of CPU processing time, and thus can affect BGP routing convergence. Also, show bgp commands that get filtered through regular expressions can take a lot of CPU cycles, especially when the database is large. FTOS optimizes processing time when using regular expressions by caching and re-using regular expression evaluated results, at the expense of some memory in RP1 processor. This feature is turned on by default. Use the command bgp regex-eval-optz-disable in CONFIGURATION ROUTER BGP mode

to disable it if necessary.

Debugging BGP
Use any of the commands in EXEC Privilege mode to enable BGP debugging.
Command Syntax
debug ip bgp [ip-address | peer-group peer-group-name] [in | out] debug ip bgp dampening [in | out] debug ip bgp [ip-address | peer-group peer-group-name] events [in | out] debug ip bgp [ip-address | peer-group peer-group-name] keepalive [in | out] debug ip bgp [ip-address | peer-group peer-group-name] notifications [in | out] debug ip bgp [ip-address | peer-group peer-group-name] updates [in | out] [prefix-list name] debug ip bgp {ip-address | peer-group-name} soft-reconfiguration

Command Mode EXEC Privilege EXEC Privilege EXEC Privilege EXEC Privilege EXEC Privilege EXEC Privilege

Purpose View all information on BGP, including BGP events, keepalives, notifications, and updates. View information on BGP route being dampened. View information on local BGP state changes and other BGP events. View information about BGP KEEPALIVE messages. View information about BGP notifications received from or sent to neighbors. View information about BGP updates and filter by prefix name

EXEC Privilege

Enable soft-reconfiguration debug. Enable soft-reconfiguration debug. To enhance debugging of soft reconfig, use the following command only when route-refresh is not negotiated to avoid the peer from resending messages:
bgp soft-reconfig-backup

In-BGP is shown via the show ip protocols command.

Border Gateway Protocol IPv4 (BGPv4) | 181

www.dell.com | support.dell.com

FTOS displays debug messages on the console. To view which debugging commands are enabled, use the show debugging command in EXEC Privilege mode. Use the keyword no followed by the debug command To disable a specific debug command. For example, to disable debugging of BGP updates, enter no debug ip bgp updates command. Use no debug ip bgp to disable all BGP debugging. Use undebug all to disable all debugging.

Storing Last and Bad PDUs

FTOS stores the last notification sent/received, and the last bad PDU received on per peer basis. The last bad PDU is the one that causes a notification to be issued. These PDUs are shown in the output of the command show ip bgp neighbor, as shown in Figure 7-34.

182

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-34. Viewing the Last Bad PDU from BGP Peers
FTOS(conf-router_bgp)#do show ip bgp neighbors 1.1.1.2 BGP neighbor is 1.1.1.2, remote AS 2, external link BGP version 4, remote router ID 2.4.0.1 BGP state ESTABLISHED, in this state for 00:00:01 Last read 00:00:00, last write 00:00:01 Hold time is 90, keepalive interval is 30 seconds Received 1404 messages, 0 in queue 3 opens, 1 notifications, 1394 updates 6 keepalives, 0 route refresh requests Sent 48 messages, 0 in queue 3 opens, 2 notifications, 0 updates 43 keepalives, 0 route refresh requests Minimum time between advertisement runs is 30 seconds Minimum time before advertisements start is 0 seconds Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) For address family: IPv4 Unicast BGP table version 1395, neighbor version 1394 Prefixes accepted 1 (consume 4 bytes), 0 withdrawn by peer Prefixes advertised 0, rejected 0, 0 withdrawn from peer Connections established 3; dropped 2 Last reset 00:00:12, due to Missing well known attribute Notification History 'UPDATE error/Missing well-known attr' Sent : 1 'Connection Reset' Sent : 1 Recv: 0

Recv: 0

Last notification (len 21) sent 00:26:02 ago Last PDUs ffffffff ffffffff ffffffff ffffffff 00160303 03010000 Last notification (len 21) received 00:26:20 ago ffffffff ffffffff ffffffff ffffffff 00150306 00000000 Last PDU (len 41) received 00:26:02 ago that caused notification to be issued ffffffff ffffffff ffffffff ffffffff 00290200 00000e01 02040201 00024003 04141414 0218c0a8 01000000 Local host: 1.1.1.1, Local port: 179 Foreign host: 1.1.1.2, Foreign port: 41758

Capturing PDUs
Capture incoming and outgoing PDUs on a per-peer basis using the command capture bgp-pdu neighbor direction. Disable capturing using the no form of this command.

Border Gateway Protocol IPv4 (BGPv4) | 183

www.dell.com | support.dell.com

The buffer size supports a maximum value between 40 MB (the default) and 100 MB. The capture buffers are cyclic and reaching the limit prompts the system to overwrite the oldest PDUs when new ones are received for a given neighbor or direction. Setting the buffer size to a value lower than the current max, might cause captured PDUs to be freed to set the new limit.
Note: Memory on RP1 is not pre-allocated, and is allocated only when a PDU needs to be captured.

Use the command capture bgp-pdu max-buffer-size (Figure 7-35) to change the maximum buffer size. View the captured PDUs using the command show capture bgp-pdu neighbor.
Figure 7-35. Viewing Captured PDUs
FTOS#show capture bgp-pdu neighbor 20.20.20.2 Incoming packet capture enabled for BGP neighbor 20.20.20.2 Available buffer size 40958758, 26 packet(s) captured using 680 bytes PDU[1] : len 101, captured 00:34:51 ago ffffffff ffffffff ffffffff ffffffff 00650100 00000013 00000000 00000000 419ef06c 00000000 00000000 00000000 00000000 00000000 0181a1e4 0181a25c 41af92c0 00000000 00000000 00000000 00000000 00000001 0181a1e4 0181a25c 41af9400 00000000 PDU[2] : len 19, captured 00:34:51 ago ffffffff ffffffff ffffffff ffffffff 00130400 PDU[3] : len 19, captured 00:34:51 ago ffffffff ffffffff ffffffff ffffffff 00130400 PDU[4] : len 19, captured 00:34:22 ago ffffffff ffffffff ffffffff ffffffff 00130400 [. . .] Outgoing packet capture enabled for BGP neighbor 20.20.20.2 Available buffer size 40958758, 27 packet(s) captured using 562 bytes PDU[1] : len 41, captured 00:34:52 ago ffffffff ffffffff ffffffff ffffffff 00290104 000100b4 14141401 0c020a01 04000100 01020080 00000000 PDU[2] : len 19, captured 00:34:51 ago ffffffff ffffffff ffffffff ffffffff 00130400 PDU[3] : len 19, captured 00:34:50 ago ffffffff ffffffff ffffffff ffffffff 00130400 PDU[4] : len 19, captured 00:34:20 ago ffffffff ffffffff ffffffff ffffffff 00130400 [. . .]

The buffers storing the PDU free memory when: BGP is disabled A neighbor is unconfigured clear ip bgp is issued New PDU are captured and there is no more space to store them The max buffer size is reduced. (This may cause PDUs to be cleared depending upon the buffer space consumed and the new limit.)

184

Border Gateway Protocol IPv4 (BGPv4)

With full internet feed (205K) captured, approximately 11.8MB is required to store all of the PDUs, as shown in Figure 7-36.
Figure 7-36. Required Memory for Captured PDUs
FTOS(conf-router_bgp)#do show capture bgp-pdu neighbor 172.30.1.250 Incoming packet capture enabled for BGP neighbor 172.30.1.250 Available buffer size 29165743, 192991 packet(s) captured using 11794257 bytes [. . .] FTOS(conf-router_bgp)#do sho ip bg s BGP router identifier 172.30.1.56, local AS number 65056 BGP table version is 313511, main routing table version 313511 207896 network entrie(s) and 207896 paths using 42364576 bytes of memory 59913 BGP path attribute entrie(s) using 2875872 bytes of memory 59910 BGP AS-PATH entrie(s) using 2679698 bytes of memory 3 BGP community entrie(s) using 81 bytes of memory Neighbor 1.1.1.2 172.30.1.250 AS 2 18508 MsgRcvd 17 243295 MsgSent 18966 25 TblVer 0 313511 InQ 0 0 OutQ Up/Down State/Pfx

0 00:08:19 Active 0 00:12:46 207896

PDU Counters
FTOS version 7.5.1.0 introduces additional counters for various types of PDUs sent and received from neighbors. These are seen in the output of the command show ip bgp neighbor.

Sample Configurations
The following configurations are examples for enabling BGP and setting up some peer groups. These are not comprehensive directions. They are intended to give you a some guidance with typical configurations. You can copy and paste from these examples to your CLI. Be sure you make the necessary changes to support your own IP Addresses, Interfaces, Names, etc. Figure 7-37 is a graphic illustration of the configurations shown on the following pages. These configurations show how to create BGP areas using physical and virtual links. They include setting up the interfaces and peers groups with each other.

Border Gateway Protocol IPv4 (BGPv4) | 185

www.dell.com | support.dell.com

Figure 7-37. Sample Configuration Illustration

AS 99
GigE 1/21 10.0.1.21 /24

Physical Links

Virtual Links Peer Group AAA GigE 2/11 10.0.1.22 /24 Loopback 1 Lo 192.168.128.2 /24 19

Loopback ck 1 192.168.128.1 /24 GigE 1/31 10.0.3.31 /24

er Pe

o Gr

C CC p u

GigE 3/11 10.0.3.33 /24 Loopback 1 192.168.128.3 /24

e Pe rG u ro p BB B

GigE 2/31 10.0.2.2 /24

GigE 3/21 10.0.2.3 /24

AS 100

186

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-38. Enable BGP - Router 1

R1# conf R1(conf)#int loop 0 R1(conf-if-lo-0)#ip address 192.168.128.1/24 R1(conf-if-lo-0)#no shutdown R1(conf-if-lo-0)#show config ! interface Loopback 0 ip address 192.168.128.1/24 no shutdown R1(conf-if-lo-0)#int gig 1/21 R1(conf-if-gi-1/21)#ip address 10.0.1.21/24 R1(conf-if-gi-1/21)#no shutdown R1(conf-if-gi-1/21)#show config ! interface GigabitEthernet 1/21 ip address 10.0.1.21/24 no shutdown R1(conf-if-gi-1/21)#int gig 1/31 R1(conf-if-gi-1/31)#ip address 10.0.3.31/24 R1(conf-if-gi-1/31)#no shutdown R1(conf-if-gi-1/31)#show config ! interface GigabitEthernet 1/31 ip address 10.0.3.31/24 no shutdown R1(conf-if-gi-1/31)#router bgp 99 R1(conf-router_bgp)#network 192.168.128.0/24 R1(conf-router_bgp)#neighbor 192.168.128.2 remote 99 R1(conf-router_bgp)#neighbor 192.168.128.2 no shut R1(conf-router_bgp)#neighbor 192.168.128.2 update-source loop 0 R1(conf-router_bgp)#neighbor 192.168.128.3 remote 100 R1(conf-router_bgp)#neighbor 192.168.128.3 no shut R1(conf-router_bgp)#neighbor 192.168.128.3 update-source loop 0 R1(conf-router_bgp)#show config ! router bgp 99 network 192.168.128.0/24 neighbor 192.168.128.2 remote-as 99 neighbor 192.168.128.2 update-source Loopback 0 neighbor 192.168.128.2 no shutdown neighbor 192.168.128.3 remote-as 100 neighbor 192.168.128.3 update-source Loopback 0 neighbor 192.168.128.3 no shutdown R1(conf-router_bgp)#end R1# R1#show ip bgp summary BGP router identifier 192.168.128.1, local AS number 99 BGP table version is 4, main routing table version 4 4 network entrie(s) using 648 bytes of memory 6 paths using 408 bytes of memory BGP-RIB over all using 414 bytes of memory 3 BGP path attribute entrie(s) using 144 bytes of memory 2 BGP AS-PATH entrie(s) using 74 bytes of memory 2 neighbor(s) using 8672 bytes of memory Neighbor 192.168.128.2 192.168.128.3 R1# AS 99 100 MsgRcvd 4 5 MsgSent 5 4 TblVer 4 1 InQ 0 0 OutQ Up/Down 0 00:00:32 0 00:00:09 State/Pfx 1 4

Border Gateway Protocol IPv4 (BGPv4) | 187

www.dell.com | support.dell.com

Figure 7-39. Enable BGP - Router 2

R2# conf R2(conf)#int loop 0 R2(conf-if-lo-0)#ip address 192.168.128.2/24 R2(conf-if-lo-0)#no shutdown R2(conf-if-lo-0)#show config ! interface Loopback 0 ip address 192.168.128.2/24 no shutdown R2(conf-if-lo-0)#int gig 2/11 R2(conf-if-gi-2/11)#ip address 10.0.1.22/24 R2(conf-if-gi-2/11)#no shutdown R2(conf-if-gi-2/11)#show config ! interface GigabitEthernet 2/11 ip address 10.0.1.22/24 no shutdown R2(conf-if-gi-2/11)#int gig 2/31 R2(conf-if-gi-2/31)#ip address 10.0.2.2/24 R2(conf-if-gi-2/31)#no shutdown R2(conf-if-gi-2/31)#show config ! interface GigabitEthernet 2/31 ip address 10.0.2.2/24 no shutdown R2(conf-if-gi-2/31)# R2(conf-if-gi-2/31)#router bgp 99 R2(conf-router_bgp)#network 192.168.128.0/24 R2(conf-router_bgp)#neighbor 192.168.128.1 remote 99 R2(conf-router_bgp)#neighbor 192.168.128.1 no shut R2(conf-router_bgp)#neighbor 192.168.128.1 update-source loop 0 R2(conf-router_bgp)#neighbor 192.168.128.3 remote 100 R2(conf-router_bgp)#neighbor 192.168.128.3 no shut R2(conf-router_bgp)#neighbor 192.168.128.3 update loop 0 R2(conf-router_bgp)#show config ! router bgp 99 bgp router-id 192.168.128.2 network 192.168.128.0/24 bgp graceful-restart neighbor 192.168.128.1 remote-as 99 neighbor 192.168.128.1 update-source Loopback 0 neighbor 192.168.128.1 no shutdown neighbor 192.168.128.3 remote-as 100 neighbor 192.168.128.3 update-source Loopback 0 neighbor 192.168.128.3 no shutdown R2(conf-router_bgp)#end R2#show ip bgp summary BGP router identifier 192.168.128.2, local AS number 99 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 3 paths using 204 bytes of memory BGP-RIB over all using 207 bytes of memory 2 BGP path attribute entrie(s) using 128 bytes of memory 2 BGP AS-PATH entrie(s) using 90 bytes of memory 2 neighbor(s) using 9216 bytes of memory Neighbor 192.168.128.1 192.168.128.3 R2# AS 99 100 MsgRcvd 40 4 MsgSent 35 4 TblVer 1 1 InQ 0 0 OutQ Up/Down State/Pfx 0 00:01:05 1 0 00:00:16 1

188

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-40. Enable BGP - Router 3

R3# conf R3(conf)# R3(conf)#int loop 0 R3(conf-if-lo-0)#ip address 192.168.128.3/24 R3(conf-if-lo-0)#no shutdown R3(conf-if-lo-0)#show config ! interface Loopback 0 ip address 192.168.128.3/24 no shutdown R3(conf-if-lo-0)#int gig 3/11 R3(conf-if-gi-3/11)#ip address 10.0.3.33/24 R3(conf-if-gi-3/11)#no shutdown R3(conf-if-gi-3/11)#show config ! interface GigabitEthernet 3/11 ip address 10.0.3.33/24 no shutdown R3(conf-if-lo-0)#int gig 3/21 R3(conf-if-gi-3/21)#ip address 10.0.2.3/24 R3(conf-if-gi-3/21)#no shutdown R3(conf-if-gi-3/21)#show config ! interface GigabitEthernet 3/21 ip address 10.0.2.3/24 no shutdown R3(conf-if-gi-3/21)# R3(conf-if-gi-3/21)#router bgp 100 R3(conf-router_bgp)#show config ! router bgp 100 R3(conf-router_bgp)#network 192.168.128.0/24 R3(conf-router_bgp)#neighbor 192.168.128.1 remote 99 R3(conf-router_bgp)#neighbor 192.168.128.1 no shut R3(conf-router_bgp)#neighbor 192.168.128.1 update-source loop 0 R3(conf-router_bgp)#neighbor 192.168.128.2 remote 99 R3(conf-router_bgp)#neighbor 192.168.128.2 no shut R3(conf-router_bgp)#neighbor 192.168.128.2 update loop 0 R3(conf-router_bgp)#show config ! router bgp 100 network 192.168.128.0/24 neighbor 192.168.128.1 remote-as 99 neighbor 192.168.128.1 update-source Loopback 0 neighbor 192.168.128.1 no shutdown neighbor 192.168.128.2 remote-as 99 neighbor 192.168.128.2 update-source Loopback 0 neighbor 192.168.128.2 no shutdown R3(conf)#end R3#show ip bgp summary BGP router identifier 192.168.128.3, local AS number 100 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 3 paths using 204 bytes of memory BGP-RIB over all using 207 bytes of memory 2 BGP path attribute entrie(s) using 128 bytes of memory 2 BGP AS-PATH entrie(s) using 90 bytes of memory 2 neighbor(s) using 9216 bytes of memory Neighbor AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/Pfx 192.168.128.1 99 24 25 1 0 0 00:14:20 1 192.168.128.2 99 14 14 1 0 0 00:10:22 1

Border Gateway Protocol IPv4 (BGPv4) | 189

www.dell.com | support.dell.com

Figure 7-41. Enable Peer Group - Router 1

R1#conf R1(conf)#router bgp 99 R1(conf-router_bgp)# network 192.168.128.0/24 R1(conf-router_bgp)# neighbor AAA peer-group R1(conf-router_bgp)# neighbor AAA no shutdown R1(conf-router_bgp)# neighbor BBB peer-group R1(conf-router_bgp)# neighbor BBB no shutdown R1(conf-router_bgp)# neighbor 192.168.128.2 peer-group AAA R1(conf-router_bgp)# neighbor 192.168.128.3 peer-group BBB R1(conf-router_bgp)# R1(conf-router_bgp)#show config ! router bgp 99 network 192.168.128.0/24 neighbor AAA peer-group neighbor AAA no shutdown neighbor BBB peer-group neighbor BBB no shutdown neighbor 192.168.128.2 remote-as 99 neighbor 192.168.128.2 peer-group AAA neighbor 192.168.128.2 update-source Loopback 0 neighbor 192.168.128.2 no shutdown neighbor 192.168.128.3 remote-as 100 neighbor 192.168.128.3 peer-group BBB neighbor 192.168.128.3 update-source Loopback 0 neighbor 192.168.128.3 no shutdown R1# R1#show ip bgp summary BGP router identifier 192.168.128.1, local AS number 99 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 3 paths using 204 bytes of memory BGP-RIB over all using 207 bytes of memory 2 BGP path attribute entrie(s) using 96 bytes of memory 2 BGP AS-PATH entrie(s) using 74 bytes of memory 2 neighbor(s) using 8672 bytes of memory Neighbor AS MsgRcvd 23 30 MsgSent 24 29 TblVer 1 1 InQ 0 0 OutQ Up/Down (0) 00:00:17 (0) 00:00:14 State/Pfx 1 1

192.168.128.2 99 192.168.128.3 100 ! R1#show ip bgp neighbors

BGP neighbor is 192.168.128.2, remote AS 99, internal link Member of peer-group AAA for session parameters BGP version 4, remote router ID 192.168.128.2 BGP state ESTABLISHED, in this state for 00:00:37 Last read 00:00:36, last write 00:00:36 Hold time is 180, keepalive interval is 60 seconds Received 23 messages, 0 in queue 2 opens, 0 notifications, 2 updates 19 keepalives, 0 route refresh requests Sent 24 messages, 0 in queue 2 opens, 1 notifications, 2 updates 19 keepalives, 0 route refresh requests Minimum time between advertisement runs is 5 seconds Minimum time before advertisements start is 0 seconds

190

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-42. Enable Peer Groups - Router 1 continued

Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 1, neighbor version 1 Prefixes accepted 1 (consume 4 bytes), withdrawn 0 by peer Prefixes advertised 1, denied 0, withdrawn 0 from peer Connections established 2; dropped 1 Last reset 00:00:57, due to user reset Notification History 'Connection Reset' Sent : 1 Recv: 0 Last notification (len 21) sent 00:00:57 ago ffffffff ffffffff ffffffff ffffffff 00150306 00000000 Local host: 192.168.128.1, Local port: 179 Foreign host: 192.168.128.2, Foreign port: 65464 BGP neighbor is 192.168.128.3, remote AS 100, external link Member of peer-group BBB for session parameters BGP version 4, remote router ID 192.168.128.3 BGP state ESTABLISHED, in this state for 00:00:37 Last read 00:00:36, last write 00:00:36 Hold time is 180, keepalive interval is 60 seconds Received 30 messages, 0 in queue 4 opens, 2 notifications, 4 updates 20 keepalives, 0 route refresh requests Sent 29 messages, 0 in queue 4 opens, 1 notifications, 4 updates 20 keepalives, 0 route refresh requests Minimum time between advertisement runs is 30 seconds Minimum time before advertisements start is 0 seconds Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 1, neighbor version 1 Prefixes accepted 1 (consume 4 bytes), withdrawn 0 by peer Prefixes advertised 1, denied 0, withdrawn 0 from peer Connections established 4; dropped 3 Last reset 00:00:54, due to user reset R1#

Border Gateway Protocol IPv4 (BGPv4) | 191

www.dell.com | support.dell.com

Figure 7-43. Enable Peer Groups - Router 2

R2#conf R2(conf)#router bgp 99 R2(conf-router_bgp)# neighbor CCC peer-group R2(conf-router_bgp)# neighbor CCC no shutdown R2(conf-router_bgp)# neighbor BBB peer-group R2(conf-router_bgp)# neighbor BBB no shutdown R2(conf-router_bgp)# neighbor 192.168.128.1 peer AAA R2(conf-router_bgp)# neighbor 192.168.128.1 no shut R2(conf-router_bgp)# neighbor 192.168.128.3 peer BBB R2(conf-router_bgp)# neighbor 192.168.128.3 no shut R2(conf-router_bgp)#show conf ! router bgp 99 network 192.168.128.0/24 neighbor AAA peer-group neighbor AAA no shutdown neighbor BBB peer-group neighbor BBB no shutdown neighbor 192.168.128.1 remote-as 99 neighbor 192.168.128.1 peer-group CCC neighbor 192.168.128.1 update-source Loopback 0 neighbor 192.168.128.1 no shutdown neighbor 192.168.128.3 remote-as 100 neighbor 192.168.128.3 peer-group BBB neighbor 192.168.128.3 update-source Loopback 0 neighbor 192.168.128.3 no shutdown R2(conf-router_bgp)#end R2# R2#show ip bgp summary BGP router identifier 192.168.128.2, local AS number 99 BGP table version is 2, main routing table version 2 1 network entrie(s) using 132 bytes of memory 3 paths using 204 bytes of memory BGP-RIB over all using 207 bytes of memory 2 BGP path attribute entrie(s) using 128 bytes of memory 2 BGP AS-PATH entrie(s) using 90 bytes of memory 2 neighbor(s) using 9216 bytes of memory Neighbor 192.168.128.1 192.168.128.3 AS 99 100 MsgRcvd 140 138 MsgSent 136 140 TblVer 2 2 InQ 0 0 OutQ Up/Down State/Pfx (0) 00:11:24 1 (0) 00:18:31 1

R2#show ip bgp neighbor BGP neighbor is 192.168.128.1, remote AS 99, internal link Member of peer-group AAA for session parameters BGP version 4, remote router ID 192.168.128.1 BGP state ESTABLISHED, in this state for 00:11:42 Last read 00:00:38, last write 00:00:38 Hold time is 180, keepalive interval is 60 seconds Received 140 messages, 0 in queue 6 opens, 2 notifications, 19 updates 113 keepalives, 0 route refresh requests Sent 136 messages, 0 in queue 12 opens, 3 notifications, 6 updates 115 keepalives, 0 route refresh requests Minimum time between advertisement runs is 5 seconds Minimum time before advertisements start is 0 seconds

192

Border Gateway Protocol IPv4 (BGPv4)

Figure 7-44. Enable Peer Group - Router 3

R3#conf R3(conf)#router bgp 100 R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# neighbor R3(conf-router_bgp)# R3(conf-router_bgp)#end R3#show ip bgp summary BGP router identifier 192.168.128.3, local AS number 100 BGP table version is 1, main routing table version 1 1 network entrie(s) using 132 bytes of memory 3 paths using 204 bytes of memory BGP-RIB over all using 207 bytes of memory 2 BGP path attribute entrie(s) using 128 bytes of memory 2 BGP AS-PATH entrie(s) using 90 bytes of memory 2 neighbor(s) using 9216 bytes of memory Neighbor AS MsgRcvd 93 122 MsgSent 99 120 TblVer 1 1 InQ 0 0 OutQ Up/Down (0) 00:00:15 (0) 00:00:11 State/Pfx 1 1 AAA peer-group AAA no shutdown CCC peer-group CCC no shutdown 192.168.128.2 peer-group BBB 192.168.128.2 no shutdown 192.168.128.1 peer-group BBB 192.168.128.1 no shutdown

192.168.128.1 99 192.168.128.2 99 R3#show ip bgp neighbor

BGP neighbor is 192.168.128.1, remote AS 99, external link Member of peer-group BBB for session parameters BGP version 4, remote router ID 192.168.128.1 BGP state ESTABLISHED, in this state for 00:00:21 Last read 00:00:09, last write 00:00:08 Hold time is 180, keepalive interval is 60 seconds Received 93 messages, 0 in queue 5 opens, 0 notifications, 5 updates 83 keepalives, 0 route refresh requests Sent 99 messages, 0 in queue 5 opens, 4 notifications, 5 updates 85 keepalives, 0 route refresh requests Minimum time between advertisement runs is 30 seconds Minimum time before advertisements start is 0 seconds Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 1, neighbor version 1 Prefixes accepted 1 (consume 4 bytes), withdrawn 0 by peer Prefixes advertised 1, denied 0, withdrawn 0 from peer

Border Gateway Protocol IPv4 (BGPv4) | 193

www.dell.com | support.dell.com

Figure 7-45. Enable Peer Groups - Router 3 continued

Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 2, neighbor version 2 Prefixes accepted 1 (consume 4 bytes), withdrawn 0 by peer Prefixes advertised 1, denied 0, withdrawn 0 from peer Connections established 6; dropped 5 Last reset 00:12:01, due to Closed by neighbor Notification History 'HOLD error/Timer expired' Sent : 1 Recv: 0 'Connection Reset' Sent : 2 Recv: 2 Last notification (len 21) received 00:12:01 ago ffffffff ffffffff ffffffff ffffffff 00150306 00000000 Local host: 192.168.128.2, Local port: 65464 Foreign host: 192.168.128.1, Foreign port: 179 BGP neighbor is 192.168.128.3, remote AS 100, external link Member of peer-group BBB for session parameters BGP version 4, remote router ID 192.168.128.3 BGP state ESTABLISHED, in this state for 00:18:51 Last read 00:00:45, last write 00:00:44 Hold time is 180, keepalive interval is 60 seconds Received 138 messages, 0 in queue 7 opens, 2 notifications, 7 updates 122 keepalives, 0 route refresh requests Sent 140 messages, 0 in queue 7 opens, 4 notifications, 7 updates 122 keepalives, 0 route refresh requests Minimum time between advertisement runs is 30 seconds Minimum time before advertisements start is 0 seconds Capabilities advertised to neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) Capabilities received from neighbor for IPv4 Unicast : MULTIPROTO_EXT(1) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) ROUTE_REFRESH(2) CISCO_ROUTE_REFRESH(128) Update source set to Loopback 0 Peer active in peer-group outbound optimization For address family: IPv4 Unicast BGP table version 2, neighbor version 2 Prefixes accepted 1 (consume 4 bytes), withdrawn 0 by peer Prefixes advertised 1, denied 0, withdrawn 0 from peer

194

Border Gateway Protocol IPv4 (BGPv4)

8
Bare Metal Provisioning 2.0
Bare Metal Provisioning 2.0 is included as part of the FTOS image. It is supported on the following platforms: z Bare Metal Provisioning (BMP) improves accessibility to the switch by automatically loading pre-defined configurations and boot images that are stored in file servers. BMP can be used on a single switch or on multiple switches. For more information on using BMP and the different types of modes, refer to the Open Automation Guide. BMP eases configuration by automating the following steps: Boot images and running configurations are specified in a DHCP server. Switch boots up in Layer 3 mode with interfaces already in no shutdown mode and only enabling some basic protocols to protect the switch and the network. The first port that receives the DHCP server response retains the IP address provided by the DHCP server during the BMP process. All other management and user ports are shut down. Files are automatically downloaded from a file server. After the BMP process is complete, the IP address is released and the configuration is applied by the switch.

BMP is enabled on a brand new, factory-loaded switch. You can enable and disable BMP using the following steps: 1. Configure a reload mode using the reload-type command. 2. Reload the switch in the configured mode using the reload command.

Prerequisites
Before you use BMP 2.0 to auto-configure a supported Dell Networking switch, you must first configure a Dynamic Host Configuration Protocol (DHCP) server and a file server in the network. Optionally, you can also configure a Domain Name Server (DNS). For more information, refer to DHCP Server, Domain Name Server, and File Server.
Note: If the switch is connected to upstream aggregation switches that have VLT enabled, and the DHCP and file servers are reachable through the VLT LAG interface, you must configure the VLT members with the lacp ungroup member-independent vlt command. This allows the bottom switch to establish communication with the VLT switches.

Bare Metal Provisioning 2.0 | 195

www.dell.com | support.dell.com

Restrictions
BMP 2.0 is supported on the user ports and management ports of a switch. BMP 2.0 is not supported in a stacking environment.

Overview
On a new factory-loaded switch, the switch boots up in JumpStart mode. You can reconfigure a switch to reload between Normal and JumpStart mode. JumpStart (BMP) mode: The switch automatically configures all ports (management and user ports) as Layer 3 physical ports and acts as a DHCP client on the ports for a user-configured time (DHCP timeout). This is the default startup mode. It is set with the reload-type jump-start command. Normal mode: The switch loads the FTOS image and startup configuration file stored in the local flash. New configurations require that the Management IP and Management Interface be configured manually. This mode is set with the reload-type normal-reload command.

To reconfigure a switch to reload between Normal and JumpStart mode, use the reload-type command.
Command Syntax
reload-type {normal-reload | jump-start [config-download {enable | disable}] [dhcp-timeout minutes]}

Command Mode EXEC Privilege

Purpose Reload a switch running BMP version 2.0 in either Normal or JumpStart mode. If you reload in JumpStart mode, you can configure: config-download: Whether the switch boots up using the configuration file downloaded from the DHCP/file servers (enable) OR if the downloaded file will be discarded and the startup configuration file stored in the local flash will be used (disable). dhcp-timeout: The amount of time the switch waits for a DHCP server response before reverting to Normal mode and loading the startup configuration from the flash. The default time is infinity, which makes the switch continue to wait forever unless the stop jump-start command is given. Range: 1 to 50 minutes. Default: The switch tries to contact a DHCP server an infinite number of times. This command stops the jump-start reload process while it is in progress and changes the reload type to Normal mode. If the command is initiated while the switch is downloading an image or configuration file, the command takes effect when the DHCP release is sent.

stop jump-start

EXEC Privilege

The reload settings that you configure with the reload-type command are stored in non-volatile memory and retained for future reboots. Enter the reload command to reload the switch in the current configured mode: Normal or JumpStart mode.

196

Bare Metal Provisioning 2.0

To display the currently configured reload mode for a switch running BMP version 2.0, enter the show reload-type or show bootvar command.
FTOS#show reload type Reload-Type : config-download : dhcp-timeout : jump-start [Next boot :jump-start] enable 10

FTOS#show bootvar . . content truncated.. Reload Mode = jump-start File URL = tftp:/30.0.0.1/FTOS-SE-8-3-8-17.bin

Note: If a switch enters a loop while reloading in JumpStart mode because it continuously tries to contact

a DHCP server and a DHCP server is not found, connect to the switch using the console terminal and enter the stop jump-start command to interrupt the repeated discovery attempts. The startup configuration file stored in the local flash on the switch is loaded and the auto-configuration mode is automatically changed to Normal reload, i.e., BMP is disabled.

JumpStart mode
JumpStart (BMP) mode is the default boot mode configured for a new switch arriving from Dell Networking. This mode obtains the FTOS image and configuration file from a network source (DHCP server and file server).

DHCP Server
DHCP Configuration
You must first configure an external DHCP server before you can use the JumpStart mode on a switch. Configure the DHCP server with the set of parameters described below for each client switch. Refer to the FTOS Configuration Guide: Dynamic Host Configuration Protocol chapter for detailed information. The DHCP server is configured to assign an IP address to the switch and specify the files to download. One or more of the following parameters must be configured on the DHCP server. Boot File Name: The FTOS image to be loaded on the switch. The boot file name is expected to use Option 67 or the boot filename in the boot payload of the DHCP offer. If both are specified, Option 67 will be used. Configuration File Name: The configurations to be applied to the switch. The configuration file name is expected to use Option 209. File Server Address: The server where the Image and Configurations file are placed. The address is assumed to be a TFTP address unless it is given as a URL. The switch supports TFTP, HTTP, and FTP protocols, as well as files stored in Flash and external USB memory. If TFTP is used, you can add Option 66 or Option 150. Domain Name Server: (Optional.) The DNS server to be contacted to resolve the hostname through Option 6. IP Address: Dynamic IP address for the switch. This is used only for file transfers.

Bare Metal Provisioning 2.0 | 197

www.dell.com | support.dell.com

The DHCP option codes used are: 6 66 67 150 209 Domain Name Server IP TFTP Server name Boot filename TFTP server IP address Configuration File

Note: The boot file name and configuration file name must be in the correct format. If it is not, the switch will be unable to download the file from the DHCP server, and will behave as if the server could not be reached. The discovery process will continue, despite configured time-out, until the stop jump-start command is given.

URL Examples
##### FTOS image option bootfile-name "ftp://user:passwd@myserver/ FTOS-SE-8.3.10.1.bin";

Description

FTP URL with hostname (requires DNS)

option bootfile-name "http://10.20.4.1/FTOS-SE-8.3.10.1.bin"; HTTP URL with IP address option bootfile-name "tftp://10.20.4.1/ FTOS-SE-8.3.10.1.bin"; option bootfile-name "flash://FTOS-SE-8.3.10.1.bin"; ##### Configuration file could be given in the following way option config-file "ftp://user:[email protected]//home/user/ S4810-1.conf"; option config-file "http://myserver/S4810-1.conf"; option config-file "tftp://10.10.4.1/S4810-1.conf"; option config-file "flash://S4810-1.conf"; option config-file "usbflash://S60-1.conf";

TFTP URL with IP address Flash path relative to /f10/flash directory FTP URL with IP address HTTP URL with hostname (requires DNS) TFTP URL with IP address Flash path relative to /f10/flash directory External USB memory

MAC-Based IP assignment
One way to use the BMP mode most efficiently is to configure the DHCP server to assign a fixed IP address, FTOS image, and configuration file based on the switchs MAC address. When this is done, the same IP address is assigned to the switch even on repetitive reloads and the same configuration file will be retrieved when using the DNS server or the network-config file to determine the hostname. The assigned IP address is only used to retrieve the files from the file server. It is discarded after the files are retrieved.

198

Bare Metal Provisioning 2.0

Following is an example of a configuration of the DHCP server included on the most popular Linux distributions. The dhcpd.conf file shows assignment of a fixed IP address and configuration file based on the MAC address of the switch.
Parameter Example
option boot-filename code 67 = text; option tftp-server-address code 150 = ip-address; option config-file code 209 = text; subnet 10.20.30.0 netmask 255.255.255.0 { option domain-name-servers 20.30.40.1, 20.30.40.2; host S4810-1 { hardware ethernet 00:01:e8:8c:4d:0e; fixed-address 10.20.30.41; option boot-filename "tftp://10.20.4.1/FTOS-SE-8.3.10.1.bin"; option config-file "http://10.20.4.1/S4810-1.conf"; } host S4810-2 { hardware ethernet 00:01:e8:8c:4c:04; fixed-address 10.20.30.42; option tftp-server-address 10.20.4.1; filename "FTOS-SE-8.3.10.1.bin"; option config-file "S4810-2.conf"; }

Description

BMP 2.0 Syntax

MAC to IP mapping FTOS image Config file

BMP1.0 syntax

MAC to IP mapping FTOS image Config file

DHCP Retry Mechanism

BMP will request a different DHCP offer in the following scenarios: If the command reload-type config-download is enabled, the DHCP offer specifies both the boot image and the configuration file, and either download is successful, BMP will not request another DHCP offer. If the offer contains only a boot image that cannot be downloaded, BMP will request another DHCP offer. If the command reload-type config-download is enabled and the configuration file in the offer cannot be downloaded, BMP will request another DHCP offer.

DHCP Server IP Blacklist

If the process does not complete successfully, the DHCP server IP will be blacklisted and the DHCP process will be re-initiated. A DHCP server is maintained in the blacklist for ten minutes. If a DHCP offer is received from the blacklisted DHCP server, the offer will be rejected.

Bare Metal Provisioning 2.0 | 199

www.dell.com | support.dell.com

File Server
Set up a file server and ensure connectivity. The server that holds the boot and configuration files must be configured as the network source for the switch. The switch recognizes HTTP, TFTP, FTP, external USB memory and Flash URLs. For example:
tftp://serverip/filename tftp://hostname/filename ftp://user:passwd@serverip//mypath/filename ftp://user:passwd@hostname//mypath/filename http://serverip/filename http://hostname/filename flash://filename usbflash://filename filename (Assumes

TFTP)

When loading the FTOS image, if the FTOS image on the server is different from the image on the local flash, the switch downloads the image from the server onto the local flash and reloads using that image. Next, the switch tries to load the configuration file. If the configuration file is not specified or if the parameter is disabled, the switch loads the startup-config from the local flash.

config-download

Domain Name Server

Set up a Domain Name Server (DNS) to determine the host name applied in the switch startup configuration when no configuration file is retrieved from the DHCP server. The DNS server is contacted only when no configuration file is contained in a DHCP server response and the host name is not resolved from the network-config file on the switch. Refer to the FTOS Configuration Guide IPv4 Addressing chapter, Resolution of Host Names for information.

Switch boot and set-up behavior in JumpStart Mode

When the switch boots up in JumpStart mode all ports, including management ports, are placed in L3 mode in a no shut state. The switch acts as a DHCP client on these ports for a period of time (dhcp-timeout). This allows the switch time to send out a DHCP DISCOVER on all the interface up ports to the DHCP Server in order to obtain its IP address, boot image filename and configuration file from the DHCP server. 1. The switch begins boot up process in JumpStart mode (default mode)

200

Bare Metal Provisioning 2.0

2. The switch sends DHCP Discover on all the interface up ports.

00:01:31: 00:01:31: 00:01:31: 00:01:31: 00:01:47: 00:01:47: 00:01:47: 00:01:47: 00:01:47: %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: %JUMPSTART-5-JUMPSTART_DISCOVER: DHCP DHCP DHCP DHCP DHCP DHCP DHCP DHCP DHCP DISCOVER DISCOVER DISCOVER DISCOVER DISCOVER DISCOVER DISCOVER DISCOVER DISCOVER sent sent sent sent sent sent sent sent sent on on on on on on on on on Ma Te Te Te Te Te Fo Fo Ma 0/0. 0/0. 0/5. 0/6. 0/8. 0/35. 0/56. 0/60. 0/0.

3. The IP address, boot image filename and the configuration filename are reserved for the switch and provided in the DHCP reply (one-file read method). The switch receives its IP address, subnet mask, DHCP server IP, TFTP server address, DNS server IP, bootfile name and the configuration filename from the DHCP server. If a DHCP offer has no image path or configuration file path it is considered to be an invalid BMP DHCP offer, the offer is ignored. The first DHCP offer with IP address, FTOS image and configuration file, or the IP address and FTOS image, or the IP address and configuration file is chosen. 4. The DHCP OFFER is selected. a All other ports are set to shutdown mode.
%STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %JUMPSTART-5-DHCP_OFFER: %JUMPSTART-5-DHCP_OFFER: %JUMPSTART-5-DHCP_OFFER: %JUMPSTART-5-DHCP_OFFER: DHCP DHCP DHCP DHCP acquired IP 30.0.0.20 mask 255.255.0.0 server IP 30.1.1.1. tftp IP 30.0.0.1 dns IP 30.0.0.1 router IP 30.0.0.14. image file FTOS-SE-8.3.10.1.bin. config file pt-s4810-12.

00:01:33: 00:01:33: 00:01:33: 00:01:33:

5. The switch sends a unicast message to the file server to retrieve the named FTOS file and/or the configuration file from the base directory of the server. a If an option bootfile-name is used, the file name can be 256 bytes. If a filename field is specified in the DHCP Offer, the filename can be 128 bytes. The name can be a fully qualified URL or it can be a file name only. When an FTOS image is found, the switch compares that image to the version the chassis currently has loaded. If there is a mismatch, the switch applies the downloaded version and reloads.
*********VALID IMAGE*********** DOWNLOADED RELEASE HEADER : Release Image Major Version Release Image Minor Version Release Image Main Version Release Image Patch Version

: : : :

8 3 8 33

FLASH RELEASE HEADER B : Release Image Major Version : 8 Release Image Minor Version : 3 Release Image Main Version : 10 Release Image Patch Version : 1 00:04:05: %STKUNIT0-M:CP %JUMPSTART-5-JUMPSTART_DOWNLOAD: The FTOS image download is successful.

Erasing Sseries Primary Image, please wait

Bare Metal Provisioning 2.0 | 201

www.dell.com | support.dell.com

.......................................................................................... .......................................................................................... .......................................................................................... .......................................................................................... .......................................................................................... ...............................00:09:50: %STKUNsyncing disks... IT0-M:CP %CHMGR-1 5-RELOAD: User done request to reload the chassis rebooting

00:27:12: 00:27:12: 00:27:12: 00:27:12: 00:27:37: 00:27:37: 00:27:37:

If there is no version mismatch the switch downloads the configuration file.

%JUMPSTART-2-JUMPSTART_DOWNLOAD_START: The config file download has started. %JUMPSTART-5-JUMPSTART_RELEASE: DHCP RELEASE sent on Gi 0/1. %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Gi 0/1 %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 0/1 %JUMPSTART-5-JUMPSTART_DOWNLOAD: The config file download is successful. %JUMPSTART-5-CFG_APPLY: The downloaded config from dhcp server is being applied %SYS-5-CONFIG_LOAD: Loading configuration file,

%STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP %STKUNIT0-M:CP

If the configuration file is downloaded from the server, any saved startup-configuration on the flash is ignored. If no configuration file is downloaded from the server or the config-download parameter is disable, the startup-configuration file on the flash is loaded as in normal reload.

6. When the FTOS image and the configuration file have been downloaded, the IP address is released.
00:04:06: %STKUNIT0-M:CP %JUMPSTART-5-JUMPSTART_RELEASE: DHCP RELEASE sent on Fo 0/56. 00:04:06: %STKUNIT0-M:CP %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Fo 0/56

7. The switch applies the configuration. The switch is now up and running. It can be managed as usual.

202

Bare Metal Provisioning 2.0

9
Content Addressable Memory
Content Addressable Memory is supported on platforms Content Addressable Memory CAM Profiles Microcode CAM Profiling for ACLs When to Use CAM Profiling Important Points to Remember Select CAM Profiles CAM Allocation Test CAM Usage View CAM Profiles View CAM-ACL settings View CAM-ACL settings Configure IPv4Flow Sub-partitions Configure Ingress Layer 2 ACL Sub-partitions Return to the Default CAM Configuration CAM Optimization Applications for CAM Profiling Troubleshoot CAM Profiling

c et s

Content Addressable Memory

Content Addressable Memory (CAM) is a type of memory that stores information in the form of a lookup table. On Dell Networking systems, the CAM stores Layer 2 and Layer 3 forwarding information, access-lists (ACL), flows, and routing policies. On Dell Networking systems, there are one or two CAM (Dual-CAM) modules per port-pipe depending on the type of line card. The ExaScale EH and EJ series line cards are single-CAM line cards that support 10M and 40M CAM for storing the lookup information.

Content Addressable Memory | 203

www.dell.com | support.dell.com

The TeraScale EG-series line cards are dual-CAM and use two 18 Megabit CAM modules with a dedicated 512 IPv4 Forwarding Information Base (FIB), and flexible CAM allocations for Layer2, FIB, and ACLs. Either ExaScale 10G or 40G CAM line cards can be used in a system.

CAM Profiles
Dell Networking systems partition each CAM module so that it can store the different types of information. The size of each partition is specified in the CAM profile. A CAM profile is stored on every card, including each RPM. The same profile must be on every line card and RPM in the chassis. There is a default CAM profile and several other CAM profiles available so that you can partition the CAM according to your performance requirements. For example, the default profile has 1K Layer 2 ingress ACL entries. If you need more memory for Layer 2 ingress ACLs, select the profile l2-ipv4-inacl. Table 9-1 describes the available profiles. The default profile is an all-purpose profile that allocates CAM space according to the way Dell Networking systems are most commonly used. In general, non-default profiles allocate more space to particular regions to accommodate specific applications. The size of CAM partitions is measured in entries. The total CAM space is finite, therefor adding entries to one region necessarily decreases the number available to other regions.
Note: Not all CAM profiles and microcodes are available for all systems. Refer to the Command Line Interface Reference Guide for details regarding available profiles for each system. Table 9-1. CAM Profile Descriptions Description An all-purpose profile that allocates CAM space according to the way Dell Networking systems are most commonly used. Available Microcodes: default, lag-hash-align, lag-hash-mpls For EG-series line cards only. EG series line cards have two CAM modules per Port-pipe. Available Microcodes: default, ipv6-extacl Provides 320K entries for the IPv4 Forwarding Information Base (FIB) and reduces the IPv4 Flow partition to 12K. Available Microcodes: default, lag-hash-mpls Provides 16K entries for egress ACLs Available Microcodes: acl-group Provides IPv6 functionality. Available Microcodes: ipv6-extacl Provides 32K entries for Layer 2 ingress ACLs and 28K entries for Layer 3 IPv4 ingress ACLs. Available Microcodes: default

CAM Profile Default

eg-default ipv4-320k

ipv4-egacl-16k ipv6-extacl l2-ipv4-inacl

204

Content Addressable Memory

Table 9-1.

CAM Profile Descriptions (continued) Description Maintains the CAM allocations for the and IPv4 FIB while allocating more CAM space for the Ingress and Egress Layer 2 ACL, and IPv4 ACL regions. Available Microcodes: ipv6-extacl Provides VRF functionality for IPv4. Available Microcodes:ipv4-vrf Provides VRF functionality for both IPv4 and I.Pv6 Available Microcodes: ipv4-v6-vrf Provides IPv6 functionality; an alternate to ipv6-extacl that redistributes CAM space from the IPv4FIB to IPv4Flow and IPv6FIB. Available Microcodes: ipv6-extacl

CAM Profile unified-default

ipv4-VRF ipv4-v6-VRF ipv4-64k-ipv6

The size of CAM partitions is measured in entries. Table 9-1 shows the number of entries available in each partition for all CAM profiles. The total CAM space is finite, therefor adding entries to one region necessarily decreases the number available to other regions.
Table 9-2. CAM entries per partition
EgIPv4ACL EgIPv6ACL

EgL2ACL

Reserved

IPv4Flow

Profile

Default eg-default ipv4-320k pv4-egacl-16k ipv6-extacl l2-ipv4-inacl unified-default IPv4-VRF IPv4-v6-VRF ipv4-64k-ipv6

32K 32K 32K 32K 32K 32K 32K 32K 32K 32K

2K 2K 2K 2K 2K 33K 3K 3K 3K 2K

256K 512K 320K 192K 192K 64K 192K 160K 64K 64K

12K 12K 12K 8K 12K 27K 9K 2K 1K 12K

24K 24K 12K 24K 8K 8K 8K 12K 12K 24K

1K 1K 1K 0 1K 2K 2K 1K 1K 1K

1K 1K 1K 16K 1K 2K 2K 12K 11K 1K

8K 8K 4K 8K 2K 2K 2K 2K 2K 8K

0 32K 0 0 6K 0 6K 0 18K 16K

0 3K 0 0 3K 0 2K 0 4K 3K

IPv6Flow

Partition

IPv4ACL

IPv6ACL

IPv4FIB

IPv6FIB

L2ACL

L2FIB

0 4K 0 0 4K 0 4K 0 3K 4K

0 1K 0 0 2K 0 2K 0 1K 1K

Content Addressable Memory | 205

www.dell.com | support.dell.com

Microcode
Microcode is a compiled set of instructions for a CPU. On Dell Networking systems, the microcode controls how packets are handled. There is a default microcode, and several other microcodes are available, so that you can adjust packet handling according to your application. Specifying a microcode is mandatory when selecting a CAM profile (though you are not required to change it).
Note: Not all CAM profiles and microcodes are available for all systems. Refer to the Command Line Interface Reference Guide for details regarding available profiles for each system. Table 9-3. Microcode Descriptions Description Distributes CAM space for a typical deployment

Microcode default lag-hash-align

For applications that require the same hashing for bi-directional traffic (for example, VoIP call or P2P file sharing). For port-channels, this microcode maps both directions of a bi-directional flow to the same output link.
For hashing based on MPLS labels (up to five labels deep). With the default microcode, MPLS packets are distributed over a port-channel based on the MAC source and destination address. With the lag-hash-mpls microcode, MPLS packets are distributed across the port-channel based on IP source and destination address and IP protocol. This is applicable for MPLS packets with up to five labels. When the IP header is not available after the 5th label, hashing for default load-balance is based on MPLS labels. For packets with more than 5 labels, hashing is always based on the MAC source and destination address. Use this microcode when IPv6 is enabled. For applications that need 16k egress IPv4 ACLs (for example, the VLAN ACL Group feature, which permits group VLANs IP egress ACLs. Apply to IPv4 VRF CAM profile. Enable IPv4 and IPv6 CAM profiles for VRF.

lag-hash-mpls

ipv6-extacl acl-group ipv4-vrf ipv4-v6-vrf

CAM Profiling for ACLs

CAM Profiling for ACLs is supported on platform

et only.

The default CAM profile has 1K Layer 2 ingress ACL entries. If you need more memory for Layer 2 ingress ACLs, select the profile l2-ipv4-inacl. When budgeting your CAM allocations for ACLs and QoS configurations, remember that ACL and QoS rules might consume more than one CAM entry depending on complexity. For example, TCP and UDP rules with port range options might require more than one CAM entry.

206

Content Addressable Memory

The Layer 2 ACL CAM partition has sub-partitions for several types of information. Table 9-4 lists the sub-partition and the percentage of the Layer 2 ACL CAM partition that FTOS allocates to each by default.
Table 9-4. Layer 2 ACL CAM Sub-partition Sizes % Allocated 6 14 50 12 13 5

Partition Sysflow L2ACL *PVST QoS L2PT FRRP

You can re-configure the amount of space, in percentage, allocated to each sub-partition. As with the IPv4Flow partition, you can configure the Layer 2 ACL partition from EXEC Privilege mode or CONFIGURATION mode. The amount of space that you can distribute to the sub-partitions is equal to the amount of CAM space that the selected CAM profile allocates to the Layer 2 ACL partition. FTOS requires that you specify the amount of CAM space for all sub-partitions and that the sum of all sub-partitions is 100%. FTOS displays the following message if the total allocated space is not correct:
% Error: Sum of all regions does not total to 100%.

Boot Behavior
The profile and microcode loaded on the primary RPM determines the profile and microcode that is required on all other chassis components and is called the chassis profile. A profile mismatch condition exists if either the CAM profile or the microcode does not match. The following points describe line

card boot behavior when the line card profile does not match the chassis profile.
A microcode mismatch constitutes a profile mismatch. When the line card profile and chassis profile are of the same type (single-CAM or dual-CAM), but their CAM profiles do not match, the line card must load a new profile and therefore takes longer to come online. If you insert a single-CAM line card into a chassis with a dual-CAM profile, the system displays Message 1. The line card boots with the default (single-CAM) profile and remains in a problem state (Figure 9-1). The line card cannot forward traffic in a problem state.

Content Addressable Memory | 207

www.dell.com | support.dell.com

If you insert a dual-CAM line card into a chassis with a single-CAM profile, the line card boots with a matching profile, but operates with a lower capability.

Message 1 EF Line Card with EG Chassis Profile Error

# Before reload: 01:09:56: %RPM0-P:CP %CHMGR-4-EG_PROFILE_WARN: If EG CAM profile is selected, non-EG cards will be in problem state after reload # After reload: 00:04:46: %RPM0-P:CP %CHMGR-3-PROFILE_MISMATCH: Mismatch: line card 1 has mismatch CAM profile or microcode

Message 2 EH Line Card with EG Chassis Profile Error

# Before reload: 01:09:56: %RPM0-P:CP %CHMGR-4-EH_PROFILE_WARN: If EH CAM profile is selected, non-EJ cards will be in problem state after reload # After reload: 00:04:46: %RPM0-P:CP %CHMGR-3-PROFILE_MISMATCH: Mismatch: line card 1 has mismatch CAM profile or microcode

Figure 9-1.

EF Line Card with EG Chassis ProfileCard Problem

R1#show linecard 1 brief -- Line card Status Next Boot Required Type Current Type Hardware Rev Num Ports Up Time FTOS Version Jumbo Capable 1 : : : : : : : : : -card problem - mismatch cam profile online E48TF - 48-port 10/100/1000Base-T line card with RJ-45 interfaces (EF) E48TF - 48-port 10/100/1000Base-T line card with RJ-45 interfaces (EF) Base - 1.1 PP0 - 1.1 PP1 - 1.1 48 0 sec 7.6.1.0 yes

Figure 9-2.

EH Line Card with EG Chassis ProfileCard Problem

R1#show linecard 1 brief -- Line card Status Next Boot Required Type Current Type Hardware Rev Num Ports Up Time FTOS Version Jumbo Capable 1 : : : : : : : : : -card problem - mismatch cam profile online E90MH - 90-port 10/100/1000Base-T line card with mini RJ-21 interfaces (EH) E90MH - 90-port 10/100/1000Base-T line card with mini RJ-21 interfaces (EH) Base - 0.3 PP0 - 1.1 PP0 - PP1 90 0 sec 8.1.1.0 yes

208

Content Addressable Memory

When to Use CAM Profiling

The CAM profiling feature enables you to partition the CAM to best suit your application. For example: Configure more Layer 2 FIB entries when the system is deployed as a switch. Configure more Layer 3 FIB entries when the system is deployed as a router. Configure more ACLs (when IPv6 is not employed). Hash MPLS packets based on source and destination IP addresses for LAGs. See LAG Hashing. Hash based on bidirectional flow for LAGs. See LAG Hashing based on Bidirectional Flow. Optimize the VLAN ACL Group feature, which permits group VLANs for IP egress ACLs. See CAM profile for the VLAN ACL group feature.

Important Points to Remember

CAM Profiling is available on the E-Series TeraScale with FTOS versions 6.3.1.1 and later. All line cards within a single system must have the same CAM profile; this profile must match the system CAM profile (the profile on the primary RPM). FTOS automatically reconfigures the CAM profile on line cards and the secondary RPM to match the system CAM profile by saving the correct profile on the card and then rebooting it. The CAM configuration is applied to entire system when you use CONFIGURATION mode commands. You must save the running-configuration to affect the change. All CAM configuration commands require you to reboot the system. When budgeting your CAM allocations for ACLs and QoS configurations, remember that ACL and QoS rules might consume more than one CAM entry depending on complexity. For example, TCP and UDP rules with port range options might require more than one CAM entry. See Pre-calculating Available QoS CAM Space. After you install a secondary RPM, copy the running-configuration to the startup-configuration so that the new RPM has the correct CAM profile.

Select CAM Profiles

A CAM profile is selected in CONFIGURATION mode. The CAM profile is applied to entire system, however, you must save the running-configuration to affect the change. All components in the chassis must have the same CAM profile and microcode. The profile and microcode loaded on the primary RPM determines the profile that is required on all other chassis components. If a newly installed line card has a profile different from the primary RPM, the card reboots so that it can load the proper profile. If a the standby RPM has a profile different from the primary RPM, the card reboots so that it can load the proper profile.

Content Addressable Memory | 209

www.dell.com | support.dell.com

To change the CAM profile on the entire system:

Step 1 Task Select a CAM profile. Command Syntax
cam-profile profile microcode
microcode

Command Mode CONFIGURATION

Note: If selecting a cam-profile for VRF (cam-profile ipv4-vrf or ipv4-v6-vrf), implement the command in the CONFIGURATION mode only. If you use EXEC Privilege mode, the linecards may go into an error state. 2 Save the running-configuration.
copy running-config startup-config show cam-profile summary

EXEC Privilege

Verify that the new CAM profile will be written to the CAM on the next boot. Reload the system.

EXEC Privilege

reload

EXEC Privilege

CAM Allocation
User Configurable CAM Allocations is available on platforms:

cs

Allocate space for IPV4 ACLs and QoS regions, and IPv6 6 ACLs and QoS regions on the C-Series and S-Series by using the cam-acl command in CONFIGURATION mode. The CAM space is allotted in FP blocks. The total space allocated must equal 13 FP blocks. Note that there are 16 FP blocks, but the System Flow requires 3 blocks that cannot be reallocated. The default CAM Allocation settings on a C-Series system are: L3 ACL (ipv4acl): 6 L2 ACL(l2acl) : 5 IPv6 L3 ACL (ipv6acl): 0 L3 QoS (ipv4qos): 1 L2 QoS (l2qos): 1 L2PT (l2pt): 1 MAC ACLs (ipmacacl): 2 ECFMACL (ecfmacl): 0 VMAN QoS (vman-qos): 0 VMAN Dual QoS (vman-dual-qos): 0

The ipv6acl and vman-dual-qos allocations must be entered as a factor of 2 (2, 4, 6, 8, 10). All other profile allocations can use either even or odd numbered ranges. You must save the new CAM settings to the startup-config (write-mem or copy run start) then reload the system for the new settings to take effect.

210

Content Addressable Memory

To configure the IPv4 and IPv6 ACLs and Qos regions on the entire system:
Step 1 Task Select a cam-acl action Command Syntax
cam-acl [default | l2acl]

Command Mode CONFIGURATION

Note: Selecting default resets the CAM entries to the default settings. Select l2acl to allocate space for the ACLs, and QoS regions. 2 Enter the number of FP blocks for each region.
l2acl number ipv4acl number ipv6acl number, ipv4qos number l2qos number, l2pt number ipmacacl number ecfmacl number [vman-qos | vman-dual-qos number

EXEC Privilege

3 4

Verify that the new settings will be written to the CAM on the next boot. Reload the system.

show cam-acl reload

EXEC Privilege EXEC Privilege

Test CAM Usage

The test cam-usage command is supported on platforms

ces

This command applies to both IPv4 and IPv6 CAM profiles, but is best used when verifying QoS optimization for IPv6 ACLs. Use this command to determine whether sufficient ACL CAM space is available to enable a service-policy. Create a Class Map with all required ACL rules, then execute the test cam-usage command in Privilege mode to verify the actual CAM space required. Figure 9-3 gives a sample of the output shown when executing the command. The status column indicates whether or not the policy can be enabled.
Figure 9-3. Command Example: test cam-usage (C-Series)

FTOS#test cam-usage service-policy input TestPolicy linecard all Linecard | Portpipe | CAM Partition | Available CAM | Estimated CAM per Port | Status -----------------------------------------------------------------------------------------2 | 1 | IPv4Flow | 232 | 0 | Allowed 2 | 1 | IPv6Flow | 0 | 0 | Allowed 4 | 0 | IPv4Flow | 232 | 0 | Allowed 4 | 0 | IPv6Flow | 0 | 0 | Allowed FTOS#

Content Addressable Memory | 211

www.dell.com | support.dell.com

View CAM Profiles

View the current CAM profile for the chassis and each component using the command show cam-profile, as shown in Figure 9-4. This command also shows the profile that will be loaded upon the next chassis or component reload.
Figure 9-4. Viewing CAM Profiles on E-Series TeraScale

FTOS#show cam-profile -- Chassis CAM Profile -CamSize Profile Name L2FIB L2ACL IPv4FIB IPv4ACL IPv4Flow EgL2ACL EgIPv4ACL Reserved FIB : ACL : Flow : EgACL : MicroCode Name --More-: : : : : : : : : : : 0 0 0 0 18-Meg Current Settings Default 32K entries 1K entries 256K entries 12K entries 24K entries 1K entries 1K entries 8K entries entries : 0 entries : 0 entries : 0 entries : 0 : Default

: : : : : : : : : :

Next Boot Default 32K entries 1K entries 256K entries 12K entries 24K entries 1K entries 1K entries 8K entries entries entries entries entries : Default

View a brief output of the command show cam-profile using the summary option. The command show running-config cam-profile shows the current profile and microcode (Figure 9-5).
Note: If you select the CAM profile from CONFIGURATION mode, the output of this command does not reflect any changes until you save the running-configuration and reload the chassis. Figure 9-5. Viewing CAM Profile Information in the Running-configuration

FTOS#show running-config cam-profile ! cam-profile default microcode default FTOS#

View CAM-ACL settings

View the current cam-acl settings for the C-Series and S-Series systems chassis and each component using the command show cam-acl, as shown in Figure 9-6.

212

Content Addressable Memory

Figure 9-6.

View CAM-ACl settings on C-Series and S-Series

FTOS# show cam-acl -- Chassis Cam ACL -Current Settings(in block sizes) L2Acl : 2 Ipv4Acl : 2 Ipv6Acl : 2 Ipv4Qos : 2 L2Qos : 2 L2PT : 1 IpMacAcl : 2 VmanQos : 0 VmanDualQos : 0 -- Line card 0 -Current Settings(in block sizes) L2Acl : 2 Ipv4Acl : 2 Ipv6Acl : 2 Ipv4Qos : 2 L2Qos : 2 L2PT : 1 IpMacAcl : 2 VmanQos : 0 VmanDualQos : 0 -- Line card 6 -Current Settings(in block sizes) L2Acl : 2 Ipv4Acl : 2 Ipv6Acl : 2 Ipv4Qos : 2 L2Qos : 2 L2PT : 1 IpMacAcl : 2 VmanQos : 0 VmanDualQos : 0

View CAM Usage

View the amount of CAM space available, used, and remaining in each partition (including IPv4Flow and Layer 2 ACL sub-partitions) using the command show cam-usage from EXEC Privilege mode, as shown in Figure 9-7.

Content Addressable Memory | 213

www.dell.com | support.dell.com

Figure 9-7.

Viewing CAM Usage Information

R1#show cam-usage Linecard|Portpipe| CAM Partition | Total CAM | Used CAM |Available CAM ========|========|=================|=============|=============|============== 1 | 0 | IN-L2 ACL | 1008 | 320 | 688 | | IN-L2 FIB | 32768 | 1132 | 31636 | | IN-L3 ACL | 12288 | 2 | 12286 | | IN-L3 FIB | 262141 | 14 | 262127 | | IN-L3-SysFlow | 2878 | 45 | 2833 | | IN-L3-TrcList | 1024 | 0 | 1024 | | IN-L3-McastFib | 9215 | 0 | 9215 | | IN-L3-Qos | 8192 | 0 | 8192 | | IN-L3-PBR | 1024 | 0 | 1024 | | IN-V6 ACL | 0 | 0 | 0 | | IN-V6 FIB | 0 | 0 | 0 | | IN-V6-SysFlow | 0 | 0 | 0 | | IN-V6-McastFib | 0 | 0 | 0 | | OUT-L2 ACL | 1024 | 0 | 1024 | | OUT-L3 ACL | 1024 | 0 | 1024 | | OUT-V6 ACL | 0 | 0 | 0 1 | 1 | IN-L2 ACL | 320 | 0 | 320 | | IN-L2 FIB | 32768 | 1136 | 31632 | | IN-L3 ACL | 12288 | 2 | 12286 | | IN-L3 FIB | 262141 | 14 | 262127 | | IN-L3-SysFlow | 2878 | 44 | 2834 --More--

Configure IPv4Flow Sub-partitions

IPv4Flow sub-partitions are supported on platform

The IPv4Flow CAM partition has sub-partitions for several types of information. Table 9-5 lists the types of information stored in this partition and the number of entries that FTOS allocates to each type.
Table 9-5. IPv4Flow CAM Sub-partition Sizes Space Allocated (TeraScale) 3K 1K 2K 5K 1K Space Allocated (ExaScale) 3K 1K 2K 5K 1K

Partition ACL Multicast FIB/ACL PBR QoS System Flow Trace Lists

You can re-configure the amount of space allocated for each type of entry. FTOS requires that you specify an amount of CAM space for all types and in the order shown in Table 9-5.

214

Content Addressable Memory

The IPv4Flow configuration is applied to entire system when you enter the command cam-ipv4flow from CONFIGURATION mode, however, you must save the running-configuration to affect the change.

The amount of space that is allocated among the sub-partitions must be equal to the amount of CAM space allocated to IPv4Flow by the selected CAM profile (see Table 9-1.); Message 3 is displayed if the total allocated space is not correct.
Message 3 IPv4Flow Configuration Error
% Error: Total size must add up to match IPv4flow size of 24K required by the configured profile.

The minimum amount of space that can be allocated to any sub-partition is 1K, except for System flow, for which the minimum is 4K. To re-allocate CAM space within the IPv4Flow partition on the entire system:
Step 1 2 Task Re-allocate CAM space within the IPv4Flow partition. Save the running-configuration. Command Syntax
cam-ipv4flow

Command Mode CONFIGURATION EXEC Privilege

copy running-config startup-config show cam-ipv4flow reload

3 4

Verify that the new CAM configuration will be written to the CAM on the next boot. Reload the system.

EXEC Privilege EXEC Privilege

Content Addressable Memory | 215

www.dell.com | support.dell.com

Figure 9-8.

Configuring IPv4Flow on the Entire System

FTOS(conf)#cam-ipv4flow default FTOS#copy running-config startup-config File with same name already exist. Proceed to copy the file [confirm yes/no]: yes ! 3914 bytes successfully copied FTOS#sh cam-ipv4flow -- Chassis Cam Ipv4Flow -Current Settings Multicast Fib/Acl : 8K Pbr : 2K Qos : 7K System Flow : 6K Trace Lists : 1K -- Line card 0 -Current Settings 8K 2K 7K 6K 1K Next Boot 9K 1K 8K 5K 1K

Next Boot 9K 1K 8K 5K 1K

Multicast Fib/Acl Pbr Qos System Flow Trace Lists -- Line card 1 --

: : : : :

Current Settings Multicast Fib/Acl Pbr Qos System Flow Trace Lists : : : : : 8K 2K 7K 6K 1K

Next Boot 9K 1K 8K 5K 1K

Configure Ingress Layer 2 ACL Sub-partitions

IPv4Flow sub-partitions are supported on platform

The Ingress Layer 2 ACL CAM partition has sub-partitions for several types of information. Table 9-9 lists the sub-partition and the percentage of the Ingress Layer 2 ACL CAM partition that FTOS allocates to each by default.
Figure 9-9. Partition Sysflow L2ACL *PVST QoS Layer 2 ACL CAM Sub-partition Sizes % Allocated 6 14 50 12

216

Content Addressable Memory

Figure 9-9. Partition L2PT FRRP

Layer 2 ACL CAM Sub-partition Sizes % Allocated 13 5

You can re-configure the amount of space, in percentage, allocated to each sub-partition. Apply the Ingress Layer 2 ACL configuration to entire system by entering the command cam-l2acl from CONFIGURATION mode, however, you must save the running-configuration to affect the change.

The amount of space that you can distribute to the sub-partitions is equal to the amount of CAM space that the selected CAM profile allocates to the Ingress Layer 2 ACL partition (see Table 9-1). FTOS requires that you specify the amount of CAM space for all sub-partitions and that the sum of all sub-partitions is 100%. FTOS displays message Message 4 if the total allocated space is not correct.
Message 4 Layer 2 ACL Configuration Error % Error: Sum of all regions does not total to 100%.

*
Step 1 2 3 4

Note: You must allocate at least (<number of VLANs> * <Number of switching ports per port-pipe>) entries at least when employing PVST+ . For example, the default CAM Profile allocates 1000 entries to the Ingress Layer 2 ACL CAM region, and a 48-port linecard has two port-pipes with 24 ports each. If you have 5 VLANs, then you must allocate at least 120 (5*24) entries to the PVST Ingress Layer 2 ACL CAM region, which is 12% of the total 1000 available entries.

To re-allocate CAM space within the Ingress Layer 2 ACL partition on the entire system (Figure 9-10):
Task Re-allocate CAM space within the Ingress Layer 2 ACL partition. Save the running-configuration. Command Syntax
cam-l2acl

Command Mode CONFIGURATION EXEC Privilege

copy running-config startup-config show cam-l2acl reload

Verify that FTOS will write the new CAM configuration to the CAM on the next boot. Reload the system.

EXEC Privilege EXEC Privilege

Content Addressable Memory | 217

www.dell.com | support.dell.com

Figure 9-10. Configuring Ingress Layer 2 ACL on the Entire System

FTOS(conf)#do show cam-l2acl | find Line card 1 -- Line card 1 -Current Settings(in percent) Sysflow : 6 L2Acl : 14 Pvst : 50 Qos : 12 L2pt : 13 Frrp : 5 [output omitted] FTOS(conf)#cam-l2acl system-flow 100 l2acl 0 p 0 q 0 l 0 f 0 FTOS(conf)#do show cam-l2acl | find Line card 1 -- Line card 1 -Current Settings(in percent) Sysflow : 6 L2Acl : 14 Pvst : 50 Qos : 12 L2pt : 13 Frrp : 5 [output omitted] FTOS(conf)#do copy run start File with same name already exist. Proceed to copy the file [confirm yes/no]: yes ! 8676 bytes successfully copied 02:00:49: %RPM0-P:CP %FILEMGR-5-FILESAVED: Copied running-config to startup-config in flash by default FTOS(conf)#do show cam-l2acl | find Line card 1 -- Line card 1 -Current Settings(in percent) Next Boot(in percent) Sysflow : 6 100 L2Acl : 14 5 Pvst : 50 5 Qos : 12 5 L2pt : 13 5 Frrp : 5 5

Return to the Default CAM Configuration

Return to the default CAM Profile, microcode, IPv4Flow, or Layer 2 ACL configuration using the keyword default from EXEC Privilege mode or from CONFIGURATION mode, as shown in Figure 9-11.

218

Content Addressable Memory

Figure 9-11.

Returning to the default Configuration

FTOS(conf)#cam-profile ? default Enable default CAM profile eg-default Enable eg-default CAM profile ipv4-320k Enable 320K CAM profile ipv4-egacl-16k Enable CAM profile with 16K IPv4 egress ACL ipv6-extacl Enable CAM profile with extended ACL l2-ipv4-inacl Enable CAM profile with 32K L2 and 28K IPv4 ingress ACL unified-default Enable default unified CAM profile FTOS(conf)#cam-profile default microcode ? default Enable default microcode lag-hash-align Enable microcode with LAG hash align lag-hash-mpls Enable microcode with LAG hash MPLS FTOS(conf)#cam-profile default microcode default FTOS(conf)#cam-ipv4flow ? default Reset IPv4flow CAM entries to default setting multicast-fib Set multicast FIB entries FTOS(conf)#cam-l2acl ? default Reset L2-ACL CAM entries to default setting system-flow Set system flow entries

CAM Optimization
CAM optimization is supported on platforms

cs

When this command is enabled, if a Policy Map containing classification rules (ACL and/or dscp/ ip-precedence rules) is applied to more than one physical interface on the same port-pipe, only a single copy of the policy is written (only 1 FP entry will be used). When the command is disabled, the system behaves as described in this chapter.

Applications for CAM Profiling

LAG Hashing
FTOS includes a CAM profile and microcode that treats MPLS packets as non-IP packets. Normally, switching and LAG hashing is based on source and destination MAC addresses. Alternatively, you can base LAG hashing for MPLS packets on source and destination IP addresses. This type of hashing is allowed for MPLS packets with 5 labels or less. MPLS packets are treated as follows: When MPLS IP packets are received, FTOS looks up to 5 labels deep for the IP header. When an IP header is present, hashing is based on IP 3 tuple (source IP address, destination IP address, and IP protocol). If an IP header is not found after the 5th label, hashing is based on the MPLS labels.

Content Addressable Memory | 219

www.dell.com | support.dell.com

If the packet has more than 5 MPLS labels, hashing is based on the source and destination MAC address.

To enable this type of hashing, use the default CAM profile with the microcode lag-hash-mpls.

LAG Hashing based on Bidirectional Flow

To hash LAG packets such that both directions of a bidirectional flow (for example, VoIP or P2P file sharing) are mapped to the same output link in the LAG bundle, use the default CAM profile with the microcode lag-hash-align.

CAM profile for the VLAN ACL group feature

IPv4Flow sub-partitions are supported on platform

et only.

To optimize for the VLAN ACL Group feature, which permits group VLANs for the IP egress ACL, use the CAM profile ipv4-egacl-16k with the default microcode.
Note: Do not use this CAM profile for Layer 2 egress ACLs.

Troubleshoot CAM Profiling

CAM Profile Mismatches
The CAM profile on all cards must match the system profile. In most cases, the system corrects mismatches by copying the correct profile to the card, and rebooting the card. If three resets do not bring up the card, or if the system is running an FTOS version prior to 6.3.1.1, the system presents an error message. In this case, manually adjust the CAM configuration on the card to match the system configuration. Dell Networking recommends the following to prevent mismatches: Use the eg-default CAM profile in a chassis that has only EG Series line cards. If this profile is used in a chassis with non-EG line cards, the non-EG line cards enter a problem state. Before moving a card to a new chassis, change the CAM profile on a card to match the new system profile. After installing a secondary RPM into a chassis, copy the running-configuration to the startup-configuration. Change to the default profile if downgrading to and FTOS version earlier than 6.3.1.1. Use the CONFIGURATION mode commands so that the profile is change throughout the system. Use the EXEC Privilege mode commands to match the profile of a component to the profile of the target system.

220

Content Addressable Memory

QoS CAM Region Limitation

The default CAM profile allocates a partition within the IPv4Flow region to store QoS service policies. If the QoS CAM space is exceeded, messages similar to the ones in Message 5 are displayed.
Message 5 QoS CAM Region Exceeded
%EX2YD:12 %DIFFSERV-2-DSA_QOS_CAM_INSTALL_FAILED: Not enough space in L3 Cam(PolicyQos) for class 2 (Gi 12/20) entries on portpipe 1 for linecard 12 %EX2YD:12 %DIFFSERV-2DSA_QOS_CAM_INSTALL_FAILED: Not enough space in L3 Cam(PolicyQos) for class 5 (Gi 12/22) entries on portpipe 1 for linecard 12

If you exceed the QoS CAM space:

Step 1 2 Task Verify that you have configured a CAM profile that allocates 24K entries to the IPv4 system flow region. See View CAM Profiles. Allocate more entries in the IPv4Flow region to QoS. See Configure IPv4Flow Sub-partitions.

FTOS version 7.4.1 introduced the ability to view the actual CAM usage before applying a service-policy. The command test cam-usage service-policy provides this test framework, see Pre-calculating Available QoS CAM Space.
Note: For troubleshooting other CAM issues see the E-Series Network Operations Guide.

Content Addressable Memory | 221

222 www.dell.com | support.dell.com

Content Addressable Memory

10
Debugging and Diagnostics
The chapter contains the following major sections: Offline diagnostics Trace logs Last restart reason (S60) show hardware commands (S60) Hardware watchdog timer Buffer tuning Troubleshooting packet loss Application core dumps Mini core dumps

Offline diagnostics
The offline diagnostics test suite is useful for isolating faults and debugging hardware.The diagnostics tests are grouped into three levels: Level 0Level 0 diagnostics check for the presence of various components and perform essential path verifications. In addition, they verify the identification registers of the components on the board. Level 1A smaller set of diagnostic tests. Level 1 diagnostics perform status/self-test for all the components on the board and test their registers for appropriate values. In addition, they perform extensive tests on memory devices (e.g., SDRAM, flash, NVRAM, EEPROM, and CPLD) wherever possible. Level 2The full set of diagnostic tests. Level 2 diagnostics are used primarily for on-board loopback tests and more extensive component diagnostics. Various components on the board are put into loopback mode, and test packets are transmitted through those components. These diagnostics also perform snake tests using VLAN configurations.

Debugging and Diagnostics | 223

www.dell.com | support.dell.com

Important Points to Remember

You can only perform offline diagnostics on an offline standalone unit or offline member unit of a stack of three or more. You cannot perform diagnostics on the management or standby unit in a stack of two or more (Message 1).
Message 1 Offline Diagnostics on Master/Standby Error
Running Diagnostics on master/standby unit is not allowed on stack.

Perform offline diagnostics on one stack member at a time. Diagnostics only test connectivity, not the entire data path. Diagnostic results are stored on the flash of the unit on which you performed the diagnostics. When offline diagnostics are complete, the unit or stack member reboots automatically.

Running Offline Diagnostics

1. Place the unit in the offline state using the offline stack-unit command from EXEC Privilege mode, as shown in Figure 10-1. YOu cannot enter the command on a Master or Standby stack unit.
The system reboots when the off-line diagnostics complete. This is an automatic process. A warning message appears when the offline stack-unit command is implemented.
Warning - Diagnostic execution will cause stack-unit to reboot after completion of diags. Proceed with Offline-Diags [confirm yes/no]:y

Figure 10-1. Taking an S-Series Stack Unit Offline

FTOS#offline stack-unit 2 Warning - Diagnostic execution will cause stack-unit to reboot after completion of diags. Proceed with Offline-Diags [confirm yes/no]:y 5w6d12h: %STKUNIT0-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 2 down - stack unit offline 5w6d12h: %STKUNIT0-M:CP %IFMGR-1-DEL_PORT: Removed port: Gi 2/1-48 FTOS#5w6d12h: %STKUNIT1-S:CP %IFMGR-1-DEL_PORT: Removed port: Gi 2/1-48

2. Use the show system brief command from EXEC Privilege mode to confirm offline status, as shown in Figure 10-2.

224

Debugging and Diagnostics

Figure 10-2. Verifying the Offline/Online Status of an S-Series Stack Unit

FTOS#show system brief | no-more Stack MAC : 00:01:e8:d6:02:39 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S25V S25V 4.7.7.220 28 1 Management offline S50N S50N 4.7.7.220 52 2 Member online S25P S25P 4.7.7.220 28 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present -- Module Info -Unit Module No Status Module Type Ports --------------------------------------------------------------------------0 0 online S50-01-10GE-2C 2 0 1 online S50-01-12G-2S 2 1 0 online S50-01-10GE-2P 2 1 1 online S50-01-12G-2S 2 2 0 not present No Module 0 2 1 offline S50-01-12G-2S 2 -- Power Supplies -Unit Bay Status Type --------------------------------------------------------------------------0 0 up AC 0 1 absent 1 0 up AC 1 1 absent 2 0 up AC 2 1 absent

3. Start diagnostics on the unit using the command diag, as shown in Figure 10-3. When the tests are complete, the system displays syslog Message 2, and automatically reboots the unit. Diagnostic results are printed to a file in the flash using the filename format TestReport-SU-<stack-unit>.txt.
Message 2 Offline Diagnostics Complete
FTOS#00:09:32 : Diagnostic test results are stored on file: flash:/TestReport-SU-1.txt 00:09:37: %S50N:1 %DIAGAGT-6-DA_DIAG_DONE: Diags finished on stack unit 1 Diags completed... Rebooting the system now!!!

As shown in Figure 10-3 and Figure 10-4, log messages differ somewhat when diagnostics are done on a standalone unit and on a stack member.

Debugging and Diagnostics | 225

www.dell.com | support.dell.com

Figure 10-3. Running Offline Diagnostics on an S-Series Standalone Unit

FTOS#diag stack-unit 1 alllevels Warning - diagnostic execution will cause multiple link flaps on the peer side - advisable to shut directly connected ports Proceed with Diags [confirm yes/no]: yes 00:03:35: %S50N:1 %DIAGAGT-6-DA_DIAG_STARTED: Starting diags on stack unit 1 00:03:35 : Approximate time to complete these Diags ... 6 Min S50N#00:09:32 : Diagnostic test results are stored on file: flash:/TestReport-SU-0.txt 00:09:37: %S50N:0 %DIAGAGT-6-DA_DIAG_DONE: Diags finished on stack unit 0 Diags completed... Rebooting the system now!!! [reboot output omitted] S50N#00:01:35: %STKUNIT0-M:CP %SYS-5-CONFIG_I: Configured from console by dir Directory of flash: 1 2 3 4 5 6 drwdrwx drwd---rw-rw16384 1536 512 512 3854 12632 Jan Feb Aug Aug Sep Nov 01 29 15 15 24 05 1980 1996 1996 1996 1996 2008 00:00:00 00:05:22 23:09:48 23:09:52 03:43:46 17:15:16 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 . .. TRACE_LOG_DIR ADMIN_DIR startup-config TestReport-SU-1.txt console

flash: 3104256 bytes total (3086336 bytes free)

Figure 10-4 shows the output of the master and member units when you run offline diagnostics on a member unit.
Figure 10-4. Running Offline Diagnostics on an S-Series Stack Member
[output from master unit] FTOS#diag stack-unit 2 Warning - the stack unit will be pulled out of the stack for diagnostic execution Proceed with Diags [confirm yes/no]: yes Warning - diagnostic execution will cause multiple link flaps on the peer side - advisable to shut directly connected ports Proceed with Diags [confirm yes/no]: yes FTOS#00:03:13: %S25P:2 %DIAGAGT-6-DA_DIAG_STARTED: Starting diags on stack unit 2 00:03:13 : Approximate time to complete these Diags ... 6 Min 00:03:13 : Diagnostic test results will be stored on stack unit 2 file: flash:/ TestReport-SU-2.txt FTOS#00:03:35: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 2 down - card removed 00:08:50: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present 00:09:00: %STKUNIT1-M:CP %CHMGR-5-CHECKIN: Checkin from Stack unit 2 (type S25P, 28 ports) 00:09:00: %S25P:2 %CHMGR-0-PS_UP: Power supply 0 in unit 2 is up 00:09:00: %STKUNIT1-M:CP %CHMGR-5-STACKUNITUP: Stack unit 2 is up [output from the console of the unit in which diagnostics are performed] FTOS(stack-member-2)# Diagnostic test results are stored on file: flash:/TestReport-SU-2.txt Diags completed... Rebooting the system now!!!

4. View the results of the diagnostic tests using the command show file flash:// from EXEC Privilege mode, as shown in Figure 10-5.

226

Debugging and Diagnostics

Figure 10-5. Viewing the Results of Offline Diagnostics on a Standalone Unit

FTOS#show file flash://TestReport-SU-0.txt **********************************S-Series Diagnostics******************** Stack Unit Board Serial Number : DL267160098 CPU Version : MPC8541, Version: 1.1 PLD Version : 5 Diag image based on build : E_MAIN4.7.7.206 Stack Unit Board Voltage levels - 3.300000 V, 2.500000 V, 1.800000 V, 1.250000 V, 1.200000 V, 2.000000 V Stack Unit Board temperature : 26 Degree C Stack Unit Number : 0 ****************************Stack Unit EEPROM INFO******************************* ********MFG INFO******************* Data in Chassis Eeprom Mfg Info is listed as... Vendor Id: 07 Country Code: 01 Date Code: 12172007 Serial Number: DL267160098 Part Number: 7590003600 Product Revision: B Product Order Number: ${ **************************** LEVEL 0 DIAGNOSTICS************************** Test 0 - CPLD Presence Test ......................................... Hardware PCB Revision is - Revision B Test 1 - CPLD Hardware PCB Revision Test ............................ Test 2.000 - CPLD Fan-0 Presence Test ............................... Test 2.001 - CPLD Fan-1 Presence Test ............................... Test 2.002 - CPLD Fan-2 Presence Test ............................... Test 2.003 - CPLD Fan-3 Presence Test ............................... Test 2.004 - CPLD Fan-4 Presence Test ............................... Test 2.005 - CPLD Fan-5 Presence Test ............................... Test 3.000 - CPLD Power Bay-0 Presence Test ......................... Test 3.001 - CPLD Power Bay-1 Presence Test ......................... Test 4 - SDRAM Access Test .......................................... Test 5 - CPU Access Test ............................................ Test 6 - I2C Temp Access Test CPU Board ............................. Test 7 - I2C Temp Access Test Main Board ............................ Test 8 - RTC Access Test ............................................ --More-PASS PASS PASS PASS PASS PASS PASS PASS PASS NOT PRESENT PASS PASS PASS PASS PASS

Trace logs
In addition to the syslog buffer, FTOS buffers trace messages which are continuously written by various FTOS software tasks to report hardware and software events and status information. Each trace message provides the date, time, and name of the FTOS process. All messages are stored in a ring buffer and can be saved to a file either manually or automatically upon failover.

Debugging and Diagnostics | 227

www.dell.com | support.dell.com

Auto Save on Crash or Rollover

Exception information on for master or standby units is stored in the flash:/TRACE_LOG_DIR directory. This directory contains files that save trace information when there has been a task crash or timeout. On a master unit, the TRACE_LOG_DIR files can be reached by FTP or by using the show file command from the flash://TRACE_LOG_DIR directory. On a Standby unit, the TRACE_LOG_DIR files can be reached only by using the show file command from the flash://TRACE_LOG_DIR directory.
Note: Non-management member units do not support this functionality.

Last restart reason (S60)

If an S60 system restarted for some reason (automatically or manually), the show system command output includes the reason for the restart. The following table shows the reasons displayed in the output and their corresponding causes.
Table 10-1. Causes Remote power cycle of the chassis reload reboot after a crash Line card restart causes and reasons Displayed Reasons push button reset soft reset soft reset

show hardware commands (S60)

Note: The show hardware command tree is supported on the S60 only.

The show hardware command tree consists of EXEC Privilege commands used with the S60 system. These commands display information from a hardware sub-component and from hardware-based feature tables. Table 10-2 lists the show hardware commands available as of the latest FTOS version on the S60.
Note: The show hardware commands should only be used under the guidance of Dell Networking Technical Assistance Center.

228

Debugging and Diagnostics

Table 10-2. Command

show hardware Commands Description

show hardware stack-unit {0-11} fpga registers View the FPGA registers or FPGA interrupt statistics on the stack-unit. | stats show hardware stack-unit {0-11} cpu management statistics show hardware stack-unit {0-11} cpu data-plane statistics

View internal interface status of the stack-unit CPU port which connects to the external management interface. View driver-level statistics for the data-plane port on the CPU for the specified stack-unit. It provides insight into the packet types entering the CPU to see whether CPU-bound traffic is internal (IPC traffic) or network control traffic, which the CPU must process. View input and output statistics on the party bus, which carries inter-process communication traffic between CPUs. View the packet-drop counters on the buffering chip. View the ingress and egress internal packet-drop counters, MAC counters drop, and FP packet drops for the stack unit on per port basis. It assists in identifying the stack unit/port pipe/port that may experience internal drops. View the input and output statistics for a stack-port interface. View the counters in the field processors of the stack unit. View the details of the the FP Devices, and Hi gig ports on the stack-unit. Execute a specified bShell commands from the CLI without going into the bShell. View the Multicast IPMC replication table from the bShell. View the internal statistics for each port-pipe (unit) on per port basis. View the stack-unit internal registers for each port-pipe. View the tables from the bShell through the CLI without going into the bShell. View the buffering chip counters. View the details of the devices and Hi gig ports on the buffering chip. Execute a specified bShell command on the buffering chip from the CLI without going into the bShell. View the flow control table contents. This will be useful in finding out the ports that are back-pressured. View the Multicast IPMC indices from the bShell. When the start and end index are specified, display all the IPMC indices between them. Default indices: 1 -4094

show hardware stack-unit {0-11} cpu party-bus statistics show hardware stack-unit {0-11} drops buffering unit show hardware stack-unit {0-11} drops unit {0-1} port {0-1} show hardware stack-unit {0-11} stack-port {0-47} show hardware stack-unit {0-11} unit {0-1} counters show hardware stack-unit {0-11} unit {0-1} details show hardware stack-unit {0-11} unit {0-1} execute-shell-cmd {command} show hardware stack-unit {0-11} unit {0-1} ipmc-replication show hardware stack-unit {0-11} unit {0-1} port-stats [detail] show hardware stack-unit {0-11} unit {0-1} register show hardware stack-unit {0-11} unit {0-1} table-dump {table name} show hardware stack-unit {0-11} buffering-unit counters show hardware stack-unit {0-11} buffering-unit details show hardware stack-unit {0-11} buffering-unit execute-shell-cmd {command} show hardware stack-unit {0-11} buffering-unit flow-control-dump show hardware stack-unit {0-11} buffering-unit ipmc-show [start-index] [start-index]

Debugging and Diagnostics | 229

www.dell.com | support.dell.com

Table 10-2. Command

show hardware Commands Description View the internal statistics on the buffering chi for the Hi gig ports. View the internal queuing inside the buffer chip . This command displays the internal queue statistics for CPU Q statistics (per Cos Q), Multicast Q statistics ( per cos Q, and Unicast Q statistics (per port, per cos Q basis ). View the internal registers for the buffering chip. View the Buffering Unit scheduler (bcm drivers) output. This command displays scheduler reports for all fabric ports. View the tables in the buffering chip from the bShell without going into the bShell.

show hardware stack-unit {0-11} buffering-unit port-stats show hardware stack-unit {0-11} buffering-unit queue-stats [cpu | multicast | unicast] show hardware stack-unit {0-11} buffering-unit registers show hardware stack-unit {0-11} buffering-unit scheduler show hardware stack-unit {0-11} buffering-unit table-dump

Hardware watchdog timer

The hardware watchdog command automatically reboots an FTOS switch/router when a unit is unresponsive. This is a last resort mechanism intended to prevent a manual power cycle.
Command
hardware watchdog

Description Enable the hardware watchdog mechanism.

Buffer tuning
Buffer Tuning allows you to modify the way your switch allocates buffers from its available memory, and helps prevent packet drops during a temporary burst of traffic.
Note: Buffer Tuning is not supported on the S60 system.

The S-Series ASICs implement the key functions of queuing, feature lookups, and forwarding lookups in hardware. Forwarding Processor (FP) ASICs provide Ethernet MAC functions, queueing and buffering, as well as store feature and forwarding tables for hardware-based lookup and forwarding decisions. 1G and 10G interfaces use different FPs.

230

Debugging and Diagnostics

Table 10-3 describes the type and number of ASICs per platform.
Table 10-3. Hardware S50N, S50V S25V, S25P, S25N ASICS by Platform FP 2 1 CSF 0 0

You can tune buffers at three locations, as shown in Figure 10-6. 1. CSF Output queues going from the CSF. 2. FP UplinkOutput queues going from the FP to the CSF IDP links. 3. Front-End LinkOutput queues going from the FP to the front-end PHY. All ports support eight queues, 4 for data traffic and 4 for control traffic. All 8 queues are tunable. Physical memory is organized into cells of 128 bytes. The cells are organized into two buffer pools dedicated buffer and dynamic buffer. Dedicated buffer is reserved memory that cannot be used by other interfaces on the same ASIC or by other queues on the same interface. This buffer is always allocated, and no dynamic recarving takes place based on changes in interface status. Dedicated buffers introduce a trade-off. They provide each interface with a guaranteed minimum buffer to prevent an overused and congested interface from starving all other interfaces. However, this minimum guarantee means the buffer manager does not reallocate the buffer to an adjacent congested interface, which means that in some cases, memory is underused. Dynamic buffer is shared memory that is allocated as needed, up to a configured limit. Using dynamic buffers provides the benefit of statistical buffer sharing. An interface requests dynamic buffers when its dedicated buffer pool is exhausted. The buffer manager grants the request based on three conditions: The number of used and available dynamic buffers The maximum number of cells that an interface can occupy Available packet pointers (2k per interface). Each packet is managed in the buffer using a unique packet pointer. Thus, each interface can manage up to 2k packets.

You can configure dynamic buffers per port on both 1G and 10G FPs and per queue on CSFs. By default, the FP dynamic buffer allocation is 10 times oversubscribed. For the 48-port 1G card: Dynamic Pool= Total Available Pool(16384 cells) Total Dedicated Pool = 5904 cells Oversubscription ratio = 10 Dynamic Cell Limit Per port = 59040/29 = 2036 cells

Debugging and Diagnostics | 231

www.dell.com | support.dell.com

Figure 10-6. Buffer Tuning Points

CSF Unit 3

1 IDP Switch Links 2

FP Unit 1

Front-end Links

PHY

PHY

Deciding to tune buffers

Dell Networking recommends exercising caution when configuring any non-default buffer settings, as tuning can significantly affect system performance. The default values work for most cases. As a guideline, consider tuning buffers if traffic is very bursty (and coming from several interfaces). In this case: Reduce the dedicated buffer on all queues/interfaces. Increase the dynamic buffer on all interfaces. Increase the cell pointers on a queue that you are expecting will receive the largest number of packets.

232

Debugging and Diagnostics

Buffer tuning commands

Task Define a buffer profile for the FP queues. Define a buffer profile for the CSF queues. Change the dedicated buffers on a physical 1G interface. Change the maximum amount of dynamic buffers an interface can request. Change the number of packet-pointers per queue. Apply the buffer profile to a line card. Apply the buffer profile to a CSF to FP link. Command
buffer-profile fp fsqueue buffer-profile csf csqueue buffer dedicated

Command Mode CONFIGURATION CONFIGURATION BUFFER PROFILE BUFFER PROFILE BUFFER PROFILE CONFIGURATION CONFIGURATION

buffer dynamic

buffer packet-pointers buffer fp-uplink linecard buffer csf linecard

FTOS Behavior: If you attempt to apply a buffer profile to a non-existent port-pipe, FTOS displays the following message. However, the configuration still appears in the running-config.
%DIFFSERV-2-DSA_BUFF_CARVING_INVALID_PORT_SET: Invalid FP port-set 2 for linecard 2. Valid range of port-set is <0-1>

Configuration changes take effect immediately and appear in the running configuration. Since under normal conditions all ports do not require the maximum possible allocation, the configured dynamic allocations can exceed the actual amount of available memory; this is called oversubscription. If you choose to oversubscribe the dynamic allocation, a burst of traffic on one interface might prevent other interfaces from receiving the configured dynamic allocation, which causes packet loss. You cannot allocate more than the available memory for the dedicated buffers. If the system determines that the sum of the configured dedicated buffers allocated to the queues is more than the total available memory, the configuration is rejected, returning a syslog message similar to the following.
Table 10-4. Buffer Allocation Error

00:04:20: %S50N:0 %DIFFSERV-2-DSA_DEVICE_BUFFER_UNAVAILABLE: Unable to allocate dedicated buffers for stack-unit 0, port pipe 0, egress port 25 due to unavailability of cells

FTOS Behavior: When you remove a buffer-profile using the command no buffer-profile [fp | csf] from CONFIGURATION mode, the buffer-profile name still appears in the output of show buffer-profile [detail | summary]. After a line card reset, the buffer profile correctly returns to the default values, but the profile name remains. Remove it from the show buffer-profile [detail | summary] command output by entering no buffer [fp-uplink |csf] linecard port-set buffer-policy from CONFIGURATION mode and no buffer-policy from INTERFACE mode.

Display the allocations for any buffer profile using the show commands in Figure 10-8. Display the default buffer profile using the command show buffer-profile {summary | detail} from EXEC Privilege mode, as shown in Figure 10-7.

Debugging and Diagnostics | 233

www.dell.com | support.dell.com

Figure 10-7. Display the Default Buffer Profile

FTOS#show buffer-profile detail interface gigabitethernet 0/1 Interface Gi 0/1 Buffer-profile Dynamic buffer 194.88 (Kilobytes) Queue# Dedicated Buffer Buffer Packets (Kilobytes) 0 2.50 256 1 2.50 256 2 2.50 256 3 2.50 256 4 9.38 256 5 9.38 256 6 9.38 256 7 9.38 256

Figure 10-8. Displaying Buffer Profile Allocations

FTOS#show running-config interface tengigabitethernet 2/0 ! interface TenGigabitEthernet 2/0 no ip address mtu 9252 switchport no shutdown buffer-policy myfsbufferprofile FTOS#sho buffer-profile detail int gi 0/10 Interface Gi 0/10 Buffer-profile fsqueue-fp Dynamic buffer 1256.00 (Kilobytes) Queue# Dedicated Buffer Buffer Packets (Kilobytes) 0 3.00 256 1 3.00 256 2 3.00 256 3 3.00 256 4 3.00 256 5 3.00 256 6 3.00 256 7 3.00 256 FTOS#sho buffer-profile detail fp-uplink stack-unit 0 port-set 0 Linecard 0 Port-set 0 Buffer-profile fsqueue-hig Dynamic Buffer 1256.00 (Kilobytes) Queue# Dedicated Buffer Buffer Packets (Kilobytes) 0 3.00 256 1 3.00 256 2 3.00 256 3 3.00 256 4 3.00 256 5 3.00 256 6 3.00 256 7 3.00 256

234

Debugging and Diagnostics

Using a pre-defined buffer profile

FTOS provides two pre-defined buffer profiles, one for single-queue (i.e non-QoS) applications, and one for four-queue (i.e QoS) applications.
Task Apply one of two pre-defined buffer profiles for all port pipes in the system. Command
buffer-profile global [1Q|4Q]

Mode CONFIGURATION

You must reload the system for the global buffer profile to take effect (Message 3).
Message 3 Reload After Applying Global Buffer Profile
% Info: For the global pre-defined buffer profile to take effect, please save the config and reload the system.

FTOS Behavior: After you configure buffer-profile global 1Q, Message 3 is displayed during every bootup. Only one reboot is required for the configuration to take effect; afterwards this bootup message may be ignored.

FTOS Behavior: The buffer profile does not returned to the default, 4Q, if you configure 1Q, save the running-config to the startup-config, and then delete the startup-config and reload the chassis. The only way to return to the default buffer profile is to explicitly configure 4Q, and then reload the chassis.

The buffer-profile global command fails if you have already applied a custom buffer profile on an interface.
Message 4 Global Buffer Profile Error
% Error: User-defined buffer profile already applied. Failed to apply global pre-defined buffer profile. Please remove all user-defined buffer profiles.

Similarly, when buffer-profile global is configured, you cannot not apply a buffer profile on any single interface.
Message 5 Global Buffer Profile Error
% Error: Global pre-defined buffer profile already applied. Failed to apply user-defined buffer profile on interface Gi 0/1. Please remove global pre-defined buffer profile.

If the default buffer profile (4Q) is active, FTOS displays an error message instructing you to remove the default configuration using the command no buffer-profile global.

Sample buffer profile configuration

The two general types of network environments are sustained data transfers and voice/data. Dell Networking recommends a single-queue approach for data transfers, as shown in Figure 10-9.

Debugging and Diagnostics | 235

www.dell.com | support.dell.com

Figure 10-9. Single Queue Application for S50N with Default Packet Pointers
! buffer-profile fp fsqueue-fp buffer dedicated queue0 3 queue1 3 queue2 3 queue3 3 queue4 3 queue5 3 queue6 3 queue7 3 buffer dynamic 1256 ! buffer-profile fp fsqueue-hig buffer dedicated queue0 3 queue1 3 queue2 3 queue3 3 queue4 3 queue5 3 queue6 3 queue7 3 buffer dynamic 1256 ! buffer fp-uplink stack-unit 0 port-set 0 buffer-policy fsqueue-hig buffer fp-uplink stack-unit 0 port-set 1 buffer-policy fsqueue-hig ! Interface range gi 0/1 - 48 buffer-policy fsqueue-fp FTOS#sho run int gi 0/10 ! interface GigabitEthernet 0/10 no ip address

Multicast Buffering on the S60

The S60 supports multicast buffering for all multicast traffic on the standalone buffering unit. The multicast-buffering enable CLI enables messages for multicast traffic and provides the capability to buffer multicast traffic. Use the shell command dump chg allocbuffscnt to count all buffer usage for multicast traffic in the standalone unit.
Figure 10-10. Count multicast traffic buffer usage

ALLOCBUFFSCNT.qmb0[8]: <ALLOCBUFFCNT=0x59>ALLOCBUFFSCNT.qmb0[56]: <ALLOCBUFFCNT=0x56>ALLOCBUFFSCNT.qmb0[96]: <ALLOCBUFFCNT=0x1e>ALLOCBUFFSCNT.qmb0[104]: <ALLOCBUFFCNT=0x53>ALLOCBUFFSCNT.qmb0[184]: <ALLOCBUFFCNT=0x1a>ALLOCBUFFSCNT.qmb0[200]: <ALLOCBUFFCNT=0x11>ALLOCBUFFSCNT.qmb0[248]: <ALLOCBUFFCNT=0x4f>ALLOCBUFFSCNT.qmb0[288]: <ALLOCBUFFCNT=0x59>ALLOCBUFFSCNT.qmb0[432]: <ALLOCBUFFCNT=0x1f49>

Use the pe clear shell command to clear the QMC_ERROR register. This indicates the QMC_ERROR0 register is aged out and has been flushed. The buffers are not completely used in the buffering unit, however, the packets are dropped due to the aging timer trigger.
Figure 10-11. Clear the QMC_Error register

QMC_ERROR0.qmc0[0x40680009]=0x1e: <VOQ_ARRIVAL_SATURATION=1,QUEUE_AGED_CNT=7>

236

Debugging and Diagnostics

The S60 switch has two ASICS; each physical port on the ASICS has 8 virtual queues for unicast traffic. There are a total of 53 ports, including the CPU ports which result in 440 total virtual queues. Virtual queues 0 through 431 are for unicast traffic, while virtual queues 432 through 440 are used for multicast traffic on the entire system. Each of the 8 virtual queues are used for each corresponding CoS value.
Figure 10-12. Viewing multicast traffic on virtual queues.

gport:(0x640001b0) cos=0 GreenAcceptedPkts=219055937 GreenAcceptedBytes=288715724966 GreenMarkedPkts=0 GreenMarkedBytes=0 GreenDiscardPkts=0 GreenDiscardBytes=0 YellowAcceptedPkts=0 YellowAcceptedBytes=0 YellowMarkedPkts=0 YellowMarkedBytes=0 YellowDiscardPkts=0 YellowDiscardBytes=0 RedAcceptedPkts=0 RedAcceptedBytes=0 RedMarkedPkts=0 RedMarkedBytes=0 RedDiscardPkts=0 RedDiscardBytes=0 BlackAcceptedPkts=0 BlackAcceptedBytes=0 BlackMarkedPkts=0 BlackMarkedBytes=0 BlackDiscardPkts=0 BlackDiscardBytes=0 NonWredDroppedPkts=0 NonWredDroppedBytes=0 OverSubTtlDrpdPkts=0 OverSubTtlDrpdBytes=0 OverSubGuarDrpdPkts=176139 OverSubGuarDrpdBytes=232151202 <TAIL drops seen after maximum buffer utilization> DequeuedPkts=146366795 DequeuedBytes=191447767860 Gport nCos Type Modid Port Basequeue -------------------------------------------------------------------------------------------0x640001b0 8 Subscriber Multicast Queue Group 10000 0 432 cos mode weight kbits_sec_min kbits_sec_max flags 0 BCM_COSQ_AF0 1 1000000 0 1 1 BCM_COSQ_AF0 2 1000000 0 1 2 BCM_COSQ_AF0 4 1000000 0 1 3 BCM_COSQ_AF0 8 1000000 0 1 4 BCM_COSQ_AF0 127 1000000 0 1 5 BCM_COSQ_AF0 127 1000000 0 1 6 BCM_COSQ_AF0 127 1000000 0 1 7 expedited forwarding 0 0 0 0

Troubleshooting packet loss

The show hardware stack-unit command is intended primarily to troubleshoot packet loss.
show hardware stack-unit cpu data-plane statistics show hardware stack-unit cpu party-bus statistics show hardware stack-unit 0-11 drops unit 0-1 port 0-49 show hardware stack-unit 0-11 stack-port 48-51 show hardware stack-unit 0-11 unit 0-1 {counters | details | port-stats [detail] | register | execute-shell-cmd | ipmc-replication | table-dump}: show hardware {layer2| layer3} {e.g. acl |in acl} stack-unit 0-11 port-set 0-1 show hardware layer3 qos stack-unit 0-7 port-set 0-1

Debugging and Diagnostics | 237

www.dell.com | support.dell.com

show hardware ipv6 {e.g.-acl |in-acl} stack-unit 0-11 port-set 0-1 show hardware system-flow layer2 stack-unit 0-11 port-set 0-1 [counters] clear hardware stack-unit 0-11 counters clear hardware stack-unit 0-11 unit 0-1 counters clear hardware stack-unit 0-11 cpu data-plane statistics clear hardware stack-unit 0-11 cpu party-bus statistics clear hardware stack-unit 0-11 stack-port 48-51

Displaying Drop Counters

The show hardware stack-unit 011 drops [unit 01 [port 049]] command assists in identifying which stack unit, port pipe, and port is experiencing internal drops, as shown in Figure 10-13 and Figure 10-14.
Figure 10-13. Displaying Drop Counter Statistics

FTOS#show hardware stack-unit 0 drops UNIT No: 0 Total Ingress Drops :0 Total IngMac Drops :0 Total Mmu Drops :0 Total EgMac Drops :0 Total Egress Drops :0 UNIT No: 1 Total Ingress Drops :0 Total IngMac Drops :0 Total Mmu Drops :0 Total EgMac Drops :0 Total Egress Drops :0 FTOS#show hardware stack-unit 0 drops unit 0 Port# :Ingress Drops :IngMac Drops :Total Mmu Drops :EgMac Drops :Egress Drops 1 0 0 0 0 0 2 0 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 6 0 0 0 0 0 7 0 0 0 0 0 8 0 0 0 0 0

Display drop counters with the show hardware stack-unit drops unit port command:

238

Debugging and Diagnostics

Figure 10-14.

Displaying Drop Counters

0 drops unit 0 port 1 : : : : : : : : : 30 0 0 0 0 14 0 16 0

FTOS#show hardware stack-unit --- Ingress Drops --Ingress Drops IBP CBP Full Drops PortSTPnotFwd Drops IPv4 L3 Discards Policy Discards Packets dropped by FP (L2+L3) Drops Port bitmap zero Drops Rx VLAN Drops --- Ingress MAC counters--Ingress FCSDrops Ingress MTUExceeds --- MMU Drops HOL DROPS TxPurge CellErr Aged Drops ---

: 0 : 0

: 0 : 0 : 0

--- Egress MAC counters--Egress FCS Drops

: 0 ---

--- Egress FORWARD PROCESSOR Drops IPv4 L3UC Aged & Drops : 0 TTL Threshold Drops : 0 INVALID VLAN CNTR Drops : 0 L2MC Drops : 0 PKT Drops of ANY Conditions : 0 Hg MacUnderflow : 0 TX Err PKT Counter : 0

Dataplane Statistics
The show hardware stack-unit cpu data-plane statistics command provides insight into the packet types coming to the CPU. As shown in Figure 10-15, the command output has been augmented, providing detailed RX/TX packet statistics on a per-queue basis. The objective is to see whether CPU-bound traffic is internal (so-called party bus or IPC traffic) or network control traffic, which the CPU must process.

Debugging and Diagnostics | 239

www.dell.com | support.dell.com

Figure 10-15.

Displaying Dataplane Statistics

FTOS#show hardware stack-unit 2 cpu data-plane statistics bc pci driver statistics for device: rxHandle :0 noMhdr :0 noMbuf :0 noClus :0 recvd :0 dropped :0 recvToNet :0 rxError :0 rxDatapathErr :0 rxPkt(COS0) :0 rxPkt(COS1) :0 rxPkt(COS2) :0 rxPkt(COS3) :0 rxPkt(COS4) :0 rxPkt(COS5) :0 rxPkt(COS6) :0 rxPkt(COS7) :0 rxPkt(UNIT0) :0 rxPkt(UNIT1) :0 rxPkt(UNIT2) :0 rxPkt(UNIT3) :0 transmitted :0 txRequested :0 noTxDesc :0 txError :0 txReqTooLarge :0 txInternalError :0 txDatapathErr :0 txPkt(COS0) :0 txPkt(COS1) :0 txPkt(COS2) :0 txPkt(COS3) :0 txPkt(COS4) :0 txPkt(COS5) :0 txPkt(COS6) :0 txPkt(COS7) :0 txPkt(UNIT0) :0 txPkt(UNIT1) :0 txPkt(UNIT2) :0 txPkt(UNIT3) :0

The show hardware stack-unit cpu party-bus statistics command displays input and output statistics on the party bus, which carries inter-process communication traffic between CPUs, as shown in Figure 10-16.
Figure 10-16. Displaying Party Bus Statistics

FTOS#sh hardware stack-unit 2 cpu party-bus statistics Input Statistics: 27550 packets, 2559298 bytes 0 dropped, 0 errors Output Statistics: 1649566 packets, 1935316203 bytes 0 errors

240

Debugging and Diagnostics

Displaying Stack Port Statistics

The show hardware stack-unit stack-port command displays input and output statistics for a stack-port interface, as shown in Figure 10-17.
Figure 10-17. Displaying Stack Unit Statistics

FTOS#show hardware stack-unit 2 stack-port 49 Input Statistics: 27629 packets, 3411731 bytes 0 64-byte pkts, 27271 over 64-byte pkts, 207 over 127-byte pkts 17 over 255-byte pkts, 56 over 511-byte pkts, 78 over 1023-byte pkts 0 Multicasts, 5 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 1649714 packets, 1948622676 bytes, 0 underruns 0 64-byte pkts, 27234 over 64-byte pkts, 107970 over 127-byte pkts 34 over 255-byte pkts, 504838 over 511-byte pkts, 1009638 over 1023-byte pkts 0 Multicasts, 0 Broadcasts, 1649714 Unicasts 0 throttles, 0 discarded, 0 collisions Rate info (interval 45 seconds): Input 00.00 Mbits/sec, 2 packets/sec, 0.00% of line-rate Output 00.06 Mbits/sec, 8 packets/sec, 0.00% of line-rate FTOS#

Displaying Stack Member Counters

The show hardware stack-unit 011 {counters | details | port-stats [detail] | register} command displays internal receive and transmit statistics, based on the selected command option. A sample of the output is shown for the counters option in Figure 10-18.
Figure 10-18.
RIPC4.ge0 RUC.ge0 RDBGC0.ge0 RDBGC1.ge0 RDBGC5.ge0 RDBGC7.ge0 GR64.ge0 GR127.ge0 GR255.ge0 GRPKT.ge0 GRBYT.ge0 GRMCA.ge0 GRBCA.ge0 GT64.ge0 GT127.ge0 GT255.ge0 GT511.ge0 GTPKT.ge0 GTBCA.ge0 GTBYT.ge0 RUC.cpu0 TDBGC6.cpu0

Displaying Stack Unit Counters

: : : : : : : : : : : : : : : : : : : : : : 1,202 1,224 34 366 16 18 5,176 1,566 4 1,602 117,600 366 12 4 964 4 1 973 1 71,531 972 1,584 +1,202 +1,217 +24 +235 +12 +12 +24 +1,433 +4 +1,461 +106,202 +235 +9 +3 +964 +4 +1 +972 +1 +71,467 +971 +1,449=

Debugging and Diagnostics | 241

www.dell.com | support.dell.com

Application core dumps

Application core dumps are disabled by default. A core dump file can be very large. Due to memory requirements the file can only be sent directly to an FTP server. It is not stored on the local flash. Enable full application core dumps with the following:
Task Enable RPM core dumps and specify the shutdown mode. Command Syntax
logging coredump server

Command Mode CONFIGURATION

Undo this command using the no logging coredump server.

Mini core dumps

FTOS supports mini core dumps on the for application and kernel crashes. The mini core dump apply to Master, Standby and Member units. Application and kernel mini core dumps are always enabled. The mini core dumps contain the stack space and some other very minimal information that can be used to debug a crash. These files are small files and are written into flash until space is exhausted. When the flash is full, the write process is stopped.
Note: If the Hardware watchdog timer is enabled and it triggers, no kernel core dump is generated because this is considered a power-off. In all other kernel crashes, a kernel mini core dump is be generated. When an application crashes, only the application coredump is generated.

A mini core dump contains critical information in the event of a crash. Mini core dump files are located in flash:/ (root dir). The application mini core file name format is f10StkUnit<Stack_unit_no>.<Application name>.acore.mini.txt. The kernel mini core file name format is f10StkUnit<Stack_unit_no>.kcore.mini.txt. Sample files names are shown in Figure 10-19 and sample file text is shown in Figure 10-20.

242

Debugging and Diagnostics

Figure 10-19.

Mini application core file naming example

FTOS#dir Directory of flash: 1 2 3 4 5 6 7 8 9 10 11 12 13 drwdrwx drwd---rw-rw-rw-rw-rw-rw-rw-rw-rw16384 1536 512 512 8693 8693 156 156 156 156 156 156 156 Jan Sep Aug Aug Sep Sep Aug Aug Aug Aug Aug Aug Aug 01 03 07 07 03 03 28 28 28 28 31 29 31 1980 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 00:00:00 16:51:02 13:05:58 13:06:00 16:50:56 16:44:22 16:16:10 17:17:24 18:25:18 19:07:36 16:18:50 14:28:34 16:14:56 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 +00:00 . .. TRACE_LOG_DIR ADMIN_DIR startup-config startup-config.bak f10StkUnit0.mrtm.acore.mini.txt f10StkUnit0.vrrp.acore.mini.txt f10StkUnit0.sysd.acore.mini.txt f10StkUnit0.frrp.acore.mini.txt f10StkUnit2.sysd.acore.mini.txt f10StkUnit0.ipm1.acore.mini.txt f10StkUnit0.acl.acore.mini.txt

flash: 3104256 bytes total (2959872 bytes free) FTOS#

When a member or standby unit crashes, the mini core file gets uploaded to master unit. When the master unit crashes, the mini core file is uploaded to new master.
Figure 10-20. Mini core text file example

VALID MAGIC ------------------------PANIC STRING ----------------panic string is :<null> ----------------------STACK TRACE START--------------0035d60c <f10_save_mmu+0x120>: 00274f8c <panic+0x144>: 0024e2b0 <db_fncall+0x134>: 0024dee8 <db_command+0x258>: 0024d9c4 <db_command_loop+0xc4>: 002522b0 <db_trap+0x158>: 0026a8d0 <mi_switch+0x1b0>: 0026a00c <bpendtsleep>: ------------------------STACK TRACE END------------------------------------------FREE MEMORY--------------uvmexp.free = 0x2312

The panic string contains key information regarding the crash. Several panic string types exist, and they are displayed in regular english text to enable easier understanding of the crash cause.

Debugging and Diagnostics | 243

244 www.dell.com | support.dell.com

Debugging and Diagnostics

Skippy812

11
Dynamic Host Configuration Protocol (DHCP)
Dynamic Host Configuration Protocol (DHCP) is available on platforms: e c s

This chapter contains the following sections: Protocol Overview Implementation Information Configuration Tasks Configure the System to be a DHCP Server Configure the System to be a Relay Agent Configure the System for User Port Stacking Configure Secure DHCP

Protocol Overview
Dynamic Host Configuration Protocol (DHCP) is an application layer protocol that dynamically assigns IP addresses and other configuration parameters to network end-stations (hosts) based on configuration policies determined by network administrators. DHCP: relieves network administrators of manually configuring hosts, which can be a tedious and error-prone process when hosts often join, leave, and change locations on the network. reclaims IP addresses that are no longer in use to prevent address exhaustion.

DHCP is based on a client-server model. A host discovers the DHCP server and requests an IP address, and the server either leases or permanently assigns one. There are three types of devices that are involved in DHCP negotiation: DHCP Servera network device offering configuration parameters to the client DHCP Clienta network device requesting configuration parameters from the server Relay agentan intermediary network device that passes DHCP messages between the client and server when the server is not on the same subnet as the host

Dynamic Host Configuration Protocol (DHCP) | 245

www.dell.com | support.dell.com

DHCP Packet Format and Options

DHCP uses UDP as its transport protocol. The server listens on port 67 and transmits to port 68; the client listens on port 68 and transmits to port 67. The configuration parameters are carried as options in the DHCP packet in Type, Length, Value (TLV) format; many options are specified in RFC 2132. To limit the number of parameters that servers must provide, hosts specify the parameters that they require, and the server sends only those; some common options are given in Table 11-1.
Figure 11-1.
op htype

DHCP Packet Format

hlen hops xid secs flags ciaddr yiaddr siaddr giaddr chaddr sname file options

Code

Length

Value

Table 11-1. Option Subnet Mask Router

Common DHCP Options Code 1 3 6 15 51 53 Description Specifies the clients subnet mask. Specifies the router IP addresses that may serve as the clients default gateway. Specifies the DNS servers that are available to the client. Specifies the domain name that clients should use when resolving hostnames via DNS. Specifies the amount of time that the client is allowed to use an assigned IP address. 1: DHCPDISCOVER 2: DHCPOFFER 3: DHCPREQUEST 4: DHCPDECLINE 5: DHCPACK 6: DHCPNACK 7: DHCPRELEASE 8: DHCPINFORM Clients use this option to tell the server which parameters it requires. It is a series of octets where each octet is DHCP option code. Specifies the amount of time after the IP address is granted that the client attempts to renew its lease with the original server. Specifies the amount of time after the IP address is granted that the client attempts to renew its lease with any server, if the original server does not respond. Signals the last option in the DHCP packet.

Domain Name Server Domain Name IP Address Lease Time DHCP Message Type

Parameter Request List Renewal Time Rebinding Time

55 58 59

End

255

246

Dynamic Host Configuration Protocol (DHCP)

Assigning an IP Address using DHCP

When a client joins a network: 1. The client initially broadcasts a DHCPDISCOVER message on the subnet to discover available DHCP servers. This message includes the parameters that the client requires and might include suggested values for those parameters. 2. Servers unicast or broadcast a DHCPOFFER message in response to the DHCPDISCOVER that offers to the client values for the requested parameters. Multiple servers might respond to a single DHCPDISCOVER; the client might wait a period of time and then act on the most preferred offer. 3. The client broadcasts a DHCPREQUEST message in response to the offer, requesting the offered values. 4. Upon receiving a DHCPREQUEST, the server binds the clients unique identifier (the hardware address plus IP address) to the accepted configuration parameters and stores the data in a database called a binding table. The server then broadcasts a DHCPACK message, which signals to the client that it may begin using the assigned parameters. 5. When the client leaves the network, or the lease time expires, returns its IP address to the server in a DHCPRELEASE message. There are additional messages that are used in case the DHCP negotiation deviates from the process previously described and shown in the illustration below. DHCPDECLINEA client sends this message to the server in response to a DHCPACK if the configuration parameters are unacceptable, for example, if the offered address is already in use. In this case, the client starts the configuration process over by sending a DHCPDISCOVER. DHCPINFORMA client uses this message to request configuration parameters when it assigned an IP address manually rather than with DHCP. The server responds by unicast. DHCPNAKA server sends this message to the client if it is not able to fulfill a DHCPREQUEST, for example if the requested address is already in use. In this case, the client starts the configuration process over by sending a DHCPDISCOVER.

Client
1. DHCPDISCOVER

Relay Agent
2. DHCPOFFER 3. DHCPREQUEST 4. DHCPACK 5. DHCPRELEASE

Server

Dynamic Host Configuration Protocol (DHCP) | 247

www.dell.com | support.dell.com

Implementation Information
The Dell Force10 implementation of DHCP is based on RFC 2131 and RFC 3046. IP Source Address Validation is a sub-feature of DHCP Snooping; FTOS uses ACLs internally to implement this feature and as such, you cannot apply ACLs to an interface which has IP Source Address Validation. If you configure IP Source Address Validation on a member port of a VLAN and then attempt to apply a access list to the VLAN, FTOS displays the first line in Message 1. If you first apply an ACL to a VLAN and then attempt enable IP Source Address Validation on one of its member ports, FTOS displays the second line in Message 1.

Message 1 DHCP Snooping with VLAN ACL Compatibility Error

% Error: Vlan member has access-list configured. % Error: Vlan has an access-list configured.

Note: If DHCP snooping is enabled globally and any L2 port is configured, any IP ACL,MAC ACL, or DHCP Source-Address validation ACL won't block DHCP packets .

FTOS provides 40K entries that can be divided between leased addresses and excluded addresses. By extension, the maximum number of pools you can configure depends on the on the subnet mask that you give to each pool. For example, if all pools were configured for a /24 mask, the total would be 40000/253 (approximately 158). If the subnet is increased, more pools can be configured. The maximum subnet that can be configured for a single pool is /17. FTOS displays an error message for configurations that exceed the allocated memory. E-Series supports 16K DHCP Snooping entries across 500 VLANs. C-Series, S-Series (S25/S50), S55, S60 and S4810 support 4K DHCP Snooping entries. All platforms support Dynamic ARP Inspection on 16 VLANs per system. Refer to Dynamic ARP Inspection.
Note: If the DHCP server is located on the ToR and the VLTi (ICL) is down due to a failed link when a VLT node is rebooted in JumpStart mode, it will not be able to reach the DHCP server, resulting in BMP failure.

Configuration Tasks
Configure the System to be a DHCP Server Configure the System to be a Relay Agent Configure Secure DHCP

248

Dynamic Host Configuration Protocol (DHCP)

Configure the System to be a DHCP Server

Configure the System to be a DHCP Server is supported only on platforms: , and . ,

c and s (S25/S50),

A DHCP server is a network device that has been programmed to provide network configuration parameters to clients upon request. Servers typically serve many clients, making host management much more organized and efficient. The key responsibilities of DHCP servers are: 1. Address Storage and Management: DHCP servers are the owners of the addresses used by DHCP clients.The server stores the addresses and manages their use, keeping track of which addresses have been allocated and which are still available. 2. Configuration Parameter Storage and Management: DHCP servers also store and maintain other parameters that are sent to clients when requested. These parameters specify in detail how a client is to operate. 3. Lease Management: DHCP servers use leases to allocate addresses to clients for a limited time. The DHCP server maintains information about each of the leases, including lease length. 4. Responding To Client Requests: DHCP servers respond to different types of requests from clients, primarily, granting, renewing, and terminating leases. 5. Providing Administration Services: The DHCP server includes functionality that allows an administrator to implement policies that govern how DHCP performs its other tasks.

Configuration Tasks
To configure DHCP, an administrator must first set up a DHCP server and provide it with configuration parameters and policy information including IP address ranges, lease length specifications, and configuration data that DHCP hosts need. Configuring the Dell Force10 system to be a DHCP server is a 3-step process: 1. Configure the Server for Automatic Address Allocation 2. Specify a Default Gateway 3. Enable DHCP Server

Related Configuration Tasks

Configure a Method of Hostname Resolution Create Manual Binding Entries Debug DHCP server DHCP Clear Commands

Dynamic Host Configuration Protocol (DHCP) | 249

www.dell.com | support.dell.com

Configure the Server for Automatic Address Allocation

This feature is available on

c and s (S25/S50),

, and

platforms only.

Automatic Address Allocation is an address assignment method by which the DHCP server leases an IP address to a client from a pool of available addresses.

Create an IP Address Pool

An address pool is a range of IP addresses that may be assigned by the DHCP server. Address pools are indexed by subnet number. To create an address pool:
Step 1 2 3 Task Access the DHCP server CLI context. Create an address pool and give it a name. Specify the range of IP addresses from which the DHCP server may assign addresses. network is the subnet address. prefix-length specifies the number of bits used for the network portion of the address you specify. Display the current pool configuration. Command Syntax
ip dhcp server pool name network network /prefix-length

Command Mode CONFIGURATION DHCP DHCP <POOL>

Prefix-length Range: 17 to 31

show config

DHCP <POOL>

Once an IP address is leased to a client, only that client may release the address. FTOS performs a IP + MAC source address validation to ensure that no client can release another clients address. This is a default behavior and is separate from IP+MAC Source Address Validation.

Exclude Addresses from the Address Pool

The DHCP server assumes that all IP addresses in a DHCP address pool are available for assigning to DHCP clients. You must specify the IP address that the DHCP server should not assign to clients.
Task Exclude an address range from DHCP assignment. The exclusion applies to all configured pools. Command Syntax
excluded-address

Command Mode DHCP

Specify an Address Lease Time

Task Specify an address lease time for the addresses in a pool. Command Syntax
lease {days [hours] [minutes] | infinite} Default: 24 hours

CommandMode DHCP <POOL>

250

Dynamic Host Configuration Protocol (DHCP)

Specify a Default Gateway

The IP address of the default router should be on the same subnet as the client.
Task Specify default gateway(s) for the clients on the subnet, in order of preference. Command Syntax
default-router address

Command Mode DHCP <POOL>

Enable DHCP Server

This feature is available on

c and s (S25/S50),

, and

platforms only.

The DHCP server is disabled by default.

Step 1 2 Task Enter the DHCP command-line context. Enable DHCP server. Command Syntax
ip dhcp server no disable

Command Mode CONFIGURATION DHCP

Default: Disabled 3 Display the current DHCP configuration.

show config

DHCP

In the illustration below, an IP phone is powered by PoE and has acquired an IP address from the Dell Force10 system, which is advertising LLDP-MED. The leased IP address is displayed using show ip dhcp binding, and confirmed with show lldp neighbors.

DNS Server

7/1

Relay Agent

Dynamic Host Configuration Protocol (DHCP) | 251

www.dell.com | support.dell.com

Configure a Method of Hostname Resolution

Dell Force10 systems are capable of providing DHCP clients with parameters for two methods of hostname resolution.

Address Resolution using DNS

A domain is a group of networks. DHCP clients query DNS IP servers when they need to correlate host names to IP addresses.
Step 1 2 Task Create a domain. Specify in order of preference the DNS servers that are available to a DHCP client. Command Syntax
domain-name name dns-server address

Command Mode DHCP <POOL> DHCP <POOL>

Address Resolution using NetBIOS WINS

Windows Internet Naming Service (WINS) is a name resolution service that Microsoft DHCP clients use to correlate host names to IP addresses within a group of networks. Microsoft DHCP clients can be one of four types of NetBIOS nodes: broadcast, peer-to-peer, mixed, or hybrid.
Step 1 Task Specify the NetBIOS Windows Internet Naming Service (WINS) name servers, in order of preference, that are available to Microsoft Dynamic Host Configuration Protocol (DHCP) clients. Specify the NetBIOS node type for a Microsoft DHCP client. Dell Force10 recommends specifying clients as hybrid. Command Syntax
netbios-name-server address

Command Mode DHCP <POOL>

netbios-node-type type

DHCP <POOL>

252

Dynamic Host Configuration Protocol (DHCP)

Create Manual Binding Entries

An address binding is a mapping between the IP address and Media Access Control (MAC) address of a client. The DHCP server assigns the client an available IP address automatically, and then creates a entry in the binding table. However, the administrator can manually create an entry for a client; manual bindings are useful when you want to guarantee that a particular network device receives a particular IP address. Manual bindings can be considered single-host address pools. There is no limit on the number of manual bindings, but you can only configure one manual binding per host.
Note: FTOS does not prevent you from using a network IP as a host IP; be sure to not use a network IP as a host IP.

To create a manual binding:

Step 1 2 3 Task Create an address pool Specify the client IP address. Specify the client hardware address. hardware-address is the client MAC address. type is the protocol of the hardware platform. The default protocol is Ethernet. Command Syntax
pool name host address hardware-address hardware-address type

Command Mode DHCP DHCP <POOL> DHCP <POOL>

Debug DHCP server

Task Display debug information for DHCP server. Command Syntax
debug ip dhcp server [events | packets]

Command Mode EXEC Privilege

DHCP Clear Commands

Task Clear DHCP binding entries for the entire binding table. Clear a DHCP binding entry for an individual IP address. Clear a DHCP address conflict. Clear DHCP server counters. Clear the DHCP binding table. Command Syntax
clear ip dhcp binding clear ip dhcp binding ip address clear ip dhcp conflict clear ip dhcp server statistics clear ip dhcp snooping

Command Mode EXEC Privilege EXEC Privilege EXEC Privilege EXEC Privilege EXEC Privilege

Dynamic Host Configuration Protocol (DHCP) | 253

www.dell.com | support.dell.com

Configure the System to be a Relay Agent

The following feature is available on platforms:

ces

DHCP clients and servers request and offer configuration information via broadcast DHCP messages. Routers do not forward broadcasts, so if there are no DHCP servers on the subnet, the client does not receive a response to its request and therefore cannot access the network. You can configure an interface on the Dell Force10 system to relay the DHCP messages to a specific DHCP server using the command ip helper-address dhcp-address from INTERFACE mode, as shown in the illustration below. Specify multiple DHCP servers by entering the ip helper-address dhcp-address command multiple times. When ip helper-address is configured, the system listens for DHCP broadcast messages on port 67. The system rewrites packets received from the client and forwards it via unicast; the system rewrites the destination IP address and writes its own address as the relay device. Responses from the server are unicast back to the relay agent on port 68 and the relay agent rewrites the destination address and forwards the packet to the client subnet via broadcast.
Note: DHCP Relay is not available on Layer 2 interfaces and VLANs.

HCP Relay Device

DHCP Server 10.11.2.5

Broadcast Source IP : 10.11.1.5 Destination IP: 255.255.255.255 Source Port: 67 Destination Port: 68

Unicast Source IP : 10.11.1.5 Destination IP: 10.11.0.3 Source Port: 67 Destination Port: 68

DHCP Server

Unicast

10.11.1.5

1/4
Broadcast Source IP : 0.0.0.0 Destination IP: 255.255.255.255 Source Port: 68 Destination Port: 67 Relay Agent Address: 0.0.0.0

1/3
Unicast Source IP : 10.11.1.3 Destination IP: 10.11.1.5 Source Port: 67 Destination Port: 67 Relay Agent Address: 10.11.0.3

R1(conf-if-gi-1/3)#show config ! interface GigabitEthernet 1/3 ip address 10.11.0.3/24 ip helper-address 10.11.1.5 ip helper-address 10.11.2.5 no shutdown

DHCP 001

254

Dynamic Host Configuration Protocol (DHCP)

To view the ip helper-address configuration for an interface, use the command show ip interface from EXEC privilege mode, as shown in the following example.
R1_E600#show ip int gig 1/3 GigabitEthernet 1/3 is up, line protocol is down Internet address is 10.11.0.1/24 Broadcast address is 10.11.0.255 Address determined by user input IP MTU is 1500 bytes Helper address is 192.168.0.1 192.168.0.2 Directed broadcast forwarding is disabled Proxy ARP is enabled Split Horizon is enabled Poison Reverse is disabled ICMP redirects are not sent ICMP unreachables are not sent

Configure the System for User Port Stacking

When you set the DHCP offer on the DHCP server, you can set the stacking-option variable to provide the stack-port detail so a stack can be formed when the units are connected.

Configure Secure DHCP

The following feature is available on platforms: c,

e, s

and z except where noted.

DHCP as defined by RFC 2131 provides no authentication or security mechanisms. Secure DHCP is a suite of features that protects networks that use dynamic address allocation from spoofing and attacks. Option 82 DHCP Snooping Dynamic ARP Inspection Source Address Validation

Option 82
RFC 3046 (Relay Agent Information option, or Option 82) is used for class-based IP address assignment. The code for the Relay Agent Information option is 82, and is comprised of two sub-options, Circuit ID and Remote ID. Circuit ID is the interface on which the client-originated message is received. Remote ID identifies the host from which the message is received. The value of this sub-option is the MAC address of the relay agent that adds Option 82.

Dynamic Host Configuration Protocol (DHCP) | 255

www.dell.com | support.dell.com

The DHCP relay agent inserts Option 82 before forwarding DHCP packets to the server. The server can use this information to: track the number of address requests per relay agent; restricting the number of addresses available per relay agent can harden a server against address exhaustion attacks. associate client MAC addresses with a relay agent to prevent offering an IP address to a client spoofing the same MAC address on a different relay agent. assign IP addresses according to the relay agent. This prevents generating DHCP offers in response to requests from an unauthorized relay agent.

The server echoes the option back to the relay agent in its response, and the relay agent can use the information in the option to forward a reply out the interface on which the request was received rather than flooding it on the entire VLAN. The relay agent strips Option 82 from DHCP responses before forwarding them to the client.
Task Insert Option 82 into DHCP packets. For routers between the relay agent and the DHCP server, enter the trust-downstream option. Configure the system to enable remote-id string in Option 82. Command Syntax
ip dhcp relay information-option

Command Mode CONFIGURATION

[trust-downstream]
ip dhcp relay information-option

CONFIGURATION

[remote-id]

DHCP Snooping
DHCP Snooping protects networks from spoofing. In the context of DHCP Snooping, all ports are either trusted or untrusted. By default, all ports are untrusted. Trusted ports are ports through which attackers cannot connect. Manually configure ports connected to legitimate servers and relay agents as trusted. When DHCP Snooping is enabled, the relay agent builds a binding tableusing DHCPACK messages containing the client MAC address, IP addresses, IP address lease time, port, VLAN ID, and binding type. Every time the relay agent receives a DHCPACK on an trusted port, it adds an entry to the table. The relay agent then checks all subsequent DHCP client-originated IP traffic (DHCPRELEASE, DHCPNACK, and DHCPDECLINE) against the binding table to ensure that the MAC-IP address pair is legitimate, and that the packet arrived on the correct port; packets that do not pass this check are forwarded to the server for validation. This check-point prevents an attacker from spoofing a client and declining or releasing the real clients address. Server-originated packets (DHCPOFFER, DHCPACK, DHCPNACK) that arrive on an untrusted port are also dropped. This check-point prevents an attacker from impostering as a DHCP server to facilitate a man-in-the-middle attack.

256

Dynamic Host Configuration Protocol (DHCP)

Binding table entries are deleted when a lease expires, or the relay agent encounters a DHCPRELEASE, DHCPNACK, DHCPDECLINE.
FTOS Behavior: Introduced in FTOS version 7.8.1.0, DHCP Snooping was available for Layer 3 only and dependent on DHCP Relay Agent (ip helper-address). FTOS version 8.2.1.0 extends DHCP Snooping to Layer 2, and you do not have to enable relay agent to snoop on Layer 2 interfaces. FTOS Behavior: Binding table entries are deleted when a lease expires or when the relay agent encounters a DHCPRELEASE. Starting with FTOS Release 8.2.1.2, line cards maintain a list of snooped VLANs. When the binding table is exhausted, DHCP packets are dropped on snooped VLANs, while these packets are forwarded across non-snooped VLANs. Since DHCP packets are dropped, no new IP address assignments are made. However, DHCPRELEASE and DHCPDECLINE packets are allowed so that the DHCP snooping table can decrease in size. Once the table usage falls below the maximum limit of 4000 entries, new IP address assignments are allowed.

Note: DHCP server packets will be dropped on all untrusted interfaces of a system configured for DHCP snooping. To prevent these packets from being dropped, configure ip dhcp snooping trust on the server-connected port.

Enable DCHP snooping

Step 1 2 3 Task Enable DHCP Snooping globally. Specify ports connected to DHCP servers as trusted. Enable DHCP Snooping on a VLAN. Command Syntax
ip dhcp snooping ip dhcp snooping trust ip dhcp snooping vlan

Command Mode CONFIGURATION INTERFACE CONFIGURATION

Add a static entry in the binding table

Task Add a static entry in the binding table. Command Syntax
ip dhcp snooping binding mac

Command Mode EXEC Privilege

Clear the binding table

Task Delete all of the entries in the binding table Command Syntax
clear ip dhcp snooping binding

Command Mode EXEC Privilege

Display the contents of the binding table

Task Display the contents of the binding table. Command Syntax
show ip dhcp snooping

Command Mode EXEC Privilege

Dynamic Host Configuration Protocol (DHCP) | 257

www.dell.com | support.dell.com

View the DHCP Snooping statistics with the show ip dhcp snooping command.
FTOS#show ip dhcp snooping IP IP IP IP DHCP DHCP DHCP DHCP Snooping Snooping Mac Verification Relay Information-option Relay Trust Downstream : : : : Enabled. Disabled. Disabled. Disabled.

Database write-delay (In minutes)

: 0

DHCP packets information Relay Information-option packets Relay Trust downstream packets Snooping packets

: 0 : 0 : 0

Packets received on snooping disabled L3 Ports : 0 Snooping packets processed on L2 vlans : 142 DHCP Binding File Details Invalid File Invalid Binding Entry Binding Entry lease expired List of Trust Ports List of DHCP Snooping Enabled Vlans List of DAI Trust ports

: 0 : 0 : 0 :Te 0/49 :Vl 10 :Te 0/49

Drop DHCP packets on snooped VLANs only

Binding table entries are deleted when a lease expires or the relay agent encounters a DHCPRELEASE. Starting with FTOS Release 8.2.1.1, line cards maintain a list of snooped VLANs. When the binding table fills, DHCP packets are dropped only on snooped VLANs, while such packets will be forwarded across non-snooped VLANs. Since DHCP packets are dropped, no new IP address assignments are made. However, DHCP Release and Decline packets are allowed so that the DHCP snooping table can decrease in size. Once the table usage falls below the max limit of 4000 entries, new IP address assignments are allowed. View the number of entries in the table with the show ip dhcp snooping binding command. This output displays the snooping binding table created using the ACK packets from the trusted port.
FTOS#show ip dhcp snooping binding Codes : S - Static D - Dynamic

IP Address MAC Address Expires(Sec) Type VLAN Interface ======================================================================== 10.1.1.251 00:00:4d:57:f2:50 172800 D Vl 10 Gi 0/2 10.1.1.252 00:00:4d:57:e6:f6 172800 D Vl 10 Gi 0/1 10.1.1.253 00:00:4d:57:f8:e8 172740 D Vl 10 Gi 0/3 10.1.1.254 00:00:4d:69:e8:f2 172740 D Vl 10 Te 0/50 Total number of Entries in the table : 4

258

Dynamic Host Configuration Protocol (DHCP)

Dynamic ARP Inspection

Dynamic ARP inspection prevents ARP spoofing by forwarding only ARP frames that have been validated against the DHCP binding table. ARP is a stateless protocol that provides no authentication mechanism. Network devices accept ARP requests and replies from any device, and ARP replies are accepted even when no request was sent. If a client receives an ARP message for which a relevant entry already exists in its ARP cache, it overwrites the existing entry with the new information. The lack of authentication in ARP makes it vulnerable to spoofing. ARP spoofing is a technique attackers use to inject false IP-to-MAC mappings into the ARP cache of a network device. It is used to launch man-in-the-middle (MITM), and denial-of-service (DoS) attacks, among others. A spoofed ARP message is one in which the MAC address in the sender hardware address field and the IP address in the sender protocol field are strategically chosen by the attacker. For example, in an MITM attack, the attacker sends a client an ARP message containing the attackers MAC address and the gateways IP address. The client then thinks that the attacker is the gateway, and sends all internet-bound packets to it. Likewise, the attacker sends the gateway an ARP message containing the attackers MAC address and the clients IP address. The gateway then thinks that the attacker is the client and forwards all packets addressed to the client to it. As a result, the attacker is able to sniff all packets to and from the client. Other attacks using ARP spoofing include: broadcastan attacker can broadcast an ARP reply that specifies FF:FF:FF:FF:FF:FF as the gateways MAC address, resulting in all clients broadcasting all internet-bound packets. MAC floodingan attacker can send fraudulent ARP messages to the gateway until the ARP cache is exhausted, after which, traffic from the gateway is broadcast.

Dynamic Host Configuration Protocol (DHCP) | 259

www.dell.com | support.dell.com

denial of servicean attacker can send a fraudulent ARP messages to a client to associate a false MAC address with the gateway address, which would blackhole all internet-bound packets from the client.

Note: DAI uses entries in the L2SysFlow CAM region, a sub-region of SystemFlow. One CAM entry is required for every DAI-enabled VLAN. You can enable DAI on up to 16 VLANs on a system. However, the ExaScale default CAM profile allocates only 9 entries to the L2SysFlow region for DAI. You can configure 10 to 16 DAI-enabled VLANs by allocating more CAM space to the L2SysFlow region before enabling DAI. SystemFlow has 102 entries by default. This region is comprised of two sub-regions: L2Protocol and L2SystemFlow. L2Protocol has 87 entries, and L2SystemFlow has 15 entries. Six L2SystemFlow entries are used by Layer 2 protocols, leaving 9 for DAI. L2Protocol can have a maximum of 100 entries, and this region must be expanded to capacity before you can increase the size of L2SystemFlow. This is relevant when you are enabling DAI on VLANs. If, for example, you want to enable DAI on 16 VLANs, you need 7 more entries; in this case, reconfigure the SystemFlow region for 122 entries:
layer-2 eg-acl value fib value frrp value ing-acl value learn value l2pt value qos value system-flow 122

The logic is as follows: L2Protocol has 87 entries by default and must be expanded to its maximum capacity, 100 entries, before L2SystemFlow can be increased; therefore 13 more L2Protocol entries are required. L2SystemFlow has 15 entries by default, but only 9 are for DAI; to enable DAI on 16 VLANs, 7 more entries are required. 87 L2Protocol + 13 additional L2Protocol + 15 L2SystemFlow + 7 additional L2SystemFlow equals 122.

Step 1 2

Task Enable DHCP Snooping. Validate ARP frames against the DHCP Snooping binding table.

Command Syntax

Command Mode

arp inspection

INTERFACE VLAN

View the number of entries in the ARP database with the show arp inspection database command.
FTOS#show arp inspection database Protocol Address Age(min) Hardware Address Interface VLAN CPU ---------------------------------------------------------------------------Internet 10.1.1.251 00:00:4d:57:f2:50 Gi 0/2 Vl 10 CP Internet 10.1.1.252 00:00:4d:57:e6:f6 Gi 0/1 Vl 10 CP Internet 10.1.1.253 00:00:4d:57:f8:e8 Gi 0/3 Vl 10 CP Internet 10.1.1.254 00:00:4d:69:e8:f2 Te 0/50 Vl 10 CP FTOS#

260

Dynamic Host Configuration Protocol (DHCP)

Use show arp inspection statistics command to see how many valid and invalid ARP packets have been processed.
FTOS#show arp inspection statistics Dynamic ARP Inspection (DAI) Statistics --------------------------------------Valid ARP Requests Valid ARP Replies Invalid ARP Requests Invalid ARP Replies FTOS#

: : : :

0 1000 1000 0

Bypass the ARP Inspection

You can configure a port to skip ARP inspection by defining the interface as trusted, which is useful in multi-switch environments. ARPs received on trusted ports bypass validation against the binding table. All ports are untrusted by default.
Task Specify an interface as trusted so that ARPs are not validated against the binding table. Command Syntax
arp inspection-trust

Command Mode INTERFACE

FTOS Behavior: Introduced in FTOS version 8.2.1.0, Dynamic ARP Inspection (DAI) was available for Layer 3 only. FTOS version 8.2.1.1 extends DAI to Layer 2.

Source Address Validation

Using the DHCP binding table, FTOS can perform three types of source address validation (SAV): IP Source Address Validation prevents IP spoofing by forwarding only IP packets that have been validated against the DHCP binding table. DHCP MAC Source Address Validation verifies a DHCP packets source hardware address matches the client hardware address field (CHADDR) in the payload. IP+MAC Source Address Validation verifies that the IP source address and MAC source address are a legitimate pair.

IP Source Address Validation

IP Source Address Validation (SAV) prevents IP spoofing by forwarding only IP packets that have been validated against the DHCP binding table. A spoofed IP packet is one in which the IP source address is strategically chosen to disguise the attacker. For example, using ARP spoofing an attacker can assume a legitimate clients identity and receive traffic addressed to it. Then the attacker can spoof the clients IP address to interact with other clients.

Dynamic Host Configuration Protocol (DHCP) | 261

www.dell.com | support.dell.com

The DHCP binding table associates addresses assigned by the DHCP servers, with the port on which the requesting client is attached. When IP Source Address Validation is enabled on a port, the system verifies that the source IP address is one that is associated with the incoming port. If an attacker is impostering as a legitimate client the source address appears on the wrong ingress port, and the system drops the packet. Likewise, if the IP address is fake, the address will not be on the list of permissible addresses for the port, and the packet is dropped.
Task Enable IP Source Address Validation Command Syntax
ip dhcp source-address-validation

Command Mode INTERFACE

Note: If IP Source Guard is enabled using the ip dhcp source-address-validation command and there are 187 entries or more in the current DHCP snooping binding table, Source Address Validation (SAV) may not be applied to all entries. To ensure that SAV is applied correctly to all entries, enable the ip dhcp source-address-validation command before adding entries to the binding table.

DHCP MAC Source Address Validation

DHCP MAC Source Address Validation (SAV) validates a DHCP packets source hardware address against the client hardware address field (CHADDR) in the payload. FTOS Release 8.2.1.1 ensures that the packets source MAC address is checked against the CHADDR field in the DHCP header only for packets from snooped VLANs.
Task Enable DHCP MAC Source Address Validation. Command Syntax
ip dhcp snooping verify mac-address

Command Mode CONFIGURATION

IP+MAC Source Address Validation

The following feature is available on platforms:

c, s and

IP Source Address Validation validates the IP source address of an incoming packet against the DHCP Snooping binding table. IP+MAC Source Address Validation ensures that the IP source address and MAC source address are a legitimate pair, rather validating each attribute individually. IP+MAC Source Address Validation cannot be configured with IP Source Address Validation.
Step 1 2 3 Task Allocate at least one FP block to the ipmacacl CAM region. Save the running-config to the startup-config. Reload the system. Command Syntax
cam-acl l2acl

Command Mode CONFIGURATION EXEC Privilege EXEC Privilege

copy running-config startup-config

reload

262

Dynamic Host Configuration Protocol (DHCP)

Step 4

Task Enable IP+MAC Source Address Validation.

Command Syntax
ip dhcp source-address-validation ipmac

Command Mode INTERFACE

FTOS creates an ACL entry for each IP+MAC address pair in the binding table and applies it to the interface.
Task Display the IP+MAC ACL for an interface for the entire system. Command Syntax
show ip dhcp snooping source-address-validation [interface]

Command Mode EXEC Privilege

Dynamic Host Configuration Protocol (DHCP) | 263

www.dell.com | support.dell.com |

264

Dynamic Host Configuration Protocol (DHCP)

12
Force10 Resilient Ring Protocol
Force10 Resilient Ring Protocol is supported on platforms

ce s

Force10 Resilient Ring Protocol (FRRP) provides fast network convergence to Layer 2 switches interconnected in a ring topology, such as a Metropolitan Area Network (MAN) or large campuses. FRRP is similar to what can be achieved with the Spanning Tree Protocol (STP), though even with optimizations, STP can take up to 50 seconds to converge (depending on the size of network and node of failure) may require 4 to 5 seconds to reconverge. FRRP can converge within 150ms to 1500ms when a link in the ring breaks (depending on network configuration). To operate a deterministic network, a network administrator must run a protocol that converges independently of the network size or node of failure. The Force10 Resilient Ring Protocol (FRRP) is a proprietary protocol that provides this flexibility, while preventing Layer 2 loops. FRRP provides sub-second ring-failure detection and convergence/re-convergence in a Layer 2 network while eliminating the need for running spanning-tree protocol. With its two-way path to destination configuration, FRRP provides protection against any single link/switch failure and thus provides for greater network uptime.

Protocol Overview
FRRP is built on a ring topology. Up to 255 rings can be configured on a system. FRRP uses one Master node and multiple Transit nodes in each ring. There is no limit to the number of nodes on a ring. The Master node is responsible for the intelligence of the Ring and monitors the status of the Ring. The Master node checks the status of the Ring by sending Ring Health Frames (RHF) around the Ring from its Primary port and returning on its Secondary port. If the Master node misses three consecutive RHFs, it determines the ring to be in a failed state. The Master then sends a Topology Change RHF to the Transit Nodes informing them that the ring has changed. This causes the Transit Nodes to flush their forwarding tables, and re-converge to the new network structure. One port of the Master node is designated the Primary port (P) to the ring; another port is designated as the Secondary port (S) to the ring. In normal operation, the Master node blocks the Secondary port for all non-control traffic belonging to this FRRP group, thereby avoiding a loop in the ring, like STP. Layer 2 switching and learning mechanisms operate per existing standards on this ring. Each Transit node is also configured with a Primary port and a Secondary port on the ring, but the port distinction is ignored as long as the node is configured as a Transit node. If the ring is complete, the Master node logically blocks all data traffic in the transmit and receive directions on the Secondary port to prevent a loop. If the Master node detects a break in the ring, it unblocks its Secondary port and allows data traffic

Force10 Resilient Ring Protocol | 265

www.dell.com | support.dell.com

to be transmitted and received through it. See Figure 12-1 for a simple example of this FRRP topology. Note that ring direction is determined by the Master nodes Primary and Secondary ports.
Figure 12-1. Normal Operating FRRP Topology

R2 TRANSIT

Primary Forwarding
R ing D ire ction

Secondary Forwarding

Primary Forwarding

Primary Forwarding

R1 MASTER

Secondary Blocking

Secondary Forwarding

R3 TRANSIT

A Virtual LAN (VLAN) is configured on all node ports in the ring. All ring ports must be members of the Member VLAN and the Control VLAN. The Member VLAN is the VLAN used to transmit data as described earlier. The Control VLAN is used to perform the health checks on the ring. The Control VLAN can always pass through all ports in the ring, including the secondary port of the Master node.

Ring Status
The Ring Failure notification and the Ring Status checks provide two ways to ensure the ring remains up and active in the event of a switch or port failure.

Ring Checking
At specified intervals, the Master Node sends a Ring Health Frame (RHF) through the ring. If the ring is complete, the frame is received on its secondary port, and the Master node resets its fail-period timer and continues normal operation.

266

Force10 Resilient Ring Protocol

If the Master node does not receive the Ring Health Frame (RHF) before the fail-period timer expires (a configurable timer), the Master node moves from the Normal state to the Ring-Fault state and unblocks its Secondary port. The Master node also clears its forwarding table and sends a control frame to all other nodes, instructing them to also clear their forwarding tables. Immediately after clearing its forwarding table, each node starts learning the new topology.

Ring Failure
If a Transit node detects a link down on any of its ports on the FRRP ring, it immediately sends a link-down control frame on the Control VLAN to the Master node. When the Master node receives this control frame, the Master node moves from the Normal state to the Ring-Fault state and unblocks its Secondary port. The Master node clears its routing table, and sends a control frame to all other ring nodes, instructing them to clear their routing tables as well. Immediately after clearing its routing table, each node begins learning the new topology.

Ring Restoration
The Master node continues sending Ring Health Frames out its primary port even when operating in the Ring-Fault state. Once the ring is restored, the next status check frame is received on the Master node's Secondary port. This will cause the Master node to transition back to the Normal state. The Master node then logically blocks non-control frames on the Secondary port, clears its own forwarding table, and sends a control frame to the Transit nodes, instructing them to clear their forwarding tables and re-learn the topology. During the time between the Transit node detecting that its link is restored and the Master node detecting that the ring is restored, the Master nodes Secondary port is still forwarding traffic. This can create a temporary loop in the topology. To prevent this, the Transit node places all the ring ports transiting the newly restored port into a temporary blocked state. The Transit node remembers which port has been temporarily blocked and places it into a pre- forwarding state. When the Transit node in the pre-forwarding state receives the control frame instructing it to clear its routing table, it does so and unblocks the previously blocked ring ports on the newly restored port. Then the Transit node returns to the Normal state.

Multiple FRRP Rings

Up to 255 rings allowed per system and multiple rings can be run on one system. However, it is not recommended on the S-Series to have more than 34 rings on the same interface (either a physical interface or a portchannel). More than the recommended number of rings may cause interface instability. Multiple rings can be configured with a single switch connection; a single ring can have multiple FRRP groups; multiple rings can be connected with a common link.

Member VLAN Spanning Two Rings Connected by One Switch

A Member VLAN can span two rings interconnected by a common switch, in a figure-eight style topology. A switch can act as a Master node for one FRRP Group and a Transit for another FRRP group, or it can be a Transit node for both rings.

Force10 Resilient Ring Protocol | 267

www.dell.com | support.dell.com

In the example shown in Figure 12-2, FRRP 101 is a ring with its own Control VLAN, and FRRP 202 has its own Control VLAN running on another ring. A Member VLAN that spans both rings is added as a Member VLAN to both FRRP groups. Switch R3 has two instances of FRRP running on it: one for each ring. The example topology that follows shows R3 assuming the role of a Transit node for both FRRP 101 and FRRP 202.
Figure 12-2. Example of Multiple Rings Connected by Single Switch

FRRP 101
MASTER R1
Primary Forwarding Secondary Blocking

Ring 101 Direction

Primary Forwarding

Primary Forwarding

TRANSIT R3 TRANSIT R3
Secondary Forwarding Secondary Forwarding Secondary Forwarding

TRANSIT R2

Primary Forwarding

TRANSIT R7

Primary Forwarding

Secondary Forwarding Secondary Forwarding

TRANSIT R4

Primary Forwarding

Ring 202 Direction

Primary Forwarding

TRANSIT R6
Secondary Forwarding

Primary Forwarding

FRRP 202
Secondary Blocking

MASTER R5

Important FRRP Points

FRRP provides a convergence time that can generally range between 150ms and 1500ms for Layer 2 networks. The master node originates a high-speed frame that circulates around the ring. This frame, appropriately, sets up or breaks down the ring. Ring Status Check Frames are transmitted by the Master Node at specified intervals

268

Force10 Resilient Ring Protocol

Multiple physical rings can be run on the same switch One Master node per ringall other nodes are Transit Each node has 2 member interfacesPrimary, Secondary No limit to the number of nodes on a ring Master node ring port statesblocking, pre-forwarding, forwarding, disabled Transit node ring port statesblocking, pre-forwarding, forwarding, disabled STP disabled on ring interfaces Master node secondary port is in blocking state during Normal operation Ring Health Frames (RHF) Hello RHF Sent at 500ms (hello interval) Transmitted and processed by Master node only Topology Change RHF Triggered updates Processed at all nodes

Important FRRP Concepts

Table 12-1 lists some important FRRP concepts.
Table 12-1. Concept Ring ID Control VLAN FRRP Components Explanation Each ring has a unique 8-bit ring ID through which the ring is identified (e.g. FRRP 101 and FRRP 202 as shown in Figure 12-2. Each ring has a unique Control VLAN through which tagged Ring Health Frames (RHF) are sent. Control VLANs are used only for sending Ring Health Frames, and cannot be used for any other purpose. Each ring maintains a list of member VLANs. Member VLANs must be consistent across the entire ring. Each node has two ports for each ring: Primary and Secondary. The Master node Primary port generates Ring Health Frames (RHF). The Master node Secondary port receives the RHF frames. On Transit nodes, there is no distinction between a Primary and Secondary interface when operating in the Normal state.

Member VLAN Port Role

Force10 Resilient Ring Protocol | 269

www.dell.com | support.dell.com

Table 12-1. Concept

FRRP Components Explanation Each interface (port) that is part of the ring maintains one of four states Blocking State: Accepts ring protocol packets but blocks data packets. LLDP, FEFD, or other Layer 2 control packets are accepted. Only the master node Secondary port can enter this state. Pre-Forwarding State: A transition state before moving to the Forward state. Control traffic is forwarded but data traffic is blocked. The Master node Secondary port transitions through this state during ring bring-up. All ports transition through this state when a port comes up. Forwarding StateBoth ring control and data traffic is passed. When the ring is in Normal operation, the Primary port on the Master node and both Primary and Secondary ports on the Transit nodes are in forwarding state. When the ring is broken, all ring ports are in this state. Disabled StateWhen the port is disabled or down, or is not on the VLAN.

Ring Interface State

Ring Protocol Timers

Hello Interval: The interval when ring frames are generated from the Master nodes Primary interface (default 500 ms). The Hello interval is configurable in 50 ms increments from 50 ms to 2000 ms. Dead Interval: The interval when data traffic is blocked on a port. The default is 3 times the Hello interval rate. The dead interval is configurable in 50 ms increments from 50 ms to 6000 ms. The state of the FRRP ring. During initialization/configuration, the default ring status is Ring-down (disabled). The Primary and Secondary interfaces, Control VLAN, and Master and Transit node information must be configured for the ring to be up. Ring-Up: Ring is up and operational Ring-Down: Ring is broken or not set up Two types of RHFs are generated by the Master node. RHFs never loop the ring because they terminate at the Master nodes secondary port. Hello RHF (HRHF): These frames are processed only on the Master nodes Secondary port. The Transit nodes pass the HRHF through the without processing it. An HRHF is sent at every Hello interval. Topology Change RHF (TCRHF): These frames contains ring status, keepalive, and the Control and Member VLAN hash. It is processed at each node of the ring. TCRHFs are sent out the Master Nodes Primary and Secondary interface when the ring is declared in a Failed state with the same sequence number, on any topology change to ensure all Transit nodes receive it. There is no periodic transmission of TCRHFs. The TCRHFs are sent on triggered events of ring failure or ring restoration only.

Ring Status

Ring Health-check Frame (RHF)

Implementing FRRP
FRRP is media and speed independent. FRRP is a Dell Networking proprietary protocol that does not interoperate with any other vendor. Spanning Tree must be disabled on both Primary and Secondary interfaces before FRRP is enabled. All ring ports must be Layer 2 ports. This is required for both Master and Transit nodes. A VLAN configured as control VLAN for a ring cannot be configured as a control or member VLAN for any other ring.

270

Force10 Resilient Ring Protocol

The Control VLAN is used to carry any data traffic; it carries only RHFs. The Control VLAN cannot have members that are not ring ports. If multiple rings share one or more member VLANs, they cannot share any links between them.

Member VLANs across multiple rings are not supported in Master nodes.
Each ring has only one Master node; all others are transit nodes.

FRRP Configuration
These are the tasks to configure FRRP. Create the FRRP group Configure the Control VLAN Configure Primary and Secondary ports Configure and add the Member VLANs Configure Primary and Secondary ports Configure the Master node Configure a Transit node Set FRRP Timers (optional) Enable FRRP

Other FRRP related commands are: Clear FRRP counters

Create the FRRP group

The FRRP group must be created on each switch in the ring. Use the commands in the following sequence to create the FRRP group.
Command Syntax
protocol frrp ring-id

Command Mode CONFIGURATION

Purpose Create the FRRP group with this Ring ID Ring ID: 1-255

Configure the Control VLAN

Control and Member VLANS are configured normally for Layer 2. Their status as Control or Member is determined at the FRRP group commands. For complete information about configuring VLANS in Layer 2 mode, see Chapter 20, Layer 2. Be sure to follow these guidelines:

Force10 Resilient Ring Protocol | 271

www.dell.com | support.dell.com

All VLANS must be in Layer 2 mode. Only ring nodes can be added to the VLAN. A Control VLAN can belong to one FRRP group only. Control VLAN ports must be tagged. All ports on the ring must use the same VLAN ID for the Control VLAN. A VLAN cannot be configured as both a Control VLAN and Member VLAN on the same ring. Only two interfaces can be members of a Control VLAN (the Master Primary and Secondary ports).

Member VLANs across multiple rings are not supported in Master nodes

Use the commands in the following sequence, on the switch that will act as the Master node, to create the Control VLAN for this FRRP group.
Step 1 2 Command Syntax
interface vlan vlan-id

Command Mode CONFIGURATION CONFIG-INT-VLAN

Purpose Create a VLAN with this ID number VLAN ID: 1-4094 Tag the specified interface or range of interfaces to this VLAN. Interface: For a 10/100/1000 Ethernet interface, enter the keyword keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. Slot/Port, Range: Slot and Port ID for the interface. Range is entered Slot/Port-Port. Assign the Primary and Secondary ports, and the Control VLAN for the ports on the ring. Interface: For a 10/100/1000 Ethernet interface, enter the keyword keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information For a SONET interface, enter the keyword sonet followed by slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. Slot/Port: Slot and Port ID for the interface. VLAN ID: The VLAN identification of the Control VLAN. Configure the Master node

tagged interface slot/ port {range}

interface primary int slot/port secondary int slot/port control-vlan vlan id

CONFIG-FRRP

mode master

CONFIG-FRRP

272

Force10 Resilient Ring Protocol

Step 5

Command Syntax
member-vlan vlan-id {range}

Command Mode CONFIG-FRRP

Purpose Identify the Member VLANs for this FRRP group VLAN-ID, Range: VLAN IDs for the rings Member VLANS. Enable FRRP

no disable

CONFIG-FRRP

Configure and add the Member VLANs

Control and Member VLANS are configured normally for Layer 2. Their status as Control or Member is determined at the FRRP group commands. For complete information about configuring VLANS in Layer 2 mode, see Chapter 20, Layer 2. Be sure to follow these guidelines: All VLANS must be in Layer 2 mode. Control VLAN ports must be tagged. Member VLAN ports except the Primary/Secondary interface can be tagged or untagged. The Control VLAN must be the same for all nodes on the ring.

Use the commands in the following sequence, on all of the Transit switches in the ring, to create the Members VLANs for this FRRP group.
Step 1 2 Command Syntax
interface vlan vlan-id

Command Mode CONFIGURATION CONFIG-INT-VLAN

Purpose Create a VLAN with this ID number VLAN ID: 1-4094 Tag the specified interface or range of interfaces to this VLAN. Interface: For a 10/100/1000 Ethernet interface, enter the keyword keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information For a SONET interface, enter the keyword sonet followed by slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. Slot/Port, Range: Slot and Port ID for the interface. Range is entered Slot/Port-Port.

tagged interface slot/ port {range}

Force10 Resilient Ring Protocol | 273

www.dell.com | support.dell.com

Step 3

Command Syntax
interface primary int slot/port secondary int slot/port control-vlan vlan id

Command Mode CONFIG-FRRP

Purpose Assign the Primary and Secondary ports, and the Control VLAN for the ports on the ring. Interface: For a 10/100/1000 Ethernet interface, enter the keyword keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information For a SONET interface, enter the keyword sonet followed by slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. Slot/Port: Slot and Port ID for the interface. VLAN ID: Identification number of the Control VLAN Configure a Transit node Identify the Member VLANs for this FRRP group VLAN-ID, Range: VLAN IDs for the rings Member VLANs. Enable this FRRP group on this switch.

4 5

mode transit member-vlan vlan-id {range}

CONFIG-FRRP CONFIG-FRRP

no disable

CONFIG-FRRP

Set FRRP Timers

Step 1 Command Syntax
timer

Command Mode CONFIG-FRRP

Purpose Enter the desired intervals for Hello-Interval or Dead-Interval times. Hello-Interval: 50-2000, in increments of 50 (default is 500) Dead-Interval: 50-6000, in increments of 50 (default is 1500)

{hello-interval|dead-interval} milliseconds

The Dead-Interval time should be set at 3x the Hello-Interval.

Clear FRRP counters

Use one of the following commands to clear the FRRP counters.
Command Syntax
clear frrp ring-id

Command Mode EXEC PRIVELEGED

Purpose Clear the counters associated with this Ring ID Ring ID: 1-255

274

Force10 Resilient Ring Protocol

Command Syntax
clear frrp

Command Mode EXEC PRIVELEGED

Purpose Clear the counters associated with all FRRP groups

Show FRRP configuration

Use the following command to view the configuration for the FRRP group.
Command Syntax
show configuration

Command Mode CONFIG-FRRP

Purpose Show the configuration for this FRRP group

Show FRRP information

Use one of the following commands show general FRRP information.
Command Syntax
show frrp ring-id

Command Mode EXEC or EXEC PRIVELEGED EXEC or EXEC PRIVELEGED

Purpose Show the information for the identified FRRP group. Ring ID: 1-255 Show the state of all FRRP groups. Ring ID: 1-255

show frrp summary

Troubleshooting FRRP
Configuration Checks
Each Control Ring must use a unique VLAN ID Only two interfaces on a switch can be Members of the same Control VLAN There can be only one Master node for any FRRP Group. FRRP can be configured on Layer 2 interfaces only Spanning Tree (if enabled globally) must be disabled on both Primary and Secondary interfaces when FRRP is enabled. When the interface ceases to be a part of any FRRP process, if Spanning Tree is enabled globally, it must be enabled explicitly for the interface. The maximum number of rings allowed on a chassis is 255.

Sample Configuration and Topology

Force10 Resilient Ring Protocol | 275

www.dell.com | support.dell.com

Figure 12-3 is an example of a basic FRRP topology. Below the figure are the associated CLI commands.
Figure 12-3. Basic Topology and CLI commands

R2 TRANSIT

R ing D irec tion

Primary Forwarding GigE 2/14

Secondary Forwarding GigE 2/31

Primary Forwarding GigE 1/24

Primary Forwarding GigE 3/21

R1 MASTER

Secondary Blocking GigE 1/34

Secondary Forwarding GigE 3/14

R3 TRANSIT

276

Force10 Resilient Ring Protocol

R1 MASTER
interface GigabitEthernet 1/24 no ip address switchport no shutdown ! interface GigabitEthernet 1/34 no ip address switchport no shutdown ! interface Vlan 101 no ip address tagged GigabitEthernet 1/24,34 no shutdown ! interface Vlan 201 no ip address tagged GigabitEthernet 1/24,34 no shutdown ! protocol frrp 101 interface primary GigabitEthernet 1/24 secondary GigabitEthernet 1/34 control-vlan 101 member-vlan 201 mode master no disable

R2 TRANSIT
interface GigabitEthernet 2/14 no ip address switchport no shutdown ! interface GigabitEthernet 2/31 no ip address switchport no shutdown ! interface Vlan 101 no ip address tagged GigabitEthernet 2/14,31 no shutdown ! interface Vlan 201 no ip address tagged GigabitEthernet 2/14,31 no shutdown ! protocol frrp 101 interface primary GigabitEthernet 2/14 secondary GigabitEthernet 2/31 control-vlan 101 member-vlan 201 mode transit no disable

R3 TRANSIT
interface GigabitEthernet 3/14 no ip address switchport no shutdown ! interface GigabitEthernet 3/21 no ip address switchport no shutdown ! interface Vlan 101 no ip address tagged GigabitEthernet 3/14,21 no shutdown ! interface Vlan 201 no ip address tagged GigabitEthernet 3/14,21 no shutdown ! protocol frrp 101 interface primary GigabitEthernet 3/21 secondary GigabitEthernet 3/14 control-vlan 101 member-vlan 201 mode transit no disable

Force10 Resilient Ring Protocol | 277

www.dell.com | support.dell.com |

278

Force10 Resilient Ring Protocol

13
GARP VLAN Registration Protocol
GARP VLAN Registration Protocol is supported on platform

ces

Protocol Overview
Typical VLAN implementation involves manually configuring each Layer 2 switch that participates in a given VLAN. GARP VLAN Registration Protocol (GVRP), defined by the IEEE 802.1q specification, is a Layer 2 network protocol that provides for automatic VLAN configuration of switches. GVRP-compliant switches use GARP to register and de-register attribute values, such as VLAN IDs, with each other. GVRP exchanges network VLAN information to allow switches to dynamically forward frames for one or more VLANs. Consequently, GVRP spreads this information and configures the needed VLAN(s) on any additional switches in the network. Data propagates via the exchange of GVRP protocol data units (PDUs). The purpose of GVRP is to simplify (but not eliminate) static configuration. The idea is to configure switches at the edge and have the information dynamically propagate into the core. As such, the edge ports must still be statically configured with VLAN membership information, and they do not run GVRP. It is this information that is propagated to create dynamic VLAN membership in the core of the network.

Important Points to Remember

GVRP propagates VLAN membership throughout a network. GVRP allows end stations and switches to issue and revoke declarations relating to VLAN membership. VLAN registration is made in the context of the port that receives the GARP PDU and is propagated to the other active ports. GVRP is disabled by default; you must enable GVRP for the switch and then for individual ports. Dynamic VLANs are aged out after the LeaveAll timer expires three times without receipt of a Join message. Use the show gvrp statistics {interface interface | summary} command to display status. On the E-Series, C-Series, and non-S60/S55 S-Series, Per-VLAN Spanning Tree (PVST+) or MSTP and GVRP cannot be enabled at the same time, as shown in Figure 13-1. If Spanning Tree and GVRP are both required, implement RSTP. The S60 and the S55 systems do support enabling GVRP and MSTP at the same time.

GARP VLAN Registration Protocol | 279

www.dell.com | support.dell.com

Figure 13-1. GVRP Compatibility Error Message

FTOS(conf)#protocol spanning-tree pvst FTOS(conf-pvst)#no disable % Error: GVRP running. Cannot enable PVST. ......... FTOS(conf)#protocol spanning-tree mstp FTOS(conf-mstp)#no disable % Error: GVRP running. Cannot enable MSTP. ......... FTOS(conf)#protocol gvrp FTOS(conf-gvrp)#no disable % Error: PVST running. Cannot enable GVRP. % Error: MSTP running. Cannot enable GVRP.

Configuring GVRP
Globally, enable GVRP on each switch to facilitate GVRP communications. Then, GVRP configuration is per interface on a switch-by-switch basis. Enable GVRP on each port that connects to a switch where you want GVRP information exchanged. In Figure 13-2, that kind of port is referred to as a VLAN trunk port, but it is not necessary to specifically identify to FTOS that the port is a trunk port.

280

GARP VLAN Registration Protocol

Figure 13-2. GVRP Configuration Overview

GVRP is configured globally and on all VLAN trunk ports for the edge and core switches. Edge Switches Core Switches VLANs 70-80 VLANs 10-20 Edge Switches

VLANs 10-20

VLANs 30-50

VLANs 30-50

VLANs 70-80

NOTES: VLAN 1 mode is always fixed and cannot be configured All VLAN trunk ports must be configured for GVRP All VLAN trunk ports must be configured as 802.1Q

Basic GVRP configuration is a 2-step process: 1. Enable GVRP globally. See page 282. 2. Enable GVRP on an interface. See page 282.

Related Configuration Tasks

Configuring GVRP Registration Configuring a GARP Timer

Enabling GVRP Globally

Enable GVRP for the entire switch using the command gvrp enable in CONFIGURATION mode, as shown in Figure 13-3. Use the show gvrp brief command to inspect the global configuration.

GARP VLAN Registration Protocol | 281

www.dell.com | support.dell.com

Figure 13-3. Enabling GVRP Globally

FTOS(conf)#protocol gvrp FTOS(config-gvrp)#no disable FTOS(config-gvrp)#show config ! protocol gvrp no disable FTOS(config-gvrp)#

Enabling GVRP on a Layer 2 Interface

Enable GVRP on a Layer 2 interface using the command gvrp enable in INTERFACE mode, as shown in Figure 13-4. Use show config from the INTERFACE mode to inspect the interface configuration, as shown in Figure 13-4, or use the show gvrp interface command in EXEC or EXEC Privilege mode.
Figure 13-4. Enabling GVRP on a Layer 2 Interface
FTOS(conf-if-gi-1/21)#switchport FTOS(conf-if-gi-1/21)#gvrp enable FTOS(conf-if-gi-1/21)#no shutdown FTOS(conf-if-gi-1/21)#show config ! interface GigabitEthernet 1/21 no ip address switchport gvrp enable no shutdown

Configuring GVRP Registration

Fixed Registration Mode: Configuring a port in fixed registration mode allows for manual creation and registration of VLANs, prevents VLAN de-registration, and registers all VLANs known on other ports on the port. For example, if an interface is statically configured via the CLI to belong to a VLAN, it should not be un-configured when it receives a Leave PDU. So, the registration mode on that interface is FIXED. Forbidden Mode: Disables the port to dynamically register VLANs, and to propagate VLAN information except information about VLAN 1. A port with forbidden registration type thus allows only VLAN 1 to pass through even though the PDU carries information for more VLANs. So, set the interface to the registration mode of FORBIDDEN if you do not want the interface to advertise or learn about particular VLANS.

282

GARP VLAN Registration Protocol

Based on the configuration in the example shown in Figure 13-5, the interface 1/21 will not be removed from VLAN 34 or VLAN 35 despite receiving a GVRP Leave message. Additionally, the interface will not be dynamically added to VLAN 45 or VLAN 46, even if a GVRP Join message is received.
Figure 13-5. Configuring GVRP Registration
FTOS(conf-if-gi-1/21)#gvrp registration fixed 34,35 FTOS(conf-if-gi-1/21)#gvrp registration forbidden 45,46 FTOS(conf-if-gi-1/21)#show conf ! interface GigabitEthernet 1/21 no ip address switchport gvrp enable gvrp registration fixed 34-35 gvrp registration forbidden 45-46 no shutdown FTOS(conf-if-gi-1/21)#

Configuring a GARP Timer

GARP timers must be set to the same values on all devices that are exchanging information using GVRP: Join: A GARP device reliably transmits Join messages to other devices by sending each Join message two times. Use this parameter to define the interval between the two sending operations of each Join message. The FTOS default is 200ms. Leave: When a GARP device expects to de-register a piece of attribute information, it will send out a Leave message and start this timer. If a Join message does not arrive before the timer expires, the information is de-registered. The Leave timer must be greater than or equal to 3x the Join timer. The FTOS default is 600ms. LeaveAll: Upon startup, a GARP device globally starts a LeaveAll timer. Upon expiration of this interval, it will send out a LeaveAll message so that other GARP devices can re-register all relevant attribute information. The device then restarts the LeaveAll timer to begin a new cycle. The LeaveAll timer must be greater than or equal to 5x of the Leave timer. The FTOS default is 10000ms.

Figure 13-6. Configuring GVRP Registration

FTOS(conf)#garp timer leav 1000 FTOS(conf)#garp timers leave-all 5000 FTOS(conf)#garp timer join 300 Verification: FTOS(conf)#do show garp timer GARP Timers Value (milliseconds) ---------------------------------------Join Timer 300 Leave Timer 1000 LeaveAll Timer 5000 FTOS(conf)#

GARP VLAN Registration Protocol | 283

www.dell.com | support.dell.com

FTOS displays Message 1 if an attempt is made to configure an invalid GARP timer.

Message 1 GARP Timer Error
FTOS(conf)#garp timers join 300 % Error: Leave timer should be >= 3*Join timer.

284

GARP VLAN Registration Protocol

14
Internet Group Management Protocol
Table 14-1. Feature IGMP version 1, 2, and 3 IGMP Snooping version 2 IGMP Snooping version 3 FTOS Support for IGMP and IGMP Snooping Platform

ces ces ces

Multicast is premised on identifying many hosts by a single destination IP address; hosts represented by the same IP address are a multicast group. Internet Group Management Protocol (IGMP) is a Layer 3 multicast protocol that hosts use to join or leave a multicast group. Multicast routing protocols (such as PIM) use the information in IGMP messages to discover which groups are active and to populate the multicast routing table.

IGMP Implementation Information

FTOS supports IGMP versions 1, 2, and 3 based on RFCs 1112, 2236, and 3376, respectively. FTOS does not support IGMP version 3 and versions 1 or 2 on the same subnet. IGMP on FTOS supports up to 512 interfaces on E-Series, 31 interfaces on C-Series and S-Series, and an unlimited number of groups on all platforms.
Note: The S60 supports up to 95 interfaces.

Dell Networking systems cannot serve as an IGMP host or an IGMP version 1 IGMP Querier. FTOS automatically enables IGMP on interfaces on which you enable a multicast routing protocol.

Internet Group Management Protocol | 285

www.dell.com | support.dell.com

IGMP Protocol Overview

IGMP has three versions. Version 3 obsoletes and is backwards-compatible with version 2; version 2 obsoletes version 1.

IGMP version 2
IGMP version 2 improves upon version 1 by specifying IGMP Leave messages, which allows hosts to notify routers that they no longer care about traffic for a particular group. Leave messages reduce the amount of time that the router takes to stop forwarding traffic for a group to a subnet (leave latency) after the last host leaves the group. In version 1 hosts quietly leave groups, and the router waits for a query response timer several times the value of the query interval to expire before it stops forwarding traffic. To receive multicast traffic from a particular source, a host must join the multicast group to which the source is sending traffic. A host that is a member of a group is called a receiver. A host may join many groups, and may join or leave any group at any time. A host joins and leaves a multicast group by sending an IGMP message to its IGMP Querier. The querier is the router that surveys a subnet for multicast receivers, and processes survey responses to populate the multicast routing table. IGMP messages are encapsulated in IP packets, as shown in Figure 14-1.
Figure 14-1. IGMP version 2 Packet Format
Preamble Start Frame Delimiter Destination MAC Source MAC Ethernet Type IP Packet Padding FCS

Version (4)

IHL

TOS (0xc0)

Total Length

Flags

Frag Offset

TTL (1)

Protocol (2)

Header Checksum

Src IP Addr

Dest IP Addr

Options (Router Alert)

Padding

IGMP Packet

Type

Max. Response Time

Checksum

Group Address

8 bits

16 bits

Code: 0x11: 0x12: 0x16: 0x17:

Membership Query IGMP version 1 Membership Report IGMP version 2 Membership Report IGMP Leave Group

May be zero and ignored by hosts for general queries or contain a group address for group-specific queries

fnC0069mp

Joining a Multicast Group

There are two ways that a host may join a multicast group: it may respond to a general query from its querier, or it may send an unsolicited report to its querier.

Responding to an IGMP Query

1. One router on a subnet is elected as the querier. The querier periodically multicasts (to all-multicast-systems address 224.0.0.1) a general query to all hosts on the subnet. 2. A host that wants to join a multicast group responds with an IGMP Membership Report that contains the multicast address of the group it wants to join (the packet is addressed to the same group). If multiple hosts want to join the same multicast group, only the report from the first host to respond reaches the querier, and the remaining hosts suppress their responses (see Adjusting Query and

286

Internet Group Management Protocol

Response Timers for how the delay timer mechanism works). 3. The querier receives the report for a group and adds the group to the list of multicast groups associated with its outgoing port to the subnet. Multicast traffic for the group is then forwarded to that subnet.

Sending an Unsolicited IGMP Report

A host does not have to wait for a general query to join a group. It may send an unsolicited IGMP Membership Report, also called an IGMP Join message, to the querier.

Leaving a Multicast Group

1. A host sends a membership report of type 0x17 (IGMP Leave message) to the all routers multicast address 224.0.0.2 when it no longer cares about multicast traffic for a particular group. 2. The querier sends a Group-Specific Query to determine whether there are any remaining hosts in the group. There must be at least one receiver in a group on a subnet for a router to forward multicast traffic for that group to the subnet. 3. Any remaining hosts respond to the query according to the delay timer mechanism (see Adjusting Query and Response Timers). If no hosts respond (because there are none remaining in the group) the querier waits a specified period, and sends another query. If it still receives no response, the querier removes the group from the list associated with forwarding port and stops forwarding traffic for that group to the subnet.

IGMP version 3
Conceptually, IGMP version 3 behaves the same as version 2. There are differences: Version 3 adds the ability to filter by multicast source, which helps multicast routing protocols avoid forwarding traffic to subnets where there are no interested receivers. To enable filtering, routers must keep track of more state information, that is, the list of sources that must be filtered. An additional query type, the Group-and-Source-Specific Query, keeps track of state changes, while the Group-Specific and General queries still refresh existing state. Reporting is more efficient and robust: hosts do not suppress query responses (non-suppression helps track state and enables the immediate-leave and IGMP Snooping features), state-change reports are retransmitted to insure delivery, and a single membership report bundles multiple statements from a single host, rather than sending an individual packet for each statement.

The version 3 packet structure is different from version 2 to accommodate these protocol enhancements. Queries (Figure 14-2) are still sent to the all-systems address 224.0.0.1, but reports (Figure 14-3) are sent to the all IGMP version 3-capable multicast routers address 244.0.0.22.

Internet Group Management Protocol | 287

www.dell.com | support.dell.com

Figure 14-2. IGMP version 3 Membership Query Packet Format

Type (0x11) Max. Response Code Checksum Group Address Reserved S Querier Robustness Value (2) Querier's Query Interval Code Number of Sources Source Addresses

Maximum Response Time derived from this value

Bit flag that when set to 1 suppresses router query response timer updates

Query Interval derived from this value

Source addresses to be filtered

Code: 0x11: Membership Query

Number of times that a router or receiver transmits a query or report to insure that it is received

Number of source addresses to be filtered

fnC0070mp

Figure 14-3. IGMP version 3 Membership Report Packet Format

Version (4) IHL TOS (0xc0) Total Length Flags Frag Offset TTL (1) Protocol (2) Header Checksum Src IP Addr Dest IP Addr (224.0.0.22) Options (Router Alert) Padding IGMP Packet

Type

Reserved

Checksum

Reserved

Number of Group Records

Group Record 1

Group Record 2

Group Record N

Value used by IGMP to calculate multicast reception state Record Type 0x12: IGMP version 1 Membership Report 0x16: IGMP version 2 Membership Report 0x17: IGMP Leave Group 0x22: IGMP version 3 Membership Report Auxiliary Data Length (0) Number of Sources Multicast Address Source Addresses

Auxiliary Data

Length of Auxiliary Data field

Group address to which the group record pertains Source addresses to be filtered

None defined in RFC 3376

Number of source addresses Range: 1-6 to be filtered Code: 1: Current state is Include 2: Current state is Exclude 3: State change to Include 4: State change to Exclude 5: Allow new sources and no state change 6: Block old sources and no state change

fnC0071mp

Joining and Filtering Groups and Sources

Figure 14-4 shows how multicast routers maintain the group and source information from unsolicited reports. 1. The first unsolicited report from the host indicates that it wants to receive traffic for group 224.1.1.1. 2. The hosts second report indicates that it is only interested in traffic from group 224.1.1.1, source 10.11.1.1. Include messages prevent traffic from all other sources in the group from reaching the subnet, so before recording this request, the querier sends a group-and-source query to verify that there are no hosts interested in any other sources. The multicast router must satisfy all hosts if they have conflicting requests. For example, if another host on the subnet is interested in traffic from 10.11.1.3, then the router cannot record the include request. There are no other interested hosts, so the request is recorded. At this point, the multicast routing protocol prunes the tree to all but the specified sources. 3. The hosts third message indicates that it is only interested in traffic from sources 10.11.1.1 and 10.11.1.2. Since this request again prevents all other sources from reaching the subnet, the router sends another group-and-source query so that it can satisfy all other hosts. There are no other interested hosts so the request is recorded.

288

Internet Group Management Protocol

Figure 14-4. IGMP Membership Reports: Joining and Filtering

Membership Reports: Joining and Filtering

3

IGMP Group-and-Source Specific Query Non-Querier

Interface Multicast Group Filter Source Source Address Timer Mode Timer 1/1 224.1.1.1 GMI Exclude None 1/1 224.1.1.1 Include 10.11.1.1 GMI 1/1 224.1.1.1 Include 10.11.1.1 GMI 10.11.1.2 GMI

Querier Type: 0x11 Group Address: 244.1.1.1 Number of Sources: 1 Source Address: 10.11.1.1 1/1

Change to Include

Type: 0x22 Number of Group Records: 1 Record Type: 3 Number of Sources: 1 Multicast Address: 224.1.1.1 Source Address: 10.11.1.1

Type: 0x22 Number of Group Records: 1 Record Type: 4 Number of Sources: 0 Multicast Address: 224.1.1.1 Type: 0x22 Number of Group Records: 1 Record Type: 5 Number of Sources: 1 Multicast Address: 224.1.1.1 Source Address: 10.11.1.2

IGMP Join message

Allow New

State-change reports retransmitted Query Robustness Value-1 times at Unsolicited Report Interval

fnC0072mp

Leaving and Staying in Groups

Figure 14-5 shows how multicast routers track and refresh state changes in response to group-and-specific and general queries. 1. Host 1 sends a message indicating it is leaving group 224.1.1.1 and that the include filter for 10.11.1.1 and 10.11.1.2 are no longer necessary. 2. The querier, before making any state changes, sends a group-and-source query to see if any other host is interested in these two sources; queries for state-changes are retransmitted multiple times. If any are, they respond with their current state information and the querier refreshes the relevant state information. 3. Separately in Figure 14-5, the querier sends a general query to 224.0.0.1. 4. Host 2 responds to the periodic general query so the querier refreshes the state information for that group.

Internet Group Management Protocol | 289

www.dell.com | support.dell.com

Figure 14-5. IGMP Membership Queries: Leaving and Staying in Groups

Membership Queries: Leaving and Staying

Querier
Interface Multicast Group Filter Source Source Address Timer Mode Timer 1/1 224.1.1.1 Include 10.11.1.1 LQMT 10.11.1.2 LQMT 224.2.2.2 GMI Exclude None Non-querier builds identical table and waits Other Querier Present Interval to assume Querier role

Non-Querier

1/1
Queries retransmitted Last Member Query Count times at Last Member Query Interval

2/1

IGMP Group-and-Source Specific Query

Type: 0x11 Group Address: 224.1.1.1 Number of Sources: 2 Source Address: 10.11.1.1, 10.11.1.2

Type: 0x11 Group Address: 224.0.0.1 Number of Sources: 0 Type: 0x22 Number of Group Records: 1 Record Type: 2 Number of Sources: 0 Multicast Address: 224.2.2.2

IGMP General Membership Query

Type: 0x17 Number of Group Records: 1 Record Type: 6 Number of Sources: 2 Multicast Address: 224.1.1.1 Source Addresses: 10.11.1.1, 10.11.1.2 IGMP Leave message

IGMP Membership Report

Host 1

Host 2

Configuring IGMP
Configuring IGMP is a two-step process: 1. Enable multicast routing using the command ip multicast-routing. 2. Enable a multicast routing protocol.

Related Configuration Tasks

Viewing IGMP Enabled Interfaces Selecting an IGMP Version Viewing IGMP Groups Adjusting Timers Configuring a Static IGMP Group Prevent a Host from Joining a Group Enabling IGMP Immediate-leave IGMP Snooping Fast Convergence after MSTP Topology Changes Designating a Multicast Router Interface

Viewing IGMP Enabled Interfaces

Interfaces that are enabled with PIM-SM are automatically enabled with IGMP. View IGMP-enabled interfaces using the command show ip igmp interface command in the EXEC Privilege mode.

290

Internet Group Management Protocol

Figure 14-6. Viewing IGMP-enabled Interfaces

FTOS#show ip igmp interface gig 7/16 GigabitEthernet 7/16 is up, line protocol is up Internet address is 10.87.3.2/24 IGMP is enabled on interface IGMP query interval is 60 seconds IGMP querier timeout is 300 seconds IGMP max query response time is 10 seconds Last member query response interval is 199 ms IGMP activity: 0 joins, 0 leaves IGMP querying router is 10.87.3.2 (this system) IGMP version is 2 FTOS#

Selecting an IGMP Version

FTOS enables IGMP version 2 by default, which supports version 1 and 2 hosts, but is not compatible with version 3 on the same subnet. If hosts require IGMP version 3, you can switch to IGMP version 3 using the command ip igmp version from INTERFACE mode, as shown in Figure 14-7.
Figure 14-7. Selecting an IGMP Version
FTOS(conf-if-gi-1/13)#ip igmp version 3 FTOS(conf-if-gi-1/13)#do show ip igmp interface GigabitEthernet 1/13 is up, line protocol is down Inbound IGMP access group is not set Interface IGMP group join rate limit is not set Internet address is 1.1.1.1/24 IGMP is enabled on interface IGMP query interval is 60 seconds IGMP querier timeout is 125 seconds IGMP max query response time is 10 seconds IGMP last member query response interval is 1000 ms IGMP immediate-leave is disabled IGMP activity: 0 joins, 0 leaves, 0 channel joins, 0 channel leaves IGMP querying router is 1.1.1.1 (this system) IGMP version is 3 FTOS(conf-if-gi-1/13)#

Viewing IGMP Groups

View both learned and statically configured IGMP groups using the command show ip igmp groups from EXEC Privilege mode.

Internet Group Management Protocol | 291

www.dell.com | support.dell.com

Figure 14-8. Viewing Static and Learned IGMP Groups

FTOS(conf-if-gi-1/0)#do sho ip igmp groups Total Number of Groups: 2 IGMP Connected Group Membership Group Address Interface Uptime 224.1.1.1 GigabitEthernet 1/0 00:00:03 224.1.2.1 GigabitEthernet 1/0 00:56:55

Expires Never 00:01:22

Last Reporter CLI 1.1.1.2

Adjusting Timers
View the current value of all IGMP timers using the command show ip igmp interface from EXEC Privilege mode, as shown in Figure 14-6.

Adjusting Query and Response Timers

The querier periodically sends a general query to discover which multicast groups are active. A group must have at least one host to be active. When a host receives a query, it does not respond immediately, but rather starts a delay timer. The delay time is set to a random value between 0 and the Maximum Response Time. The host sends a response when the timer expires; in version 2, if another host responds before the timer expires, the timer is nullified, and no response is sent. The Maximum Response Time is the amount of time that the querier waits for a response to a query before taking further action. The querier advertises this value in the query (see Figure 14-1). Lowering this value decreases leave latency but increases response burstiness since all host membership reports must be sent before the Maximum Response Time expires. Inversely, increasing this value decreases burstiness at the expense of leave latency. Adjust the period between queries using the command ip igmp query-interval from INTERFACE mode. Adjust the Maximum Response Time using the command ip igmp query-max-resp-time from INTERFACE mode.

When the querier receives a leave message from a host, it sends a group-specific query to the subnet. If no response is received, it sends another. The amount of time that the querier waits to receive a response to the initial query before sending a second one is the Last Member Query Interval (LMQI). The switch waits one LMQI after the second query before removing the group from the state table. Adjust the Last Member Query Interval using the command ip igmp last-member-query-interval from INTERFACE mode.

Adjusting the IGMP Querier Timeout Value

If there is more than one multicast router on a subnet, only one is elected to be the querier, which is the router that sends queries to the subnet. 1. Routers send queries to the all multicast systems address, 224.0.0.1. Initially, all routers send queries.

292

Internet Group Management Protocol

2. When a router receives a query it compares the IP address of the interface on which it was received with the source IP address given in the query. If the receiving router IP address is greater than the source address given in the query, the router stops sending queries. By this method, the router with the lowest IP address on the subnet is elected querier and continues to send queries. 3. If a specified amount of time elapses during which other routers on the subnet do not receive a query, those routers assume that the querier is down, and a new querier is elected. The amount of time that elapses before routers on a subnet assume that the querier is down is the Other Querier Present Interval. Adjust this value using the command ip igmp querier-timeout from INTERFACE mode.

Configuring a Static IGMP Group

Configure a static IGMP group using the command ip igmp static-group. Multicast traffic for static groups is always forwarded to the subnet even if there are no members in the group. View the static groups using the command show ip igmp groups from EXEC Privilege mode. Static groups have an expiration value of Never and a Last Reporter value of CLI, as shown in Figure 14-8.

Enabling IGMP Immediate-leave

If the querier does not receive a response to a group-specific or group-and-source query, it sends another (Querier Robustness Value). Then, after no response, it removes the group from the outgoing interface for the subnet. IGMP Immediate Leave reduces leave latency by enabling a router to immediately delete the group membership on an interface upon receiving a Leave message (it does not send any group-specific or group-and-source queries before deleting the entry). Configure the system for IGMP Immediate Leave using the command ip igmp immediate-leave. View the enable status of this feature using the command show ip igmp interface from EXEC Privilege mode, as shown in Figure 14-7.

Internet Group Management Protocol | 293

www.dell.com | support.dell.com

IGMP Snooping
Multicast packets are addressed with multicast MAC addresses, which represent a group of devices, rather than one unique device. Switches forward multicast frames out of all ports in a VLAN by default, even though there may be only some interested hosts, which is a waste of bandwidth. IGMP Snooping enables switches to use information in IGMP packets to generate a forwarding table that associates ports with multicast groups so that when they receive multicast frames, they can forward them only to interested receivers.

IGMP Snooping Implementation Information

IGMP Snooping on FTOS uses IP multicast addresses not MAC addresses. IGMP Snooping is not supported on stacked VLANs. IGMP Snooping is supported on all S-Series stack members,. IGMP Snooping reacts to STP and MSTP topology changes by sending a general query on the interface that transitions to the forwarding state.

Configuring IGMP Snooping

Configuring IGMP Snooping is a one-step process. That is, enable it on a switch using the command ip igmp snooping enable from CONFIGURATION mode. View the configuration using the command show running-config from CONFIGURATION mode, as shown in Figure 14-9. You can disable snooping on for a VLAN using the command no ip igmp snooping from INTERFACE VLAN mode.
Figure 14-9. Enabling IGMP Snooping
FTOS(conf)#ip igmp snooping enable FTOS(conf)#do show running-config igmp ip igmp snooping enable FTOS(conf)#

Related Configuration Tasks

Enabling IGMP Immediate-leave Disabling Multicast Flooding Specifying a Port as Connected to a Multicast Router Configuring the Switch as Querier

Enabling IGMP Immediate-leave

Configure the switch to remove a group-port association upon receiving an IGMP Leave message using the command ip igmp fast-leave from INTERFACE VLAN mode. View the configuration using the command show config from INTERFACE VLAN mode, as shown in Figure 14-10.

294

Internet Group Management Protocol

Figure 14-10.

Enabling IGMP Snooping

FTOS(conf-if-vl-100)#show config ! interface Vlan 100 no ip address ip igmp snooping fast-leave shutdown FTOS(conf-if-vl-100)#

Disabling Multicast Flooding

If the switch receives a multicast packet that has an IP address of a group it has not learned (unregistered frame), the switch floods that packet out of all ports on the VLAN. On the E-Series, you can configure the switch to only forward unregistered packets to ports on a VLAN that are connected to multicast routers (mrouter ports) using the command no ip igmp snooping flood from CONFIGURATION mode. When flooding is disabled, if there are no such ports in the VLAN connected to a multicast router, the switch drops the packets. On the C-Series and S-Series, when you configure no ip igmp snooping flood, the system drops the packets immediately. The system does not forward the frames on mrouter ports, even if they are present. On the C-Series and S-Series, Layer 3 multicast must be disabled (no ip multicast-routing) in order to disable multicast flooding.

Specifying a Port as Connected to a Multicast Router

snooping mrouter from

You can statically specify a port in a VLAN as connected to a multicast router using the command ip igmp INTERFACE VLAN mode. View the ports that are connected to multicast routers using the command show ip igmp snooping mrouter from EXEC Privilege mode.

Configuring the Switch as Querier

Hosts that do not support unsolicited reporting wait for a general query before sending a membership report. When the multicast source and receivers are in the same VLAN, multicast traffic is not routed, and so there is no querier. You must configure the switch to be the querier for a VLAN so that hosts send membership reports, and the switch can generate a forwarding table by snooping. Configure the switch to be the querier for a VLAN by first assigning an IP address to the VLAN interface, and then using the command ip igmp snooping querier from INTERFACE VLAN mode. IGMP snooping Querier does not start if there is a statically configured multicast router interface in the VLAN. The switch may lose the querier election if it does not have the lowest IP address of all potential queriers on the subnet.

Internet Group Management Protocol | 295

www.dell.com | support.dell.com

When enabled, IGMP snooping Querier starts after one query interval in case no IGMP general query (with IP SA lower than its VLAN IP address) is received on any of its VLAN members.

Adjusting the Last Member Query Interval

When the querier receives a leave message from a receiver, it sends a group-specific query out of the ports specified in the forwarding table. If no response is received, it sends another. The amount of time that the querier waits to receive a response to the initial query before sending a second one is the Last Member Query Interval (LMQI). The switch waits one LMQI after the second query before removing the group-port entry from the forwarding table. Adjust the Last Member Query Interval using the command ip igmp snooping last-member-query-interval from INTERFACE VLAN mode.

Fast Convergence after MSTP Topology Changes

When a port transitions to the Forwarding state as a result of an STP or MSTP topology change, FTOS sends a general query out of all ports except the multicast router ports. The host sends a response to the general query and the forwarding database is updated without having to wait for the query interval to expire. When an IGMP snooping switch is not acting as a Querier it sends out the general query, in response to the MSTP triggered link-layer topology change, with the source IP address of 0.0.0.0 to avoid triggering Querier election.

Designating a Multicast Router Interface

You can designate an interface as a multicast router interface with the command ip igmp snooping mrouter interface. FTOS also has the capability of listening in on the incoming IGMP General Queries and designate those interfaces as the multicast router interface when the frames have a non-zero IP source address. All IGMP control packets and IP multicast data traffic originating from receivers is forwarded to multicast router interfaces.

296

Internet Group Management Protocol

15
Interfaces
This chapter describes interface types, both physical and logical, and how to configure them with FTOS. 10/100/1000 Mbps Ethernet, Gigabit Ethernet, and 10 Gigabit Ethernet interfaces are supported on platforms

ces e and are covered in the SONET/SDH chapter of

SONET interfaces are only supported on platform the E-Series FTOS Configuration Guide.

Basic Interface Configuration:

Interface Types View Basic Interface Information Enable a Physical Interface Physical Interfaces Management Interfaces VLAN Interfaces Loopback Interfaces Null Interfaces Port Channel Interfaces

Advanced Interface Configuration:

Bulk Configuration Interface Range Macros Monitor and Maintain Interfaces Link Debounce Timer Link Dampening Ethernet Pause Frames Configure MTU Size on an Interface Port-pipes Auto-Negotiation on Ethernet Interfaces

Interfaces | 297

www.dell.com | support.dell.com

View Advanced Interface Information

Interface Types
Interface Type Physical Management Loopback Null Port Channel VLAN Modes Possible L2, L3 N/A L3 N/A L2, L3 L2, L3 Default Mode Unset N/A L3 N/A L3 L2 Requires Creation No No Yes No Yes Yes (except default) Default State Shutdown (disabled) No Shutdown (enabled) No Shutdown (enabled) Enabled Shutdown (disabled) L2 - No Shutdown (enabled) L3 - Shutdown (disabled)

View Basic Interface Information

The user has several options for viewing interface status and configuration parameters. The show interfaces command in EXEC mode will list all configurable interfaces on the chassis and has options to display the interface status, IP and MAC addresses, and multiple counters for the amount and type of traffic passing through the interface. If a port channel interface is configured, the show interfaces command can list the interfaces configured in the port channel.
Note: To end output from the system, such as the output from the show interfaces command, enter CTRL+C and FTOS will return to the command prompt.

Figure 15-1 displays the configuration and status information for one interface.

298

Interfaces

Figure 15-1. show interfaces Command Example

FTOS#show interfaces tengigabitethernet 1/0 TenGigabitEthernet 1/0 is up, line protocol is up Hardware is Force10Eth, address is 00:01:e8:05:f3:6a Current address is 00:01:e8:05:f3:6a Pluggable media present, XFP type is 10GBASE-LR. Medium is MultiRate, Wavelength is 1310nm XFP receive power reading is -3.7685 Interface index is 67436603 Internet address is 65.113.24.238/28 MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 10000 Mbit, Mode full duplex, Master ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:09:54 Queueing strategy: fifo Input Statistics: 0 packets, 0 bytes 0 Vlans 0 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 0 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 3 packets, 192 bytes, 0 underruns 3 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts

Use the show ip interfaces brief command in the EXEC Privilege mode to view which interfaces are enabled for Layer 3 data transmission. In Figure 15-2, GigabitEthernet interface 1/5 is in Layer 3 mode since an IP address has been assigned to it and the interfaces status is operationally up.
Figure 15-2. show ip interfaces brief Command Example (Partial)
FTOS#show ip interface Interface GigabitEthernet 1/0 GigabitEthernet 1/1 GigabitEthernet 1/2 GigabitEthernet 1/3 GigabitEthernet 1/4 GigabitEthernet 1/5 GigabitEthernet 1/6 GigabitEthernet 1/7 GigabitEthernet 1/8 brief IP-Address unassigned unassigned unassigned unassigned unassigned 10.10.10.1 unassigned unassigned unassigned

OK? NO NO YES YES YES YES NO NO NO

Method Manual Manual Manual Manual Manual Manual Manual Manual Manual

Status administratively administratively up up up up administratively administratively administratively

Protocol down down down down up up up up down down down down down down

Use the show interfaces configured command in the EXEC Privilege mode to view only configured interfaces. In Figure 15-2, GigabitEthernet interface 1/5 is in Layer 3 mode since an IP address has been assigned to it and the interfaces status is operationally up. To determine which physical interfaces are available, use the show running-config command in EXEC mode. This command displays all physical interfaces available on the line cards. (Figure 158).

Interfaces | 299

www.dell.com | support.dell.com

Figure 15-3. Interfaces listed in the show running-config Command (Partial)

FTOS#show running Current Configuration ... ! interface GigabitEthernet no ip address shutdown ! interface GigabitEthernet no ip address shutdown ! interface GigabitEthernet no ip address shutdown ! interface GigabitEthernet no ip address shutdown

9/6

9/7

9/8

9/9

Enable a Physical Interface

After determining the type of physical interfaces available, the user may enter the INTERFACE mode by entering the command interface interface slot/port to enable and configure the interfaces. To enter the INTERFACE mode, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
interface interface

Command Mode CONFIGURATION

Purpose Enter the keyword interface followed by the type of interface and slot/port information: For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For the Management interface on the RPM, enter the keyword ManagementEthernet followed by the slot/port information. For a SONET interface, enter the keyword sonet followed by slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. Enter the no shutdown command to enable the interface. If the interface is a SONET interface, enter the encap ppp command to enable PPP encapsulation.

no shutdown

INTERFACE

300

Interfaces

To confirm that the interface is enabled, use the show config command in the INTERFACE mode. To leave the INTERFACE mode, use the exit command or end command. The user can not delete a physical interface.

Physical Interfaces
The Management Ethernet interface, is a single RJ-45 Fast Ethernet port on the Route Processor Module (RPM) of the C-Series and E-Series and on each unit of the S60; it provides dedicated management access to the system. The other S-Series (non-S60) systems supported by FTOS do not have this dedicated management interface, but you can use any Ethernet port configured with an IP address and route. Line card interfaces support Layer 2 and Layer 3 traffic over the 10/100/1000, Gigabit, and 10-Gigabit Ethernet interfaces. SONET interfaces with PPP encapsulation support Layer 3 traffic. These interfaces (except SONET interfaces with PPP encapsulation) can also become part of virtual interfaces such as VLANs or port channels. Link detection on ExaScale line cards is interrupt-based rather than poll-based, which enables ExaScale cards to bring up and take down links faster. For more information on VLANs, see Bulk Configuration and for more information on port channels, see Port Channel Interfaces.
FTOS Behavior: S-Series systems use a single MAC address for all physical interfaces while E-Series and C-Series use a unique MAC address for each physical interface, though this results in no functional difference between these platforms.

Configuration Task List for Physical Interfaces

By default, all interfaces are operationally disabled and traffic will not pass through them. The following section includes information about optional configurations for physical interfaces: Overview of Layer Modes Configure Layer 2 (Data Link) Mode Management Interfaces Auto-Negotiation on Ethernet Interfaces Adjust the keepalive timer Clear interface counters

Interfaces | 301

www.dell.com | support.dell.com

Overview of Layer Modes

On all systems running FTOS, you can place physical interfaces, port channels, and VLANs in Layer 2 mode or Layer 3 mode. By default, VLANs are in Layer 2 mode.
Table 15-1. Interfaces Types Possible Modes Layer 2 Layer 3 Layer 3 n/a Layer 3 n/a Layer 2 Layer 3 Layer 2 Layer 3 Requires Creation No No No Yes No Yes Yes, except for the default VLAN

Type of Interface 10/100/1000 Ethernet, Gigabit Ethernet, 10 Gigabit Ethernet SONET (PPP encapsulation) Management Loopback Null interface Port Channel VLAN

Default State Shutdown (disabled) Shutdown (disabled) Shutdown (disabled) No shutdown (enabled) Enabled Shutdown (disabled) No shutdown (active for Layer 2) Shutdown (disabled for Layer 3)

Configure Layer 2 (Data Link) Mode

Use the switchport command in INTERFACE mode to enable Layer 2 data transmissions through an individual interface. The user can not configure switching or Layer 2 protocols such as spanning tree protocol on an interface unless the interface has been set to Layer 2 mode. Figure 15-4 displays the basic configuration found in a Layer 2 interface.
Figure 15-4. show config Command Example of a Layer 2 Interface
FTOS(conf-if)#show config ! interface Port-channel 1 no ip address switchport no shutdown FTOS(conf-if)#

To configure an interface in Layer 2 mode, use these commands in the INTERFACE mode:
Command Syntax
no shutdown switchport

Command Mode INTERFACE INTERFACE

Purpose Enable the interface. Place the interface in Layer 2 (switching) mode.

302

Interfaces

For information on enabling and configuring Spanning Tree Protocol, see Chapter 10, Layer 2, on page 47. To view the interfaces in Layer 2 mode, use the command show interfaces switchport in the EXEC mode.

Configure Layer 3 (Network) Mode

When you assign an IP address to a physical interface, you place it in Layer 3 mode. Use the ip address command and no shutdown command in INTERFACE mode to enable Layer 3 mode on an individual interface. In all interface types except VLANs, the shutdown command prevents all traffic from passing through the interface. In VLANs, the shutdown command prevents Layer 3 traffic from passing through the interface. Layer 2 traffic is unaffected by the shutdown command. One of the interfaces in the system must be in Layer 3 mode before you configure or enter a Layer 3 protocol mode (for example, OSPF). Figure 15-5 shows how the show config command displays an example of a Layer 3 interface.
Figure 15-5. show config Command Example of a Layer 3 Interface
FTOS(conf-if)#show config ! interface GigabitEthernet 1/5 ip address 10.10.10.1 /24 no shutdown FTOS(conf-if)#

If an interface is in the incorrect layer mode for a given command, an error message is displayed to the user. For example, in Figure 15-6, the command ip address triggered an error message because the interface is in Layer 2 mode and the ip address command is a Layer 3 command only.
Figure 15-6. Error Message When Trying to Add an IP Address to Layer 2 Interface
FTOS(conf-if)#show config ! interface GigabitEthernet 1/2 no ip address switchport no shutdown FTOS(conf-if)#ip address 10.10.1.1 /24 % Error: Port is in Layer 2 mode Gi 1/2. FTOS(conf-if)#

Error message

To determine the configuration of an interface, you can use the show config command in INTERFACE mode or the various show interface commands in EXEC mode. To assign an IP address, use both of the following commands in the INTERFACE mode:
Command Syntax
no shutdown

Command Mode INTERFACE

Purpose Enable the interface.

Interfaces | 303

www.dell.com | support.dell.com

Command Syntax
ip address ip-address mask [secondary]

Command Mode INTERFACE

Purpose Configure a primary IP address and mask on the interface. The ip-address must be in dotted-decimal format (A.B.C.D) and the mask must be in slash format (/xx). Add the keyword secondary if the IP address is the interfaces backup IP address.

You can only configure one (1) primary IP address per interface. You can configure up to 255 secondary IP addresses on a single interface. To view all interfaces to see with an IP address assigned, use the show ip interfaces brief command in the EXEC mode (Figure 176). To view IP information on an interface in Layer 3 mode, use the show ip interface command in the EXEC Privilege mode (Figure 159).
Figure 15-7. Command Example: show ip interface
FTOS>show ip int vlan 58 Vlan 58 is up, line protocol is up Internet address is 1.1.49.1/24 Broadcast address is 1.1.49.255 Address determined by config file MTU is 1554 bytes Inbound access list is not set Proxy ARP is enabled Split Horizon is enabled Poison Reverse is disabled ICMP redirects are not sent ICMP unreachables are not sent

Management Interfaces
The S60 system supports the Management Ethernet interface as well as the standard S-Series interface on any port. Either method can be used to connect to the system.

Configure Management Interfaces on the E-Series and C-Series and on the S60
On the E-Series, C-Series, and S60 the dedicated Management interface provides management access to the system. You can configure this interface with FTOS, but the configuration options on this interface are limited. Gateway addresses and IP addresses cannot be configured if it appears in the main routing table of FTOS. In addition, Proxy ARP is not supported on this interface.
Note: On the S60, a default IP address is assigned to the Management port. Use this IP address to set your laptop Ethernet port to the same network for test purposes.

304

Interfaces

To configure a Management interface, use the following command in the CONFIGURATION mode:
Command Syntax
interface Managementethernet interface

Command Mode CONFIGURATION

Purpose Enter the slot and the port (0). ON the E-Series and C-Series, dual RPMs can be in use. Slot range: C-Series, E-Series: 0-1 S60: 0

To view the Primary RPM Management port, use the show interface Managementethernet command in the EXEC Privilege mode. If there are 2 RPMs, the you cannot view information on that interface. To configure IP addresses on a Management interface, use the following command in the MANAGEMENT INTERFACE mode:
Command Syntax
ip address ip-address mask

Command Mode INTERFACE

Purpose Configure an IP address and mask on the interface. ip-address mask: enter an address in dotted-decimal format (A.B.C.D), the mask must be in /prefix format (/x)

If there are 2 RPMs on the system, each Management interface must be configured with a different IP address. Unless the management route command is configured, you can only access the Management interface from the local LAN. To access the Management interface from another LAN, the management route command must be configured to point to the Management interface. Alternatively, you can use virtual-ip to manage a system with one or two RPMs. A virtual IP is an IP address assigned to the system (not to any management interfaces) and is a CONFIGURATION mode command. When a virtual IP address is assigned to the system, the active management interface of the RPM is recognized by the virtual IP addressnot by the actual interface IP address assigned to it. During an RPM failover, you do not have to remember the IP address of the new RPMs management interface the system will still recognizes the virtual-IP address.

Important Things to Remember virtual-ip

virtual-ip is a CONFIGURATION mode command.

When applied, the management port on the primary RPM assumes the virtual IP address. Executing show interfaces and show ip interface brief commands on the primary RPM management interface will display the virtual IP address and not the actual IP address assigned on that interface. A duplicate IP address message is printed for management ports virtual IP address on an RPM failover. This is a harmless error that is generated due to a brief transitory moment during failover when both RPMs management ports own the virtual IP address, but have different MAC addresses. The primary management interface will use only the virtual IP address if it is configured. The system can not be accessed through the native IP address of the primary RPMs management interface.

Interfaces | 305

www.dell.com | support.dell.com

Once the virtual IP address is removed, the system is accessible through the native IP address of the primary RPMs management interface. Primary and secondary management interface IP and virtual IP must be in the same subnet.

Configure Management Interfaces on the S-Series

The user can manage the S-Series from any port. Configure an IP address for the port using the ip address command, and enable it using the command no shutdown. The user may use the command description from INTERFACE mode to note that the interface is the management interface. There is no separate management routing table, so the user must configure all routes in the IP routing table (the ip route command). As shown in Figure 15-8, from EXEC Privilege mode, display the configuration for a given port by entering the command show interface, and the routing table with the show ip route command.
Figure 15-8. Viewing Management Routes on the S-Series
FTOS#show int gig 0/48 GigabitEthernet 0/48 is up, line protocol is up Description: This is the Managment Interface Hardware is Force10Eth, address is 00:01:e8:cc:cc:ce Current address is 00:01:e8:cc:cc:ce Pluggable media not present Interface index is 46449666 Internet address is 10.11.131.240/23 [output omitted] FTOS#show ip route Codes: C - connected, S - static, R - RIP, B - BGP, IN - internal BGP, EX - external BGP,LO - Locally Originated, O - OSPF, IA - OSPF inter area, N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2, E1 - OSPF external type 1, E2 - OSPF external type 2, i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, IA - IS-IS inter area, * - candidate default, > - non-active route, + - summary route Gateway of last resort is 10.11.131.254 to network 0.0.0.0 Destination ----------0.0.0.0/0 10.11.130.0/23 Gateway ------via 10.11.131.254, Gi 0/48 Direct, Gi 0/48 Dist/Metric Last Change ----------- ----------1/0 1d2h 0/0 1d2h

*S C FTOS#

306

Interfaces

VLAN Interfaces
VLANs are logical interfaces and are, by default, in Layer 2 mode. Physical interfaces and port channels can be members of VLANs. For more information on VLANs and Layer 2, refer to Chapter 10, Layer 2, on page 47. See also Chapter 18, VLAN Stacking, on page 367.
Note: To monitor VLAN interfaces, use the Management Information Base for Network Management of TCP/IP-based internets: MIB-II (RFC 1213). Monitoring VLAN interfaces via SNMP is supported only on E-Series.

FTOS supports Inter-VLAN routing (Layer 3 routing in VLANs). You can add IP addresses to VLANs and use them in routing protocols in the same manner that physical interfaces are used. For more information on configuring different routing protocols, refer to the chapters on the specific protocol. A consideration for including VLANs in routing protocols is that the no shutdown command must be configured. (For routing traffic to flow, the VLAN must be enabled.)
Note: An IP address cannot be assigned to the Default VLAN, which, by default, is VLAN 1. To assign another VLAN ID to the Default VLAN, use the default vlan-id vlan-id command.

Assign an IP address to an interface with the following command the INTERFACE mode:
Command Syntax
ip address ip-address mask [secondary]

Command Mode INTERFACE

Purpose Configure an IP address and mask on the interface. ip-address mask: enter an address in dotted-decimal format (A.B.C.D) and the mask must be in slash format (/24). secondary: the IP address is the interfaces backup IP address. You can configure up to eight secondary IP addresses.

Figure 15-9 shows a sample configuration of a VLAN participating in an OSPF process.

Figure 15-9. Sample Layer 3 Configuration of a VLAN
interface Vlan 10 ip address 1.1.1.2/24 tagged GigabitEthernet 2/2-13 tagged TenGigabitEthernet 5/0 ip ospf authentication-key FTOS ip ospf cost 1 ip ospf dead-interval 60 ip ospf hello-interval 15 no shutdown !

Interfaces | 307

www.dell.com | support.dell.com

Loopback Interfaces
A Loopback interface is a virtual interface in which the software emulates an interface. Packets routed to it are processed locally. Since this interface is not a physical interface, you can configure routing protocols on this interface to provide protocol stability. You can place Loopback interfaces in default Layer 3 mode. To configure a Loopback interface, use the following command in the CONFIGURATION mode:
Command Syntax
interface loopback number

Command Mode CONFIGURATION

Purpose Enter a number as the loopback interface. Range: 0 to 16383.

To view Loopback interface configurations, use the show interface loopback number command in the EXEC mode. To delete a Loopback interface, use the no interface loopback number command syntax in the CONFIGURATION mode. Many of the same commands found in the physical interface are found in Loopback interfaces. See also Configuring ACLs to Loopback.

Null Interfaces
The Null interface is another virtual interface created by the E-Series software. There is only one Null interface. It is always up, but no traffic is transmitted through this interface. To enter the INTERFACE mode of the Null interface, use the following command in the CONFIGURATION mode:
Command Syntax
interface null 0

Command Mode CONFIGURATION

Purpose Enter the INTERFACE mode of the Null interface.

The only configurable command in the INTERFACE mode of the Null interface is the ip unreachable command.

308

Interfaces

Port Channel Interfaces

Port channel interfaces support link aggregation, as described in IEEE Standard 802.3ad. This section covers the following topics: Port channel definition and standards Port channel benefits Port channel implementation Configuration task list for port channel interfaces

Port channel definition and standards

Link aggregation is defined by IEEE 802.3ad as a method of grouping multiple physical interfaces into a single logical interfacea Link Aggregation Group (LAG) or port channel. A LAG is a group of links that appear to a MAC client as if they were a single link according to IEEE 802.3ad. In FTOS, a LAG is referred to as a port channel interface. A port channel provides redundancy by aggregating physical interfaces into one logical interface. If one physical interface goes down in the port channel, another physical interface carries the traffic.

Port channel benefits

For the E-Series, a port channel interface provides many benefits, including easy management, link redundancy, and sharing. Port channels are transparent to network configurations and can be modified and managed as one interface. For example, you configure one IP address for the group and that IP address is used for all routed traffic on the port channel. With this feature, the user can create larger-capacity interfaces by utilizing a group of lower-speed links. For example, the user can build a 5-Gigabit interface by aggregating five 1-Gigabit Ethernet interfaces together. If one of the five interfaces fails, traffic is redistributed across the four remaining interfaces.

Port channel implementation

FTOS supports two types of port channels: StaticPort channels that are statically configured

Interfaces | 309

www.dell.com | support.dell.com

DynamicPort channels that are dynamically configured using Link Aggregation Control Protocol (LACP). For details, see Chapter 19, Link Aggregation Control Protocol.
Number of Port-channels per Platform Port-channels 255 128 52 128 Members/Channel 16 8 8 8

Table 15-2. Platform E-Series C-Series

S-Series: S50 and S25 S-Series: S55, S60 and S4810 Table 15-3. Platform E-Series ExaScale

Maximum number of configurable Port-channels Port-channels 512 Members/Channel 64

As soon as a port channel is configured, FTOS treats it like a physical interface. For example, IEEE 802.1Q tagging is maintained while the physical interface is in the port channel. Member ports of a LAG are added and programmed into hardware in a predictable order based on the port ID, instead of in the order in which the ports come up. With this implementation, load balancing yields predictable results across line card resets and chassis reloads. A physical interface can belong to only one port channel at a time. Each port channel must contain interfaces of the same interface type/speed. Port channels can contain a mix of 10, 100, or 1000 Mbps Ethernet interfaces and Gigabit Ethernet interfaces, and the interface speed (10, 100, or 1000 Mbps) used by the port channel is determined by the first port channel member that is physically up. FTOS disables the interfaces that do match the interface speed set by the first channel member. That first interface may be the first interface that is physically brought up or was physically operating when interfaces were added to the port channel. For example, if the first operational interface in the port channel is a Gigabit Ethernet interface, all interfaces at 1000 Mbps are kept up, and all 10/100/1000 interfaces that are not set to 1000 speed or auto negotiate are disabled. FTOS brings up 10/100/1000 interfaces that are set to auto negotiate so that their speed is identical to the speed of the first channel member in the port channel.

10/100/1000 Mbps interfaces in port channels

When both 10/100/1000 interfaces and GigE interfaces are added to a port channel, the interfaces must share a common speed. When interfaces have a configured speed different from the port channel speed, the software disables those interfaces.

310

Interfaces

The common speed is determined when the port channel is first enabled. At that time, the software checks the first interface listed in the port channel configuration. If that interface is enabled, its speed configuration becomes the common speed of the port channel. If the other interfaces configured in that port channel are configured with a different speed, FTOS disables them. For example, if four interfaces (Gi 0/0, 0/1, 0/2, 0/3) in which Gi 0/0 and Gi 0/3 are set to speed 100 Mb/s and the others are set to 1000 Mb/s, with all interfaces enabled, and you add them to a port channel by entering channel-member gigabitethernet 0/0-3 while in the port channel interface mode, and FTOS determines if the first interface specified (Gi 0/0) is up. Once it is up, the common speed of the port channel is 100 Mb/s. FTOS disables those interfaces configured with speed 1000 or whose speed is 1000 Mb/s as a result of auto-negotiation. In this example, you can change the common speed of the port channel by changing its configuration so the first enabled interface referenced in the configuration is a 1000 Mb/s speed interface. You can also change the common speed of the port channel here by setting the speed of the Gi 0/0 interface to 1000 Mb/s.

Configuration task list for port channel interfaces

To configure a port channel (LAG), you use the commands similar to those found in physical interfaces. By default, no port channels are configured in the startup configuration. Create a port channel (mandatory) Add a physical interface to a port channel (mandatory) Reassign an interface to a new port channel (optional) Configure the minimum oper up links in a port channel (LAG) (optional) Add or remove a port channel from a VLAN (optional) Assign an IP address to a port channel (optional) Delete or disable a port channel (optional) Load balancing through port channels (optional)

Create a port channel

You can create up to 255 port channels on an E-Series. You can create up to 128 port channels on an C-Series, 52 port channels with 8 port members per group on an S-Series S50 or S25, and 128 port channels with 8 port members per group on an S-Series S55, S60 and S4810. To configure a port channel, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
interface port-channel id-number no shutdown

Command Mode CONFIGURATION INTERFACE PORT-CHANNEL

Purpose Create a port channel. Ensure that the port channel is active.

Interfaces | 311

www.dell.com | support.dell.com

The port channel is now enabled and you can place the port channel in Layer 2 or Layer 3 mode. Use the switchport command to place the port channel in Layer 2 mode or configure an IP address to place the port channel in Layer 3 mode. You can configure a port channel as you would a physical interface by enabling or configuring protocols or assigning access control lists.

Add a physical interface to a port channel

The physical interfaces in a port channel can be on any line card in the chassis, but must be the same physical type.
Note: Port channels can contain a mix of Gigabit Ethernet and 10/100/1000 Ethernet interfaces, but FTOS disables the interfaces that are not the same speed of the first channel member in the port channel (see 10/100/1000 Mbps interfaces in port channels).

You can add any physical interface to a port channel if the interface configuration is minimal. Only the following commands can be configured on an interface if it is a member of a port channel:
description shutdown/no shutdown mtu ip mtu (if the interface is on a Jumbo-enabled by default.)

Note: The S-Series supports jumbo frames by default (the default maximum transmission unit

(MTU) is 1554 bytes) You can configure the MTU using the mtu command from INTERFACE mode.

To view the interfaces configuration, enter the INTERFACE mode for that interface and enter the show config command or from the EXEC Privilege mode, enter the show running-config interface interface command. When an interface is added to a port channel, FTOS recalculates the hash algorithm. To add a physical interface to a port channel, use these commands in the following sequence in the INTERFACE mode of a port channel:
Step 1 Command Syntax
channel-member interface

Command Mode INTERFACE PORT-CHANNEL INTERFACE PORT-CHANNEL

Purpose Add the interface to a port channel. The interface variable is the physical interface type and slot/port information. Double check that the interface was added to the port channel.

show config

To view the port channels status and channel members in a tabular format, use the show interfaces port-channel brief (Figure 177) command in the EXEC Privilege mode.

312

Interfaces

Figure 15-10.

show interfaces port-channel brief Command Example

FTOS#show int port brief LAG Mode 1 L2L3 2 L2L3 Status up up Uptime 00:06:03 00:06:03 Ports Gi 13/6 Gi 13/12 Gi 13/7 Gi 13/8 Gi 13/13 Gi 13/14

(Up) * (Up) (Up) * (Up) (Up) (Up)

FTOS#

Figure 15-11 displays the port channels mode (L2 for Layer 2 and L3 for Layer 3 and L2L3 for a Layer 2 port channel assigned to a routed VLAN), the status, and the number of interfaces belonging to the port channel.
Figure 15-11. show interface port-channel Command Example
FTOS>show interface port-channel 20 Port-channel 20 is up, line protocol is up Hardware address is 00:01:e8:01:46:fa Internet address is 1.1.120.1/24 MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 2000 Mbit Members in this channel: Gi 9/10 Gi 9/17 ARP type: ARPA, ARP timeout 04:00:00 Last clearing of "show interface" counters 00:00:00 Queueing strategy: fifo 1212627 packets input, 1539872850 bytes Input 1212448 IP Packets, 0 Vlans 0 MPLS 4857 64-byte pkts, 17570 over 64-byte pkts, 35209 over 127-byte pkts 69164 over 255-byte pkts, 143346 over 511-byte pkts, 942523 over 1023-byte pkts Received 0 input symbol errors, 0 runts, 0 giants, 0 throttles 42 CRC, 0 IP Checksum, 0 overrun, 0 discarded 2456590833 packets output, 203958235255 bytes, 0 underruns Output 1640 Multicasts, 56612 Broadcasts, 2456532581 Unicasts 2456590654 IP Packets, 0 Vlans, 0 MPLS 0 throttles, 0 discarded Rate info (interval 5 minutes): Input 00.01Mbits/sec, 2 packets/sec Output 81.60Mbits/sec, 133658 packets/sec Time since last interface status change: 04:31:57 FTOS>

When more than one interface is added to a Layer 2 port channel, FTOS selects one of the active interfaces in the port channel to be the Primary Port. The primary port replies to flooding and sends protocol PDUs. An asterisk in the show interfaces port-channel brief command indicates the primary port. As soon as a physical interface is added to a port channel, the properties of the port channel determine the properties of the physical interface. The configuration and status of the port channel are also applied to the physical interfaces within the port channel. For example, if the port channel is in Layer 2 mode, you cannot add an IP address or a static MAC address to an interface that is part of that port channel. As Figure 15-12 illustrates, interface GigabitEthernet 1/6 is part of port channel 5, which is in Layer 2 mode, and an error message appeared when an IP address was configured.

Interfaces | 313

www.dell.com | support.dell.com

Figure 15-12.

Error Message

FTOS(conf-if-portch)#show config ! interface Port-channel 5 no ip address switchport channel-member GigabitEthernet 1/6 FTOS(conf-if-portch)#int gi 1/6 FTOS(conf-if)#ip address 10.56.4.4 /24 % Error: Port is part of a LAG Gi 1/6. FTOS(conf-if)#

Error message

Reassign an interface to a new port channel

An interface can be a member of only one port channel. If the interface is a member of a port channel, you must remove it from the first port channel and then add it to the second port channel. Each time you add or remove a channel member from a port channel, FTOS recalculates the hash algorithm for the port channel. To reassign an interface to a new port channel, use these commands in the following sequence in the INTERFACE mode of a port channel:
Step 1 Command Syntax
no channel-member interface interface port-channel id number channel-member interface

Command Mode INTERFACE PORT-CHANNEL INTERFACE PORT-CHANNEL INTERFACE PORT-CHANNEL

Purpose Remove the interface from the first port channel. Change to the second port channel INTERFACE mode. Add the interface to the second port channel.

2 3

Figure 15-13 displays an example of moving the GigabitEthernet 1/8 interface from port channel 4 to port channel 3.

314

Interfaces

Figure 15-13.

Command Example from Reassigning an Interface to a Different Port Channel

FTOS(conf-if-portch)#show config ! interface Port-channel 4 no ip address channel-member GigabitEthernet 1/8 no shutdown FTOS(conf-if-portch)#no chann gi 1/8 FTOS(conf-if-portch)#int port 5 FTOS(conf-if-portch)#channel gi 1/8 FTOS(conf-if-portch)#sho conf ! interface Port-channel 5 no ip address channel-member GigabitEthernet 1/8 shutdown FTOS(conf-if-portch)#

Configure the minimum oper up links in a port channel (LAG)

You can configure the minimum links in a port channel (LAG) that must be in oper up status for the port channel to be considered to be in oper up status. Use the following command in the INTERFACE mode:
Command Syntax
minimum-links number

Command Mode INTERFACE

Purpose Enter the number of links in a LAG that must be in oper up status. Default: 1

Figure 15-14 displays an example of configuring five minimum oper up links in a port channel.
Figure 15-14. Example of using the minimum-links Command

FTOS#config t FTOS(conf)#int po 1 FTOS(conf-if-po-1)#minimum-links 5 FTOS(conf-if-po-1)#

Add or remove a port channel from a VLAN

As with other interfaces, you can add Layer 2 port channel interfaces to VLANs. To add a port channel to a VLAN, you must place the port channel in Layer 2 mode (by using the switchport command).

Interfaces | 315

www.dell.com | support.dell.com

To add a port channel to a VLAN, use either of the following commands:

Command Syntax
tagged port-channel id number

Command Mode INTERFACE VLAN INTERFACE VLAN

Purpose Add the port channel to the VLAN as a tagged interface. An interface with tagging enabled can belong to multiple VLANs. Add the port channel to the VLAN as an untagged interface. An interface without tagging enabled can belong to only one VLAN.

untagged port-channel id number

To remove a port channel from a VLAN, use either of the following commands:
Command Syntax
no tagged port-channel id number no untagged port-channel id number

Command Mode INTERFACE VLAN INTERFACE VLAN

Purpose Remove the port channel with tagging enabled from the VLAN. Remove the port channel without tagging enabled from the VLAN.

To see which port channels are members of VLANs, enter the show vlan command in the EXEC Privilege mode.

Assign an IP address to a port channel

You can assign an IP address to a port channel and use port channels in Layer 3 routing protocols. To assign an IP address, use the following command in the INTERFACE mode:
Command Syntax
ip address ip-address mask [secondary]

Command Mode INTERFACE

Purpose Configure an IP address and mask on the interface. ip-address mask: enter an address in dotted-decimal format (A.B.C.D) and the mask must be in slash format (/24). secondary: the IP address is the interfaces backup IP address. You can configure up to eight secondary IP addresses.

Delete or disable a port channel

To delete a port channel, you must be in the CONFIGURATION mode and use the no interface portchannel channel-number command. When you disable a port channel (using the shutdown command) all interfaces within the port channel are operationally down also.

316

Interfaces

Load balancing through port channels

FTOS uses hash algorithms for distributing traffic evenly over channel members in a port channel (LAG). The hash algorithm distributes traffic among ECMP paths and LAG members. The distribution is based on a flow, except for packet-based hashing. A flow is identified by the hash and is assigned to one link. In packet-based hashing, a single flow can be distributed on the LAG and uses one link. Packet based hashing is used to load balance traffic across a port-channel based on the IP Identifier field within the packet. Load balancing uses source and destination packet information to get the greatest advantage of resources by distributing traffic over multiple paths when transferring data to a destination. FTOS allows you to modify the hashing algorithms used for flows and for fragments. The load-balance and
hash-algorithm commands are available for modifying the distribution algorithms. Their syntax and

implementation are somewhat different between the E-Series and the C-Series and S-Series.
Note: Hash-based load-balancing on MPLS does not work when packet-based hashing (load-balance
ip-selection packet-based) is enabled.

E-Series load-balancing
On the E-Series, the default load-balance criteria are a 5-tuple, as follows: IP source address IP destination address Protocol type TCP/UDP source port TCP/UDP destination port

Balancing may be applied to IPv4, switched IPv6, and non-IP traffic. For these traffic types, the IP-header-based hash and MAC-based hash may be applied to packets by using the following methods.
Table 15-4. Hash Methods as Applied to Port Channel Types Layer 2 Port Channel X X X X Layer 3 Port Channel X X X

Hash (Header Based)

5-tuple 3-tuple Packet-based MAC source address (SA) and destination address (DA)

Interfaces | 317

www.dell.com | support.dell.com

On the E-Series, to change the 5-tuple default to 3-tuple, MAC, or packet-based, use the following command in CONFIGURATION mode:
Command Syntax [no] load-balance [ip-selection {3-tuple | packet-based}] [mac] Command Mode CONFIGURATION Purpose To designate a method to balance traffic over a port channel. By default, IP 5-tuple is used to distribute traffic over members port channel. ip-selection 3-tupleDistribute IP traffic based on IP source address, IP destination address, and IP protocol type. ip-selection packet-basedDistribute IPV4 traffic based on the IP Identification field in the IPV4 header. macDistribute traffic based on the MAC source address, and the MAC destination address. See Table 15-6 for more information.

For details on the load-balance command, see the IP Routing chapter of the FTOS Command Reference. To distribute IP traffic over an E-Series port channel member, FTOS uses the 5-tuple IP default. The 5-tuple and the 3-tuple hash use the following keys:
Table 15-5. Keys 5-tuple and 3-tuple Keys 5-tuple X X X X X 3-tuple X X X

IP source address (lower 32 bits) IP destination address (lower 32 bits) Protocol type TCP/UDP source port TCP/UDP destination port

Note: For IPV6, only the first 32 bits (LSB) of IP Source Address and IP Destination Address are used for hash generation.

Figure 15-15 shows the configuration and show command for packet-based hashing on the E-Series.
Figure 15-15. Command example: load-balance ip-selection packet-based

FTOS(conf)#load-balance ip-selection packet-based

FTOS#show running-config | grep load load-balance ip-selection packet-based FTOS#

The load-balance packet based command can co-exist with load balance mac command to achieve the functionality shown in Table 15-6.

318

Interfaces

IPv4, IPv6, and non-IP traffic handling on the E-Series

The table below presents the combinations of the load-balance command and their effect on traffic types.
Table 15-6. The load-balance Commands and Port Channel Types Switched IP Traffic IP 5-tuple (lower 32 bits) IP 3-tuple (lower 32 bits) MAC-based MAC-based Packet based: IPV4 No distribution: IPV6 MAC-based Routed IP Traffic (IPv4 only) IP 5-tuple IP 3-tuple IP 5-tuple IP 3-tuple Packet-based Packet-based Switched Non-IP Traffic MAC-based MAC-based MAC-based MAC-based MAC-based MAC-based

Configuration Commands

Default (IP 5-tuple)

load-balance ip-selection 3-tuple load-balance ip-selection mac load-balance ip-selection 3-tuple load-balance ip-selection mac load-balance ip-selection packet-based load-balance ip-selection packet-based load-balance ip-selection mac

C-Series and S-Series load-balancing

For LAG hashing on C-Series and S-Series, the source IP, destination IP, source TCP/UDP port, and destination TCP/UDP port are used for hash computation by default. For packets without a Layer 3 header, FTOS automatically uses load-balance mac source-dest-mac. IP hashing or MAC hashing should not be configured at the same time. If you configure an IP and MAC hashing scheme at the same time, the MAC hashing scheme takes precedence over the IP hashing scheme. To change the IP traffic load balancing default on the C-Series and S-Series, use the following command:
Command Syntax [no] load-balance {ip-selection [dest-ip | source-ip]} | {mac [dest-mac | source-dest-mac | source-mac]} | {tcp-udp enable} Command Mode CONFIGURATION Purpose Replace the default IP 4-tuple method of balancing traffic over a port channel. You can select one, two, or all three of the following basic hash methods ip-selection [dest-ip | source-ip]Distribute IP traffic based on IP destination or source address. mac [dest-mac | source-dest-mac | source-mac]Distribute IPV4 traffic based on the destination or source MAC address, or both, along with the VLAN, Ethertype, source module ID and source port ID. tcp-udp enableDistribute traffic based on TCP/ UDP source and destination ports.

Interfaces | 319

www.dell.com | support.dell.com

Hash algorithm
The load-balance command discussed above selects the hash criteria applied to port channels. If even distribution is not obtained with the load-balance command, the hash-algorithm command can be used to select the hash scheme for LAG, ECMP and NH-ECMP. The 12 bit Lag Hash can be rotated or shifted till the desired hash is achieved. The nh-ecmp option allows you to change the hash value for recursive ECMP routes independently of non-recursive ECMP routes. This option provides for better traffic distribution over available equal cost links that involve a recursive next hop lookup. For the E-Series TeraScale and ExaScale, you can select one of 47 possible hash algorithms.
Command Syntax
hash-algorithm {algorithm-number} | {ecmp {checksum|crc|xor} [number]} lag {checksum|crc|xor][number]}nh-ecmp {[checksum|crc|xor] [number]}}| {linecard number ip-sa-mask value ip-da-mask value}

Command Mode CONFIGURATION

Purpose Change the default (0) to another algorithm and apply it to ECMP, LAG hashing, or a particular line card. Note: To achieve the functionality of hash-align on the ExaScale platform, do not use CRC as an hash-algorithm method. For ExaScale systems, set the default hash-algorithm method to ensure CRC is not used for LAG. For example,
hash-algorithm ecmp xor lag checksum nh-ecmp checksum

For details on the algorithm choices, see the command details in the IP Routing chapter of the FTOS

Command Reference.

Note: E-Series systems require the lag-hash-align microcode be configured in the in the CAM profile. E-Series TeraScale et includes this microcode as an option with the Default cam profile. E-Series ExaScale ex systems require that a CAM profile be created and specifically include lag-hash-align microcode.

Figure 15-16 shows a sample configuration for the hash-algorithm command.

Figure 15-16. Command example: hash-algorithm

FTOS(conf)#hash-algorithm ecmp xor 26 lag crc 26 nh-ecmp checksum 26 FTOS(conf)#

On C-Series and S-Series, the hash-algorithm command is specific to ECMP groups and has different defaults from the E-Series. The default ECMP hash configuration is crc-lower. This takes the lower 32 bits of the hash key to compute the egress port. Other options for ECMP hash-algorithms are:
crc-upper uses the upper 32 bits of the hash key to compute the egress port dest-ip uses destination IP address as part of the hash key

320

Interfaces

lsb always uses the least significant bit of the hash key to compute the egress port

To change to another method, use the following command in the CONFIGURATION mode:
Command Syntax
hash-algorithm ecmp {crc-upper} | {dest-ip} | {lsb}

Command Mode CONFIGURATION

Purpose Change to another algorithm.

For more on load-balancing, see Equal Cost Multipath and Link Aggregation Frequently Asked Questions in the E-Series FAQ section (login required) of iSupport: https://www.force10networks.com/CSPortal20/KnowledgeBase/ToolTips.aspx

Bulk Configuration
Bulk configuration enables you to determine if interfaces are present, for physical interfaces, or, configured, for logical interfaces.

Interface Range
An interface range is a set of interfaces to which other commands may be applied, and may be created if there is at least one valid interface within the range. Bulk configuration excludes from configuration any non-existing interfaces from an interface range. A default VLAN may be configured only if the interface range being configured consists of only VLAN ports. The interface range command allows you to create an interface range allowing other commands to be applied to that range of interfaces. The interface range prompt offers the interface (with slot and port information) for valid interfaces. The maximum size of an interface range prompt is 32. If the prompt size exceeds this maximum, it displays (...) at the end of the output.
Note: Non-existing interfaces are excluded from interface range prompt. In the following example, Tengigabit 3/0 and VLAN 1000 do not exist.

Note: When creating an interface range, interfaces appear in the order they were entered and are not sorted.

The show range command is available under interface range mode. This command allows you to display all interfaces that have been validated under the interface range context. The show configuration command is also available under the interface range mode. This command allows you to display the running configuration only for interfaces that are part of interface range.

Interfaces | 321

www.dell.com | support.dell.com

Bulk Configuration Examples

The following are examples of using the interface range command for bulk configuration: Create a single-range Create a multiple-range Exclude duplicate entries Exclude a smaller port range Overlap port ranges Commas Add ranges

Create a single-range
Figure 15-17. Creating a Single-Range Bulk Configuration
FTOS(config)# interface range gigabitethernet 5/1 - 23 FTOS(config-if-range-gi-5/1-23)# no shutdown FTOS(config-if-range-gi-5/1-23)#

Create a multiple-range
Figure 15-18. Creating a Multiple-Range Prompt

FTOS(conf)#interface range tengigabitethernet 3/0 , gigabitethernet 2/1 - 47 , vlan 1000 FTOS(conf-if-range-gi-2/1-47,so-5/0)#

Exclude duplicate entries

Duplicate single interfaces and port ranges are excluded from the resulting interface range prompt:
Figure 15-19. Interface Range Prompt Excluding Duplicate Entries

FTOS(conf)#interface range vlan 1 , vlan 1 , vlan 3 , vlan 3 FTOS(conf-if-range-vl-1,vl-3)# FTOS(conf)#interface range gigabitethernet 2/0 - 23 , gigabitethernet 2/0 - 23 , gigab 2/0 - 23 FTOS(conf-if-range-gi-2/0-23)#

Exclude a smaller port range

If interface range has multiple port ranges, the smaller port range is excluded from prompt:
Figure 15-20. Interface Range Prompt Excluding a Smaller Port Range

FTOS(conf)#interface range gigabitethernet 2/0 - 23 , gigab 2/1 - 10 FTOS(conf-if-range-gi-2/0-23)#

322

Interfaces

Overlap port ranges

If overlapping port ranges are specified, the port range is extended to the smallest start port number and largest end port number:
Figure 15-21. Interface Range Prompt Including Overlapping Port Ranges

FTOS(conf)#inte ra gi 2/1 - 11 , gi 2/1 - 23 FTOS(conf-if-range-gi-2/1-23)#

Commas
The example below shows how to use commas to add different interface types to the range, enabling all Gigabit Ethernet interfaces in the range 5/1 to 5/23 and both Ten Gigabit Ethernet interfaces 1/1 and 1/2.
FTOS(config-if)# interface range gigabitethernet 5/1 - 23, tengigabitethernet 1/1 - 2 FTOS(config-if-range-gi-5/1-23)# no shutdown FTOS(config-if-range-gi-5/1-23)#

Figure 15-22.

Multiple-Range Bulk Configuration Gigabit Ethernet and Ten-Gigabit Ethernet

Add ranges
The example below shows how to use commas to add VLAN and port-channel interfaces to the range.
Figure 15-23. Multiple-Range Bulk Configuration with VLAN, and Port-channel

FTOS(config-ifrange-gi-5/1-23-te-1/1-2)# interface range Vlan 2 100 , Port 1 25 FTOS(config-if-range-gi-5/1-23-te-1/1-2-so-5/1-vl-2-100-po-1-25)# no shutdown FTOS(config-if-range)#

Interface Range Macros

The user can define an interface-range macro to automatically select a range of interfaces for configuration. Before you can use the macro keyword in the interface-range macro command string, you must define the macro. To define an interface-range macro, enter this command:
Command Syntax FTOS (config)# define interface-range macro_name {vlan vlan_ID - vlan_ID} | {{gigabitethernet | tengigabitethernet} slot/interface - interface} [ , {vlan vlan_ID - vlan_ID} {{gigabitethernet | tengigabitethernet} slot/interface - interface}] Command Mode CONFIGURATION Purpose Defines the interface-range macro and saves it in the running configuration file.

Interfaces | 323

www.dell.com | support.dell.com

Define the Interface Range

This example shows how to define an interface-range macro named test to select Fast Ethernet interfaces 5/1 through 5/4:
FTOS(config)# define interface-range test gigabitethernet 5/1 - 4

To show the defined interface-range macro configuration, use the command show running-config in the EXEC mode. The example below shows how to display the defined interface-range macro named test:

FTOS# show running-config | include define define interface-range test GigabitEthernet5/1 - 4 FTOS#

Choose an Interface-range Macro

To use an interface-range macro in the interface range command, enter this command:
Command Syntax
interface range macro name

Command Mode CONFIGURATION

Purpose Selects the interfaces range to be configured using the values saved in a named interface-range macro.

The example below shows how to change to the interface-range configuration mode using the interface-range macro named test.
FTOS(config)# interface range macro test FTOS(config-if)#

324

Interfaces

Monitor and Maintain Interfaces

Monitor interface statistics with the monitor interface command. This command displays an ongoing list of the interface status (up/down), number of packets, traffic statistics, etc.
Command Syntax
monitor interface interface

Command Mode EXEC Privilege

Purpose View the interfaces statistics. Enter the type of interface and slot/port information: For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For the Management interface on the RPM, enter the keyword ManagementEthernet followed by the slot/ port information. For a SONET interface, enter the keyword sonet followed by slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information.

The information (Figure 15-24) displays in a continuous run, refreshing every 2 seconds by default. Use the following keys to manage the output.
m - Change mode l - Page up T - Increase refresh interval (by 1 second) q - Quit c - Clear screen a - Page down t - Decrease refresh interval (by 1 second)

Interfaces | 325

www.dell.com | support.dell.com

Figure 15-24.

Command Example: monitor interface

FTOS#monitor interface gi 3/1

FTOS uptime is 1 day(s), 4 hour(s), 31 minute(s) Monitor time: 00:00:00 Refresh Intvl.: 2s Interface: Gi 3/1, Disabled, Link is Down, Linespeed is 1000 Mbit Traffic statistics: Input bytes: Output bytes: Input packets: Output packets: 64B packets: Over 64B packets: Over 127B packets: Over 255B packets: Over 511B packets: Over 1023B packets: Error statistics: Input underruns: Input giants: Input throttles: Input CRC: Input IP checksum: Input overrun: Output underruns: Output throttles: m l T q q FTOS# Current 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Rate Bps Bps pps pps pps pps pps pps pps pps pps pps pps pps pps pps pps pps Delta 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Change mode Page up Increase refresh interval Quit

c - Clear screen a - Page down t - Decrease refresh interval

Maintenance using TDR

The Time Domain Reflectometer (TDR) is supported on all Dell Networking switch/routers. TDR is an assistance tool to resolve link issues that helps detect obvious open or short conditions within any of the four copper pairs. TDR sends a signal onto the physical cable and examines the reflection of the signal that returns. By examining the reflection, TDR is able to indicate whether there is a cable fault (when the cable is broken, becomes unterminated, or if a transceiver is unplugged). TDR is useful for troubleshooting an interface that is not establishing a link, that is, when the link is flapping or not coming up. TDR is not intended to be used on an interface that is passing traffic. When a TDR test is run on a physical cable, it is important to shut down the port on the far end of the cable. Otherwise, it may lead to incorrect test results.
Note: TDR is an intrusive test. Do not run TDR on a link that is up and passing traffic.

326

Interfaces

To test the condition of cables on 10/100/1000 BASE-T modules, use the tdr-cable-test command:
Step 1 Command Syntax
tdr-cable-test gigabitethernet <slot>/
<port>

Command Mode EXEC Privilege

Usage To test for cable faults on the GigabitEthernet cable. Between two ports, the user must not start the test on both ends of the cable. The user must enable the interface before starting the test. The port should be enabled to run the test or the test prints an error message. Displays TDR test results.

show tdr gigabitethernet <slot>/

<port>

EXEC Privilege

Link Debounce Timer

Link Debounce Timer is supported on platform

The Link Debounce Timer feature isolates upper layer protocols on Ethernet switches and routers from very short-term, possibly repetitive interface flaps often caused by network jitter on the DWDM equipment connecting the switch and other devices on a SONET ring. The Link Debounce Timer delays link change notifications, thus decreasing traffic loss due to network configuration. All interfaces have a built-in timer to manage traffic. This feature extends the time allowed by the upper layers. The SONET ring has its own restore time whenever there is a failure. During this time, however, the Ethernet interface connected to the switch will flap. Link Debounce Timer instructs the Ethernet switch to delay the notification of the link change to the upper layers. If the link state changes again within this period, no notification goes to the upper layers, so that the switch remains unaware of the change.
Note: Enabling the link debounce timer causes link up and link down detections to be delayed, resulting in traffic being blackholed during the debouncing period. This situation might affect the convergence and reconvergence of some Layer 2 and Layer 3 protocols.

Important Points to Remember about Link Debounce Timer

Link Debounce Timer is configurable on physical ports only. Only 1G fiber, 10/100/1000 copper, 10G fiber, 10G copper are supported. This feature is not supported on management interfaces or SONET interfaces. Link Debounce takes effect only when the operational state of the port is up. Link Debounce is supported on interfaces that also have link dampening configured. Unlike link dampening, link debounce timer does not notify other protocols. Changes made do not affect any ongoing debounces. The timer changes take affect from the next debounce onward.

Interfaces | 327

www.dell.com | support.dell.com

Assign a debounce time to an interface

Command Syntax
link debounce time [milliseconds]

Command Mode INTERFACE

Purpose Enter the time to delay link status change notification on this interface. Range: 100-5000 ms Default for Copper is 3100 ms Default for Fiber is 100 ms

Figure 15-25.

Setting Debounce Time

FTOS(conf)#int gi 3/1 FTOS(conf-if-gi-3/1)#link debounce time 150 FTOS(conf-if-gi-3/1)#=

Show debounce times in an interface

show interface debounce [type] [slot/

EXEC Privilege

port]

Show the debounce time for the specified interface. Enter the interface type keyword followed by the type of interface and slot/port information: For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information.

Figure 15-26.

Showing .Debounce Time

FTOS# FTOS#show interfaces debounce gigabitethernet 3/1 Interface Time(ms) GigabitEthernet 3/1 200 FTOS#

Note: FTOS rounds the entered debounce time up to the nearest hundredth. Note in Figure 15-25 that the timer was set at 150 ms, but appears as 200 in Figure 15-26.

Disable ports when one only SFM is available (E300 only)

Selected ports can be shut down when a single SFM is available on the E300 system. Each port to be shut down must be configured individually. When an E300 system boots up and a single SFM is active this configuration, any ports configured with this feature will be shut down. All other ports are booted up.

328

Interfaces

Similarly, if an SFM fails (or is removed) in an E300 system with two SFM, ports configured with this feature will be shut down. All other ports are treated normally. When a second SFM is installed or replaced, all ports are booted up and treated as normally. This feature does not take affect until a single SFM is active in the E300 system.

Disable port on one SFM

This feature must be configured for each interface to shut down in the event that an SFM is disabled. Enter the command disable-on-sfm-failure from INTERFACE mode to disable the port when only a single SFM is available.

Link Dampening
Interface state changes occur when interfaces are administratively brought up or down or if an interface state changes. Every time an interface changes state or flaps, routing protocols are notified of the status of the routes that are affected by the change in state, and these protocols go through momentous task of re-converging. Flapping therefore puts the status of entire network at risk of transient loops and black holes. Link dampening minimizes the risk created by flapping by imposing a penalty for each interface flap and decaying the penalty exponentially. Once the penalty exceeds certain threshold, the interface is put in an "error-disabled" state, and for all practical purposes of routing, the interface is deemed to be "down." Once the interface becomes stable and the penalty decays below a certain threshold, the interface comes up again and the routing protocols re-converge. Link dampening: reduces processing on the CPUs by reducing excessive interface flapping. improves network stability by penalizing misbehaving interfaces and redirecting traffic improves convergence times and stability throughout the network by isolating failures so that disturbances are not propagated.

Important Points to Remember

Link dampening is not supported on VLAN interfaces Link dampening is disabled when the interface is configured for port monitoring Link dampening can be applied to Layer 2 and Layer 3 interfaces. Link dampening can be configured on individual interfaces in a LAG.

Enable Link Dampening

Enable link dampening using the command dampening from INTERFACE mode, as shown in Figure 15-27.

Interfaces | 329

www.dell.com | support.dell.com

Figure 15-27.

Configuring Link Dampening

R1(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 ip address 10.10.19.1/24 dampening 1 2 3 4 no shutdown R1(conf-if-gi-1/1)#exit

View the link dampening configuration on an interface using the command show config, or view dampening information on all or specific dampened interfaces using the command show interfaces dampening from EXEC Privilege mode, as shown in Figure 15-28.
Figure 15-28. Viewing all Dampened Interfaces
dampening Penalty Half-LifeReuse SuppressMax-Sup 0 0 5 750 2500 2 1200 20 500 1500 4 850 30 600 2000

FTOS# show interfaces InterfaceState Flaps Gi 0/0 Up Gi 0/1 Up Gi 0/2 Down

20 300 120

View a dampening summary for the entire system using the command show interfaces dampening summary from EXEC Privilege mode, as shown in Figure 15-29.
Figure 15-29. Viewing a System-wide Dampening Summary

FTOS# show interfaces dampening summary 20 interfaces are configured with dampening. 3 interfaces are currently suppressed. Following interfaces are currently suppressed: Gi 0/2 Gi 3/1 Gi 4/2 FTOS#

Clear Dampening Counters

Clear dampening counters and accumulated penalties using the command clear dampening, as shown in Figure 15-30.
Figure 15-30. Clearing Dampening Counters

FTOS# clear dampening interface Gi 0/1 FTOS# show interfaces dampening GigabitEthernet0/0 InterfaceState Flaps Penalty Half-LifeReuse SuppressMax-Sup Gi 0/1 Up 0 0 20 500 1500

300

330

Interfaces

Link Dampening Support for XML

View the output of the following show commands in XML by adding | display xml to the end of the command: show interfaces dampening show interfaces dampening summary show interfaces interface x/y

Configure MTU size on an Interface

The E-Series supports a link Maximum Transmission Unit (MTU) of 9252 bytes and maximum IP MTU of 9234 bytes. The link MTU is the frame size of a packet, and the IP MTU size is used for IP fragmentation. If the system determines that the IP packet must be fragmented as it leaves the interface, FTOS divides the packet into fragments no bigger than the size set in the ip mtu command. In FTOS, MTU is defined as the entire Ethernet packet (Ethernet header + FCS + payload) Since different networking vendors define MTU differently, check their documentation when planing MTU sizes across a network. Table 15-7 lists the range for each transmission media.
Table 15-7. MTU Range MTU Range (in bytes) 594-9252 = link MTU 576-9234 = IP MTU

Transmission Media Ethernet

Ethernet Pause Frames

ces Threshold Settings are supported only on platforms: c s

Ethernet Pause Frames is supported on platforms Ethernet Pause Frames allow for a temporary stop in data transmission. A situation may arise where a sending device may transmit data faster than a destination device can accept it. The destination sends a PAUSE frame back to the source, stopping the senders transmission for a period of time. The globally assigned 48-bit Multicast address 01-80-C2-00-00-01 is used to send and receive pause frames. To allow full duplex flow control, stations implementing the pause operation instruct the MAC to enable reception of frames with destination address equal to this multicast address.

Interfaces | 331

www.dell.com | support.dell.com

The PAUSE frame is defined by IEEE 802.3x and uses MAC Control frames to carry the PAUSE commands. Ethernet Pause Frames are supported on full duplex only. The only configuration applicable to half duplex ports is rx off tx off.
Note that if a port is over-subscribed, Ethernet Pause Frame flow control does not ensure no loss behavior. The following error message appears when trying to enable flow control when half duplex is already configured: Cant configure flowcontrol when half duplex is configure, config ignored. The following error message appears when trying to enable half duplex and flow control configuration is on: Cant configure half duplex when flowcontrol is on, config ignored.

Threshold Settings
Threshold Settings are supported only on platforms:

cs

When the transmission pause is set (tx on), 3 thresholds can be set to define the controls more closely. Ethernet Pause Frames flow control can be triggered when either the flow control buffer threshold or flow control packet pointer threshold is reached. The thresholds are: Number of flow-control packet pointers: 1-2047 (default = 75) Flow-control buffer threshold in KB: 1-2013 (default = 49KB) Flow-control discard threshold in KB: 1-2013 (default= 75KB)

The pause is started when either the packet pointer or the buffer threshold is met (whichever is met first). When the discard threshold is met, packets are dropped. The pause ends when both the packet pointer and the buffer threshold fall below 50% of the threshold settings. The discard threshold defines when the interface starts dropping the packet on the interface. This may be necessary when a connected device doesnt honor the flow control frame sent by S-Series. The discard threshold should be larger than the buffer threshold so that the buffer holds at least hold at least 3 packets.

Enable Pause Frames

Note: On the C-Series and S-Series (non-S60) platforms, Ethernet Pause Frames TX should be enabled only after consulting with the Dell Networking Technical Assistance Center.

Note: The S60 supports only the rx control option. The S60 does not transmit pause frames.

Ethernet Pause Frames flow control must be enabled on all ports on a chassis or a line card. If not, the system may exhibit unpredictable behavior.

332

Interfaces

On the C-Series and S-Series systems, the flow-control sender and receiver must be on the same port-pipe. Flow control is not supported across different port-pipes on the C-Series or S-Series system.
Command Syntax
flowcontrol rx [off | on] tx [off | on] [threshold

Command Mode INTERFACE

Purpose Control how the system responds to and generates 802.3x pause frames on 1 and 10Gig line cards. Defaults: C-Series: rx off tx off E-Series: rx on tx on S-Series: rx off tx off S60: rx off

{<1-2047> <1-2013> <1-2013>}]

Parameters: rx on: Enter the keywords rx on to process the received flow control frames on this port. rx off: Enter the keywords rx off to ignore the received flow control frames on this port. tx on: Enter the keywords tx on to send control frames from this port to the connected device when a higher rate of traffic is received. tx off: Enter the keywords tx off so that flow control frames are not sent from this port to the connected device when a higher rate of traffic is received. threshold (C-Series and S-Series only): When tx on is configured, the user can set the threshold values for: Number of flow-control packet pointers: 1-2047 (default = 75) Flow-control buffer threshold in KB: 1-2013 (default = 49KB) Flow-control discard threshold in KB: 1-2013 (default= 75KB) Pause control is triggered when either the flow control buffer threshold or flow control packet pointer threshold is reached.

Interfaces | 333

www.dell.com | support.dell.com

Configure MTU Size on an Interface

If a packet includes a Layer 2 header, the difference in bytes between the link MTU and IP MTU must be enough to include the Layer 2 header. For example, for VLAN packets, if the IP MTU is 1400, the Link MTU must be no less than 1422:
1400-byte IP MTU + 22-byte VLAN Tag = 1422-byte link MTU

On the E-Series, the user must enter the ip mtu command to manually configure the IP MTU to compensate for the Layer 2 header. The C-Series and S-Series automatically configure the IP MTU. Table 15-8 lists the various Layer 2 overheads found in FTOS and the number of bytes.
Table 15-8. Difference between Link MTU and IP MTU Difference between Link MTU and IP MTU 18 bytes 22 bytes 22 bytes 26 bytes

Layer 2 Overhead Ethernet (untagged) VLAN Tag Untagged Packet with VLAN-Stack Header Tagged Packet with VLAN-Stack Header

Link MTU and IP MTU considerations for port channels and VLANs are as follows. Port Channels: All members must have the same link MTU value and the same IP MTU value. The port channel link MTU and IP MTU must be less than or equal to the link MTU and IP MTU values configured on the channel members.

Example: If the members have a link MTU of 2100 and an IP MTU 2000, the port channels MTU values cannot be higher than 2100 for link MTU or 2000 bytes for IP MTU. VLANs: All members of a VLAN must have the same IP MTU value. Members can have different Link MTU values. Tagged members must have a link MTU 4 bytes higher than untagged members to account for the packet tag. The VLAN link MTU and IP MTU must be less than or equal to the link MTU and IP MTU values configured on the VLAN members.

Example: The VLAN contains tagged members with Link MTU of 1522 and IP MTU of 1500 and untagged members with Link MTU of 1518 and IP MTU of 1500. The VLANs Link MTU cannot be higher than 1518 bytes and its IP MTU cannot be higher than 1500 bytes.

334

Interfaces

Port-pipes
A port pipe is a Dell Networking specific term for the hardware path that packets follow through a system. Port pipes travel through a collection of circuits (ASICs) built into line cards and RPMs on which various processing events for the packets occur. One or two port pipes process traffic for a given set of physical interfaces or a port-set. The E300 only supports one port pipe per slot. On the E1200 and E600 each slot has two port pipes with following specifications: 48 port line rate cards have two port pipes on the line card 48 port high density cards have only one port pipe on the line card

Note: All references to the E1200 in this section include the E1200i-AC and E1200i-DC. References to E600 include the E600i.

For the purposes of diagnostics, the major difference between the E-Series platforms is the number of port pipes per slot. E1200 and E600Each slot has two port-pipes. Each portpipe has nine 3.125Gbps channels to the backplane, one to each SFM. E300Each slot has one portpipe. Each port-pipe has eight 3.125Gbps channels to the backplane, with four channels to each SFM.

Table 15-9 presents these platform differences again.

Table 15-9.

Platform Differences Concerning Port-pipes Port-pipes / Slot 2 2 1 Channels / Capacity of Each Port-pipe Channel (Gbps) 9 9 8 3.125 3.125 3.125 Raw Slot Capacity (Gbps) 56.25 56.25 25

Chassis Type E1200/E1200i-AC/DC E600/E600i E300

Interfaces | 335

www.dell.com | support.dell.com

Auto-Negotiation on Ethernet Interfaces

Setting speed and duplex mode of Ethernet Interfaces
By default, auto-negotiation of speed and duplex mode is enabled on 10/100/1000 Base-T Ethernet interfaces. Only 10GE interfaces do not support auto-negotiation. When using 10GE interfaces, verify that the settings on the connecting devices are set to no auto-negotiation.
Note: Starting with FTOS 7.8.1.0, when a copper SFP2 module with catalog number GP-SFP2-1T is used in the S25P model of the S-Series, its speed can be manually set with the speed command. When the speed is set to 10 or 100 Mbps, the duplex command can also be executed.

The local interface and the directly connected remote interface must have the same setting, and auto-negotiation is the easiest way to accomplish that, as long as the remote interface is capable of auto-negotiation. Note: As a best practice, Dell Networking recommends keeping auto-negotiation enabled. Auto-negotiation should only be disabled on switch ports that attach to devices not capable of supporting negotiation or where connectivity issues arise from interoperability issues. For 10/100/1000 Ethernet interfaces, the negotiation auto command is tied to the speed command. Auto-negotiation is always enabled when the speed command is set to 1000 or auto. To discover whether the remote and local interface require manual speed synchronization, and to manually synchronize them if necessary, use the following command sequence (see Figure 15-32 on page 337):
Step 1 2 3 4 5 6 7 Task Determine the local interface status. See Figure 15-31. Determine the remote interface status. Access CONFIGURATION mode. Access the port. Set the local port speed. Optionally, set full- or half-duplex. Disable auto-negotiation on the port. If the speed was set to 1000, auto-negotiation does not need to be disabled. Verify configuration changes. Command Syntax
show interfaces [interface | linecard slot-number] status

Command Mode EXEC Privilege EXEC EXEC Privilege EXEC Privilege CONFIGURATION INTERFACE INTERFACE INTERFACE

[Use the command on the remote system that is equivalent to the above command.]
config interface interface slot/port speed {10 | 100 | 1000 | auto} duplex {half | full} no negotiation auto

show config

INTERFACE

336

Interfaces

Note: The show interfaces status command displays link status, but not administrative status. For link and administrative status, use show ip interface [interface | brief | linecard slot-number] [configuration]. Figure 15-31. show interfaces status Command Example

FTOS#show interfaces status Port Description Status Speed Duplex Vlan Gi 0/0 Up 1000 Mbit Auto -Gi 0/1 Down Auto Auto 1 Gi 0/2 Down Auto Auto 1 Gi 0/3 Down Auto Auto -Gi 0/4 FTOSPort Up 1000 Mbit Auto 30-130 Gi 0/5 Down Auto Auto -Gi 0/6 Down Auto Auto -Gi 0/7 Up 1000 Mbit Auto 1502,1504,1506-1508,1602 Gi 0/8 Down Auto Auto -Gi 0/9 Down Auto Auto -Gi 0/10 Down Auto Auto -Gi 0/11 Down Auto Auto -Gi 0/12 Down Auto Auto -[output omitted]

In the example, above, several ports display Auto in the Speed field, including port 0/1. In Figure 15-32, the speed of port 0/1 is set to 100Mb and then its auto-negotiation is disabled.
Figure 15-32. Setting Port Speed Example

FTOS#configure FTOS(config)#interface gig 0/1 FTOS(Interface 0/1)#speed 100 FTOS(Interface 0/1)#duplex full FTOS(Interface 0/1)#no negotiation auto FTOS(Interface 0/1)#show config ! interface GigabitEthernet 0/1 no ip address speed 100 duplex full no shutdown

Setting Auto-Negotiation Options

The negotiation auto command provides a mode option for configuring an individual port to forced master/forced slave once auto-negotiation is enabled. Caution: Ensure that only one end of the node is configured as forced-master and the other is configured as forced-slave. If both are configured the same (that is both as forced-master or both as forced-slave), the show interface command will flap between an auto-neg-error and forced-master/slave states.

Interfaces | 337

www.dell.com | support.dell.com

Figure 15-33.

Setting Auto-Negotiation Options

FTOS(conf)# int gi 0/0 FTOS(conf-if)#neg auto FTOS(conf-if-autoneg)# ? end Exit from configuration mode exit Exit from autoneg configuration mode mode Specify autoneg mode no Negate a command or set its defaults show Show autoneg configuration information FTOS(conf-if-autoneg)#mode ? forced-master Force port to master mode

For details on the speed, duplex, and negotiation auto commands, see the Interfaces chapter of the FTOS Command Reference.

Adjust the keepalive timer

Use the keepalive command to change the time interval between keepalive messages on the interfaces. The interface sends keepalive messages to itself to test network connectivity on the interface. To change the default time interval between keepalive messages, use the following command:

Command Syntax
keepalive [seconds]

Command Mode INTERFACE

Purpose Change the default interval between keepalive messages.

To view the new setting, use the show config command in the INTERFACE mode.

View Advanced Interface Information

Display Only Configured Interfaces
The following options have been implemented for show [ip | running-config] interfaces commands for (only) linecard interfaces. When the configured keyword is used, only interfaces that have non-default configurations are displayed. Dummy linecard interfaces (created with the linecard command) are treated like any other physical interface. Figure 15-34 lists the possible show commands that have the configured keyword available:

338

Interfaces

Figure 15-34.
FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show FTOS#show

show Commands with configured Keyword Examples

interfaces configured interfaces linecard 0 configured interfaces gigabitEthernet 0 configured ip interface configured ip interface linecard 1 configured ip interface gigabitEthernet 1 configured ip interface br configured ip interface br linecard 1 configured ip interface br gigabitEthernet 1 configured running-config interfaces configured running-config interface gigabitEthernet 1 configured

In EXEC mode, the show interfaces switchport command displays only interfaces in Layer 2 mode and their relevant configuration information. The show interfaces switchport command (Figure 15-35) displays the interface, whether the interface supports IEEE 802.1Q tagging or not, and the VLANs to which the interface belongs.
Figure 15-35. show interfaces switchport Command Example

FTOS#show interfaces switchport Name: GigabitEthernet 13/0 802.1QTagged: True Vlan membership: Vlan 2

Name: GigabitEthernet 13/1 802.1QTagged: True Vlan membership: Vlan 2

Name: GigabitEthernet 13/2 802.1QTagged: True Vlan membership: Vlan 2

Name: GigabitEthernet 13/3 802.1QTagged: True

Configure Interface Sampling Size

Use the rate-interval command, in INTERFACE mode, to configure the number of seconds of traffic statistics to display in the show interfaces output. Although any value between 30 and 299 seconds (the default) can be entered, software polling is done once every 15 seconds. So, for example, if you enter 19, you will actually get a sample of the past 15 seconds. All LAG members inherit the rate interval configuration from the LAG. Figure 15-36 shows how to configure rate interval when changing the default value:

Interfaces | 339

www.dell.com | support.dell.com

Figure 15-36.

Configuring Rate Interval Example

FTOS#show interfaces TenGigabitEthernet 10/0 is down, line protocol is down Hardware is Force10Eth, address is 00:01:e8:01:9e:d9 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 10000 Mbit ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 1d23h44m Queueing strategy: fifo 0 packets input, 0 bytes Input 0 IP Packets, 0 Vlans 0 MPLS 0 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts Received 0 input symbol errors, 0 runts, 0 giants, 0 throttles 0 CRC, 0 IP Checksum, 0 overrun, 0 discarded 0 packets output, 0 bytes, 0 underruns Output 0 Multicasts, 0 Broadcasts, 0 Unicasts 0 IP Packets, 0 Vlans, 0 MPLS 0 throttles, 0 discarded Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Time since last interface status change: 1d23h40m FTOS(conf)#interface tengigabitethernet 10/0 FTOS(conf-if-te-10/0)#rate-interval 100 FTOS#show interfaces TenGigabitEthernet 10/0 is down, line protocol is down Hardware is Force10Eth, address is 00:01:e8:01:9e:d9 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 10000 Mbit ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 1d23h45m Queueing strategy: fifo 0 packets input, 0 bytes Input 0 IP Packets, 0 Vlans 0 MPLS 0 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts Received 0 input symbol errors, 0 runts, 0 giants, 0 throttles 0 CRC, 0 IP Checksum, 0 overrun, 0 discarded 0 packets output, 0 bytes, 0 underruns Output 0 Multicasts, 0 Broadcasts, 0 Unicasts 0 IP Packets, 0 Vlans, 0 MPLS 0 throttles, 0 discarded Rate info (interval 100 seconds): Input 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Time since last interface status change: 1d23h42m

Default value of 299 seconds

Change rate interval to 100

New rate interval set to 100

Dynamic Counters
By default, counting for the following four applications is enabled: IPFLOW IPACL

340

Interfaces

L2ACL L2FIB

For remaining applications, FTOS automatically turns on counting when the application is enabled, and is turned off when the application is disabled. Please note that if more than four counter-dependent applications are enabled on a port pipe, there is an impact on line rate performance. The following counter-dependent applications are supported by FTOS: Egress VLAN Ingress VLAN Next Hop 2 Next Hop 1 Egress ACLs ILM IP FLOW IP ACL IP FIB L2 ACL L2 FIB

Interfaces | 341

www.dell.com | support.dell.com

Clear interface counters

The counters in the show interfaces command are reset by the clear counters command. This command does not clear the counters captured by any SNMP program. To clear the counters, use the following command in the EXEC Privilege mode:
Command Syntax
clear counters [interface] [vrrp [vrid] | learning-limit]

Command Mode EXEC Privilege

Purpose Clear the counters used in the show interface commands for all VRRP groups, VLANs, and physical interfaces or selected ones. Without an interface specified, the command clears all interface counters. (OPTIONAL) Enter the following interface keywords and slot/port or number information: For a 1-Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a Loopback interface, enter the keyword loopback followed by a number from 0 to 16383. For a Port Channel interface, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale. For the management interface on the RPM, enter the keyword ManagementEthernet followed by slot/port information. The slot range is 0-1, and the port range is 0. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a VLAN, enter the keyword vlan followed by a number from 1 to 4094 E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.
(OPTIONAL) Enter the keyword vrrp to clear statistics for all VRRP groups configured. Enter a number from 1 to 255 as the vrid. (OPTIONAL) Enter the keyword learning-limit to clear unknown source address (SA) drop counters when MAC learning limit is configured on the interface.

When you enter this command, you must confirm that you want FTOS to clear the interface counters for that interface (Figure 15-37).
Figure 15-37. Clearing an Interface

FTOS#clear counters gi 0/0 Clear counters on GigabitEthernet 0/0 [confirm] FTOS#

342

Interfaces

16
IPv4 Routing
IPv4 Routing is supported on platforms

ces

FTOS supports various IP addressing features. This chapter explains the basics of Domain Name Service (DNS), Address Resolution Protocol (ARP), and routing principles and their implementation in FTOS. IP Addresses Directed Broadcast Resolution of Host Names ARP UDP Helper

Table 16-1 lists the defaults for the IP addressing features described in this chapter.
Table 16-1. IP Feature DNS Directed Broadcast Proxy ARP ICMP Unreachable ICMP Redirect IP Defaults Default Disabled Disabled Enabled Disabled Disabled

IP Addresses
FTOS supports IP version 4, as described in RFC 791. It also supports classful routing and Variable Length Subnet Masks (VLSM). With VLSM one network can be can configured with different masks. Supernetting, which increases the number of subnets, is also supported. Subnetting is when a mask is added to the IP address to separate the network and host portions of the IP address. At its most basic level, an IP address is 32-bits composed of network and host portions and represented in dotted decimal format. For example,
00001010110101100101011110000011

IPv4 Routing | 343

www.dell.com | support.dell.com

is represented as 10.214.87.131 For more information on IP addressing, refer to RFC 791, Internet Protocol.

Implementation Information
In FTOS, you can configure any IP address as a static route except IP addresses already assigned to interfaces.
Note: FTOS versions 7.7.1.0 and later support 31-bit subnet masks (/31, or 255.255.255.254) as defined by RFC 3021. This feature allows you to save two more IP addresses on point-to-point links than 30-bit masks. FTOS supports RFC 3021 with ARP.

Configuration Task List for IP Addresses

The following list includes the configuration tasks for IP addresses: Assign IP addresses to an interface (mandatory) Configure static routes (optional) Configure static routes for the management interface (optional)

For a complete listing of all commands related to IP addressing, refer to FTOS Command Line Interface Reference.

Assign IP addresses to an interface

Assign primary and secondary IP addresses to physical or logical (for example, VLAN or port channel) interfaces to enable IP communication between the E-Series and hosts connected to that interface. In FTOS, you can assign one primary address and up to 255 secondary IP addresses to each interface.

344

IPv4 Routing

To assign an IP address to an interface, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
interface interface

Command Mode CONFIGURATION

Purpose Enter the keyword interface followed by the type of interface and slot/port information: For a 1-Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a Loopback interface, enter the keyword loopback followed by a number from 0 to 16383. For the Management interface on the RPM, enter the keyword ManagementEthernet followed by the slot/ port information. The slot range is 0-1 and the port range is 0. For a port channel interface, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a VLAN interface, enter the keyword vlan followed by a number from 1 to 4094. E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS. Enable the interface. Configure a primary IP address and mask on the interface. ip-address mask: IP address must be in dotted decimal format (A.B.C.D) and the mask must be in slash prefix-length format (/24). Add the keyword secondary if the IP address is the interfaces backup IP address. You can configure up to eight secondary IP addresses.

2 3

no shutdown ip address ip-address mask [secondary]

INTERFACE INTERFACE

To view the configuration, use the show config command (Figure 246) in the INTERFACE mode or show ip interface in the EXEC privilege mode (Figure 247).
Figure 16-1. show config Command Example in the INTERFACE Mode
FTOS(conf-if)#show conf ! interface GigabitEthernet 0/0 ip address 10.11.1.1/24 no shutdown ! FTOS(conf-if)#

Figure 16-2. show ip interface Command Example

IPv4 Routing | 345

www.dell.com | support.dell.com

FTOS#show ip int gi 0/8 GigabitEthernet 0/8 is up, line protocol is up Internet address is 10.69.8.1/24 Broadcast address is 10.69.8.255 Address determined by config file MTU is 1554 bytes Inbound access list is not set Proxy ARP is enabled Split Horizon is enabled Poison Reverse is disabled ICMP redirects are not sent ICMP unreachables are not sent

FTOS#

Configure static routes

A static route is an IP address that is manually configured and not learned by a routing protocol, such as OSPF. Often static routes are used as backup routes in case other dynamically learned routes are unreachable. To configure a static route, use the following command in the CONFIGURATION mode:

Command Syntax
ip route ip-address mask {ip-address | interface [ip-address]} [distance] [permanent] [tag tag-value]

Command Mode CONFIGURATION

Purpose Configure a static IP address. Use the following required and optional parameters: ip-address: Enter an address in dotted decimal format (A.B.C.D). mask: Enter a mask in slash prefix-length format (/X). interface: Enter an interface type followed by slot/port information. distance range: 1 to 255 (optional). permanent: Keep the static route in the routing table (if interface option is used) even if the interface with the route is disabled. (optional) tag tag-value range: 1 to 4294967295. (optional)

You can enter as many static IP addresses as necessary. To view the configured routes, use the show ip route static command.

346

IPv4 Routing

Figure 16-3. show ip route static Command Example (partial)

FTOS#show ip route static Destination Gateway ----------------S 2.1.2.0/24 Direct, Nu 0 S 6.1.2.0/24 via 6.1.20.2, S 6.1.2.2/32 via 6.1.20.2, S 6.1.2.3/32 via 6.1.20.2, S 6.1.2.4/32 via 6.1.20.2, S 6.1.2.5/32 via 6.1.20.2, S 6.1.2.6/32 via 6.1.20.2, S 6.1.2.7/32 via 6.1.20.2, S 6.1.2.8/32 via 6.1.20.2, S 6.1.2.9/32 via 6.1.20.2, S 6.1.2.10/32 via 6.1.20.2, S 6.1.2.11/32 via 6.1.20.2, S 6.1.2.12/32 via 6.1.20.2, S 6.1.2.13/32 via 6.1.20.2, S 6.1.2.14/32 via 6.1.20.2, S 6.1.2.15/32 via 6.1.20.2, S 6.1.2.16/32 via 6.1.20.2, S 6.1.2.17/32 via 6.1.20.2, S 11.1.1.0/24 Direct, Nu 0 Direct, Lo 0 --More--

Te Te Te Te Te Te Te Te Te Te Te Te Te Te Te Te Te

5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0 5/0

Dist/Metric Last Change ----------- ----------0/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 1/0 00:02:30 0/0 00:02:30

FTOS installs a next hop that is on the directly connected subnet of current IP address on the interface (for example, if interface gig 0/0 is on 172.31.5.0 subnet, FTOS installs the static route). FTOS also installs a next hop that is not on the directly connected subnet but which recursively resolves to a next hop on the interface's configured subnet. For example, if gig 0/0 has ip address on subnet 2.2.2.0 and if 172.31.5.43 recursively resolves to 2.2.2.0, FTOS installs the static route. When interface goes down, FTOS withdraws the route. When interface comes up, FTOS re-installs the route. When recursive resolution is broken, FTOS withdraws the route. When recursive resolution is satisfied, FTOS re-installs the route.

Configure static routes for the management interface

When an IP address used by a protocol and a static management route exists for the same prefix, the protocol route takes precedence over the static management route. To configure a static route for the management port, use the following command in the CONFIGURATION mode:

Command Syntax
management route ip-address mask {forwarding-router-address | ManagementEthernet slot/port}

Command Mode CONFIGURATION

Purpose Assign a static route to point to the management interface or forwarding router.

IPv4 Routing | 347

www.dell.com | support.dell.com

To view the configured static routes for the management port, use the show ip management-route command in the EXEC privilege mode.
Figure 16-4. show ip management-route Command Example
FTOS>show ip management-route Destination ----------1.1.1.0/24 172.16.1.0/24 172.31.1.0/24 FTOS> Gateway ------172.31.1.250 172.31.1.250 ManagementEthernet 1/0 State ----Active Active Connected

Directed Broadcast
By default, FTOS drops directed broadcast packets destined for an interface. This default setting provides some protection against Denial of Service (DOS) attacks. To enable FTOS to receive directed broadcasts, use the following command in the INTERFACE mode:

Command Syntax
ip directed-broadcast

Command Mode INTERFACE

Purpose Enable directed broadcast.

To view the configuration, use the show config command in the INTERFACE mode.

Resolution of Host Names

Domain Name Service (DNS) maps host names to IP addresses. This feature simplifies such commands as Telnet and FTP by allowing you to enter a name instead of an IP address. Dynamic resolution of host names is disabled by default. Unless the feature is enabled, the system resolves only host names entered into the host table with the ip host command. Enable dynamic resolution of host names Specify local system domain and a list of domains DNS with traceroute

Enable dynamic resolution of host names

By default, dynamic resolution of host names (DNS) is disabled. To enable DNS, use the following commands in the CONFIGURATION mode:

348

IPv4 Routing

Command Syntax
ip domain-lookup ip name-server ip-address [ip-address2 ... ip-address6]

Command Mode CONFIGURATION CONFIGURATION

Purpose Enable dynamic resolution of host names. Specify up to 6 name servers. The order you entered the servers determines the order of their use.

To view current bindings, use the show hosts command.

Figure 16-5. show hosts Command Example
FTOS>show host Default domain is force10networks.com Name/address lookup uses domain service Name servers are not set Host Flags TTL --------------ks (perm, OK) patch1 (perm, OK) tomm-3 (perm, OK) gxr (perm, OK) f00-3 (perm, OK) FTOS>

Type ---IP IP IP IP IP

Address ------2.2.2.2 192.68.69.2 192.68.99.2 192.71.18.2 192.71.23.1

To view the current configuration, use the show running-config resolve command.

Specify local system domain and a list of domains

If you enter a partial domain, FTOS can search different domains to finish or fully qualify that partial domain. A fully qualified domain name (FQDN) is any name that is terminated with a period/dot. FTOS searches the host table first to resolve the partial domain. The host table contains both statically configured and dynamically learnt host and IP addresses. If FTOS cannot resolve the domain, it tries the domain name assigned to the local system. If that does not resolve the partial domain, FTOS searches the list of domains configured To configure a domain name, use the following command in the CONFIGURATION mode:

Command Syntax
ip domain-name name

Command Mode CONFIGURATION

Purpose Configure one domain name for the E-Series

To configure a list of domain names, use the following command in the CONFIGURATION mode:

Command Syntax
ip domain-list name

Command Mode CONFIGURATION

Purpose Configure names to complete unqualified host names. Configure this command up to 6 times to specify a list of possible domain names. FTOS searches the domain names in the order they were configured until a match is found or the list is exhausted.

IPv4 Routing | 349

www.dell.com | support.dell.com

DNS with traceroute

To configure your switch to perform DNS with traceroute, follow the steps below in the CONFIGURATION mode.

Command Syntax
ip domain-lookup ip name-server ip-address [ip-address2 ... ip-address6] traceroute [host | ip-address ]

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION

Purpose Enable dynamic resolution of host names. Specify up to 6 name servers. The order you entered the servers determines the order of their use. When you enter the traceroute command without specifying an IP address (Extended Traceroute), you are prompted for a target and source IP address, timeout in seconds (default is 5), a probe count (default is 3), minimum TTL (default is 1), maximum TTL (default is 30), and port number (default is 33434). To keep the default setting for those parameters, press the ENTER key.

Figure 16-6 is an example output of DNS using the traceroute command.

Figure 16-6. Traceroute command example

FTOS#traceroute www.force10networks.com Translating "www.force10networks.com"...domain server (10.11.0.1) [OK] Type Ctrl-C to abort. -----------------------------------------------------------------------------------------Tracing the route to www.force10networks.com (10.11.84.18), 30 hops max, 40 byte packets -----------------------------------------------------------------------------------------TTL 1 2 3 4 Hostname Probe1 Probe2 Probe3 10.11.199.190 001.000 ms 001.000 ms 002.000 ms gwegress-sjc-02.force10networks.com (10.11.30.126) 005.000 ms 001.000 ms 001.000 ms fw-sjc-01.force10networks.com (10.11.127.254) 000.000 ms 000.000 ms 000.000 ms www.force10networks.com (10.11.84.18) 000.000 ms 000.000 ms 000.000 ms

ARP
FTOS uses two forms of address resolution: ARP and Proxy ARP. Address Resolution Protocol (ARP) runs over Ethernet and enables endstations to learn the MAC addresses of neighbors on an IP network. Over time, FTOS creates a forwarding table mapping the MAC addresses to their corresponding IP address. This table is called the ARP Cache and dynamically learned addresses are removed after a defined period of time. For more information on ARP, see RFC 826, An Ethernet Address Resolution Protocol.

350

IPv4 Routing

In FTOS, Proxy ARP enables hosts with knowledge of the network to accept and forward packets from hosts that contain no knowledge of the network. Proxy ARP makes it possible for hosts to be ignorant of the network, including subnetting. For more information on Proxy ARP, refer to RFC 925, Multi-LAN Address Resolution, and RFC 1027, Using
ARP to Implement Transparent Subnet Gateways.

Configuration Task List for ARP

The following list includes configuration tasks for ARP: Configure static ARP entries (optional) Enable Proxy ARP (optional) Clear ARP cache (optional) ARP Learning via Gratuitous ARP ARP Learning via ARP Request Configurable ARP Retries

For a complete listing of all ARP-related commands, refer to .

Configure static ARP entries

ARP dynamically maps the MAC and IP addresses, and while most network host support dynamic mapping, you can configure an ARP entry (called a static ARP) for the ARP cache. To configure a static ARP entry, use the following command in the CONFIGURATION mode:

Command Syntax
arp ip-address mac-address interface

Command Mode CONFIGURATION

Purpose Configure an IP address and MAC address mapping for an interface. ip-address: IP address in dotted decimal format (A.B.C.D). mac-address: MAC address in nnnn.nnnn.nnnn format interface: enter the interface type slot/port information.

These entries do not age and can only be removed manually. To remove a static ARP entry, use the no arp ip-address command syntax. To view the static entries in the ARP cache, use the show arp static command (Figure 253) in the EXEC privilege mode.

IPv4 Routing | 351

www.dell.com | support.dell.com

Figure 16-7. show arp static Command Example

FTOS#show arp Protocol Address Age(min) Hardware Address Interface VLAN CPU -------------------------------------------------------------------------------Internet 10.1.2.4 17 08:00:20:b7:bd:32 Ma 1/0 CP FTOS#

Enable Proxy ARP

By default, Proxy ARP is enabled. To disable Proxy ARP, use no proxy-arp command in the interface mode. To re-enable Proxy ARP, use the following command in the INTERFACE mode:

Command Syntax
ip proxy-arp

Command Mode INTERFACE

Purpose Re-enable Proxy ARP.

To view if Proxy ARP is enabled on the interface, use the show config command in the INTERFACE mode. If it is not listed in the show config command output, it is enabled. Only nondefault information is displayed in the show config command output.

Clear ARP cache

To clear the ARP cache of dynamically learnt ARP information, use the following command in the EXEC privilege mode:

352

IPv4 Routing

Command Syntax
clear arp-cache [interface | ip ip-address] [no-refresh]

Command Mode EXEC privilege

Purpose Clear the ARP caches for all interfaces or for a specific interface by entering the following information: For a 1-Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a port channel interface, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a VLAN interface, enter the keyword vlan followed by a number between 1 and 4094. E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS. ip ip-address (OPTIONAL) Enter the keyword ip followed by the IP address of the ARP entry you wish to clear. no-refresh (OPTIONAL) Enter the keyword no-refresh to delete the ARP entry from CAM. Or use this option with interface or ip ip-address to specify which dynamic ARP entries you want to delete.
Note: Transit traffic may not be forwarded during the period when deleted ARP entries are resolved again and re-installed in CAM. Use this option with extreme caution.

ARP Learning via Gratuitous ARP

Gratuitous ARP can mean an ARP request or reply. In the context of ARP Learning via Gratuitous ARP on FTOS, the gratuitous ARP is a request. A Gratuitous ARP Request is an ARP request that is not needed according to the ARP specification, but one that hosts may send to: detect IP address conflicts inform switches of their presence on a port so that packets can be forwarded update the ARP table of other nodes on the network in case of an address change

In the request, the host uses its own IP address in the Sender Protocol Address and Target Protocol Address fields. In FTOS versions prior to 8.3.1.0, if a gratuitous ARP is received some time after an ARP request is sent, only RP2 installs the ARP information. For example: 1. At time t=0 FTOS sends an ARP request for IP A.B.C.D 2. At time t=1 FTOS receives an ARP request for IP A.B.C.D 3. At time t=2 FTOS installs an ARP entry for A.B.C.D only on RP2.

IPv4 Routing | 353

www.dell.com | support.dell.com

Beginning with version 8.3.1.0, when a Gratuitous ARP is received, FTOS installs an ARP entry on all 3 CPUs.

Task Enable ARP learning via gratuitous ARP.

Command Syntax arp learn-enable

Command Mode CONFIGURATION

ARP Learning via ARP Request

In FTOS versions prior to 8.3.1.0, FTOS learns via ARP Requests only if the Target IP specified in the packet matches the IP address of the receiving router interface. This is the case when a host is attempting to resolve the gateway address. If the Target IP does not match the incoming interface, then the packet is dropped. If there is an existing entry for the requesting host, it is updated.
Figure 16-8. Learning via Gratuitous ARP
VLAN ID: 1.1.1.1

ARP Request Target IP: 1.1.1.3

X
Target IP is not the VLAN interface IP. Update existing Host 1 entry. Drop packet. Host 2 IP: 1.1.1.3 MAC: BB

Host 1 IP: 1.1.1.2 MAC: AA

Beginning with FTOS version 8.3.1.0, when ARP Learning via Gratuitous ARP is enabled, the system installs a new ARP entry, or updates an existing entry for all received ARP requests.
Figure 16-9. Learning via Gratuitous ARP
VLAN ID: 1.1.1.1 ARP Learning via Gratuitous ARP enabled

ARP Request Target IP: 1.1.1.3

X
Target IP is not the VLAN interface IP. Install new entry for Host 1, or update existing Host 1 entry. Drop packet. Host 2 IP: 1.1.1.3 MAC: BB

Host 1 IP: 1.1.1.2 MAC: AA

Whether ARP Learning via Gratuitous ARP is is enabled or disabled, the system does not look up the Target IP. It only updates the ARP entry for the Layer 3 interface with the source IP of the request.

354

IPv4 Routing

Configurable ARP Retries

In FTOS versions prior to 8.3.1.0 the number of ARP retries is set to 5 and is not configurable. After 5 retries, FTOS backs off for 20 seconds before it sends a new request. Beginning with FTOS version 8.3.1.0, the number of ARP retries is configurable. The backoff interval remains at 20 seconds.
Task Set the number of ARP retries. Command Syntax arp retries number Default: 5 Range: 5-20 Display all ARP entries learned via gratuitous ARP. show arp retries EXEC Privilege Command Mode CONFIGURATION

UDP Helper
UDP helper allows you to direct the forwarding IP/UDP broadcast traffic by creating special broadcast addresses and rewriting the destination IP address of packets to match those addresses. Configurations using this feature are described in the section Configurations Using UDP Helper.

Configuring UDP Helper

Configuring FTOS to direct UDP broadcast is a two-step process: 1. Enable UDP helper and specify the UDP ports for which traffic is forwarded. See Enabling UDP Helper. 2. Configure a broadcast address on interfaces that will receive UDP broadcast traffic. See Configuring a Broadcast Address.

Important Points to Remember about UDP Helper

The existing command ip directed broadcast is rendered meaningless if UDP helper is enabled on the same interface. The broadcast traffic rate should not exceed 200 packets per second when UDP helper is enabled. You may specify a maximum of 16 UDP ports. UDP helper is compatible with IP helper (ip helper-address): UDP broadcast traffic with port number 67 or 68 are unicast to the DHCP server per the ip helper-address configuration whether or not the UDP port list contains those ports. If the UDP port list contains ports 67 or 68, UDP broadcast traffic forwarded on those ports.

IPv4 Routing | 355

www.dell.com | support.dell.com

Enabling UDP Helper

Enable UPD helper using the command ip udp-helper udp-ports, as shown in Figure 16-10.
Figure 16-10. Enabling UDP Helper

FTOS(conf-if-gi-1/1)#ip udp-helper udp-port 1000 FTOS(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 ip address 2.1.1.1/24 ip udp-helper udp-port 1000 no shutdown

View the interfaces and ports on which UDP helper is enabled using the command show ip udp-helper from EXEC Privilege mode, as shown in Figure 16-11.
Figure 16-11. Viewing the UDP Broadcast Configuration

FTOS#show ip udp-helper -------------------------------------------------Port UDP port list -------------------------------------------------Gi 1/1 1000

Configuring a Broadcast Address

Configure a broadcast address on an interface using the command ip udp-broadcast-address, as shown in Figure 16-12.
Figure 16-12. Configuring a Broadcast Address

FTOS(conf-if-vl-100)#ip udp-broadcast-address 1.1.255.255 FTOS(conf-if-vl-100)#show config ! interface Vlan 100 ip address 1.1.0.1/24 ip udp-broadcast-address 1.1.255.255 untagged GigabitEthernet 1/2 no shutdown

View the configured broadcast address for an interface using the command show interfaces, as shown in Figure 16-13.

356

IPv4 Routing

Figure 16-13.

Configuring a Broadcast Address

R1_E600(conf)#do show interfaces vlan 100 Vlan 100 is up, line protocol is down Address is 00:01:e8:0d:b9:7a, Current address is 00:01:e8:0d:b9:7a Interface index is 1107787876 Internet address is 1.1.0.1/24 IP UDP-Broadcast address is 1.1.255.255 MTU 1554 bytes, IP MTU 1500 bytes LineSpeed auto ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:07:44 Queueing strategy: fifo Input Statistics: 0 packets, 0 bytes Time since last interface status change: 00:07:44

Configurations Using UDP Helper

When UDP helper is enabled and the destination IP address of an incoming packet is a broadcast address, FTOS suppresses the destination address of the packet. The following sections describe various configurations that employ UDP helper to direct broadcasts. UDP Helper with Broadcast-all Addresses UDP Helper with Subnet Broadcast Addresses UDP Helper with Configured Broadcast Addresses UDP Helper with No Configured Broadcast Addresses

UDP Helper with Broadcast-all Addresses

When the destination IP address of an incoming packet is the IP broadcast address, FTOS rewrites the address to match the configured broadcast address. In Figure 16-14: 1. Packet 1 is dropped at ingress if no UDP helper address is configured. 2. If UDP helper (using the command ip udp-helper udp-port) is enabled, and the UDP destination port of the packet matches the UDP port configured, the system changes the destination address to the configured broadcast 1.1.255.255 and routes the packet to VLANs 100 and 101. If an IP broadcast address is not configured (using the command ip udp-broadcast-address) on VLANs 100 or 101, the packet is forwarded using the original destination destination IP address 255.255.255.255. Packet 2, sent from a host on VLAN 101 has a broadcast MAC address and IP address. In this case: 1. It is flooded on VLAN 101 without changing the destination address because the forwarding process is Layer 2. 2. If UDP helper is enabled, the system changes the destination IP address to the configured broadcast address 1.1.255.255 and forwards the packet to VLAN 100.

IPv4 Routing | 357

www.dell.com | support.dell.com

3. Packet 2 is also forwarded to the ingress interface with an unchanged destination address because it does not have broadcast address configured.
Figure 16-14. UDP helper with All Broadcast Addresses
VLAN 100 IP address: 1.1.0.1/24 Subnet broadcast address: 1.1.0.255 Configured broadcast address: 1.1.255.255 Hosts on VLAN 100: 1.1.0.2, 1.1.0.3, 1.1.0.4

Packet 1 Destination Address: 255.255.255.255 1/1 1/2

Ingress interface IP Address: 2.1.1.1/24 UDP helper enabled

1/3

Packet 2 Switched Packet

VLAN 101 IP address: 1.11.1/24 Subnet broadcast address: 1.1.1.255 Configured broadcast address: 1.1.255.255 Hosts on VLAN 100: 1.1.1.2, 1.1.1.3, 1.1.1.4

UDP Helper with Subnet Broadcast Addresses

When the destination IP address of an incoming packet matches the subnet broadcast address of any interface, the system changes the address to the configured broadcast address and sends it to matching interface. In Figure 16-15, Packet 1 has the destination IP address 1.1.1.255, which matches the subnet broadcast address of VLAN 101. If UDP helper is configured and the packet matches the specified UDP port, then the system changes the address to the configured IP broadcast address and floods the packet on VLAN 101. Packet 2 is sent from host on VLAN 101. It has a broadcast MAC address and a destination IP address of 1.1.1.255. In this case, it is flooded on VLAN 101 in its original condition as the forwarding process is Layer 2.
Figure 16-15.
Preamble Start Frame Delimiter

UDP helper with Subnet Broadcast Addresses

Source MAC Ethernet Type (0x88CC) LLDPDU Padding FCS

Destination MAC (01:80:C2:00:00:0E)

TLV 1 Chassis ID

TLV 2 Port ID

TLV 3 Port Description

TLV 4 System Name

TLV 6 TLV 7 TLV 5 System Description System Capabilities Management Addr

TLV 127 Organizationally Specific

TLV 0 End of LLDPDU

UDP Helper with Configured Broadcast Addresses

Incoming packets with a destination IP address matching the configured broadcast address of any interface are forwarded to the matching interfaces.

358

IPv4 Routing

In Figure 16-16, Packet 1 has a destination IP address that matches the configured broadcast address of VLAN 100 and 101. If UDP helper is enabled and the UDP port number matches, the packet is flooded on both VLANs with an unchanged destination address. Packet 2 is sent from a host on VLAN 101. It has broadcast MAC address and a destination IP address that matches the configured broadcast address on VLAN 101. In this case, Packet 2 is flooded on VLAN 101 with the destination address unchanged because the forwarding process is Layer 2. If UDP helper is enabled, the packet is flooded on VLAN 100 as well.
Figure 16-16. UDP Helper with Configured Broadcast Addresses
VLAN 100 IP address: 1.1.0.1/24 Subnet broadcast address: 1.1.0.255 Configured broadcast address: 1.1.255.255 Hosts on VLAN 100: 1.1.0.2, 1.1.0.3, 1.1.0.4

Packet 1 Destination Address: 1.1.255.255 1/1 1/2

Ingress interface IP Address: 2.1.1.1/24 UDP helper enabled

1/3

Packet 2 Switched Packet Destination Address: 1.1.255.255

VLAN 101 IP address: 1.11.1/24 Subnet broadcast address: 1.1.1.255 Configured broadcast address: 1.1.255.255 Hosts on VLAN 100: 1.1.1.2, 1.1.1.3, 1.1.1.4

fnC0048mp

UDP Helper with No Configured Broadcast Addresses

If the incoming packet has a broadcast destination IP address, then the unaltered packet is routed to all Layer 3 interfaces. If the Incoming packet has a destination IP address that matches the subnet broadcast address of any interface, then the unaltered packet is routed to the matching interfaces.

Troubleshooting UDP Helper

Display debugging information using the command debug ip udp-helper, as shown in Figure 16-17.
Figure 16-17. Debugging UDP Broadcast

FTOS(conf)# debug ip udp-helper 01:20:22: Pkt rcvd on Gi 5/0 with IP DA (0xffffffff) will be sent on Gi 5/1 Gi 5/2 Vlan 3 01:44:54: Pkt rcvd on Gi 7/0 is handed over for DHCP processing.

Use the command debug ip dhcp when using the IP helper and UDP helper on the same interface, as shown in Figure 16-18.

IPv4 Routing | 359

www.dell.com | support.dell.com

Figure 16-18.

Debugging IP Helper with UDP Helper

Packet 0.0.0.0:68 -> 255.255.255.255:67 TTL 128 2005-11-05 11:59:35 %RELAY-I-PACKET, BOOTP REQUEST (Unicast) received at interface 172.21.50.193 BOOTP Request, XID = 0x9265f901, secs = 0 hwaddr = 00:02:2D:8D:46:DC, giaddr = 0.0.0.0, hops = 2 2005-11-05 11:59:35 %RELAY-I-BOOTREQUEST, Forwarded BOOTREQUEST for 00:02:2D:8D:46:DC to 137.138.17.6 2005-11-05 11:59:36 %RELAY-I-PACKET, BOOTP REPLY (Unicast) received at interface 194.12.129.98 BOOTP Reply, XID = 0x9265f901, secs = 0 hwaddr = 00:02:2D:8D:46:DC, giaddr = 172.21.50.193, hops = 2 2005-07-05 11:59:36 %RELAY-I-BOOTREPLY, Forwarded BOOTREPLY for 00:02:2D:8D:46:DC to 128.141.128.90 Packet 0.0.0.0:68 -> 255.255.255.255:67 TTL 128

360

IPv4 Routing

17
IPv6 Basics
IPv6 Basics, applies to platforms

ces

Note: The IPv6 basic commands are supported on all platforms. However, not all features are supported on all platforms, nor for all releases. See Table 17-2to determine the FTOS version supporting which features and platforms.

IPv6 (Internet Protocol Version 6) is the successor to IPv4. Due to the extremely rapid growth in internet users, and IP addresses, IPv4 is reaching its maximum usage. IPv6 will eventually replace IPv4 usage to allow for the constant expansion. This chapter provides a brief discussion of the differences between IPv4 and IPv6 and Dell Networking support of IPv6. This chapter discusses the following topics, but is not intended to be a comprehensive discussion of IPv6. Protocol Overview Extended Address Space Stateless Autoconfiguration IPv6 Headers Implementing IPv6 with FTOS Table 17-2 FTOS and IPv6 Feature Support Path MTU Discovery IPv6 Neighbor Discovery QoS for IPv6 IPv6 Multicast SSH over an IPv6 Transport Configuration Task List for IPv6

IPv6 Basics | 361

www.dell.com | support.dell.com

Protocol Overview
IPv6 is an evolution of IPv4. IPv6 is generally installed as an upgrade in devices and operating systems. Most new devices and operating systems support both IPv4 and IPv6. Some key changes in IPv6 are: Extended Address Space Stateless Autoconfiguration Header Format Simplification Improved Support for Options and Extensions

Extended Address Space

The address format is extended from 32 bits to 128 bits. This not only provides room for all anticipated needs, it allows for the use of a hierarchical address space structure to optimize global addressing.

Stateless Autoconfiguration
When a booting device comes up in IPv6 and asks for its network prefix, the device can get the prefix (or prefixes) from an IPv6 router on its link. It can then autoconfigure one or more global IP addresses by using either the MAC address or a private random number to build its unique IP address. Stateless auto-configuration uses three mechanisms for IPv6 address configuration: Prefix Advertisement - Routers use "Router Advertisement" messages to announce the Network Prefix. Hosts then use their interface-identifier MAC address to generate their own valid IPv6 address. Duplicate Address Detection (DAD) - Before configuring its IPv6 address, an IPv6 host node device checks whether that address is used anywhere on the network using this mechanism. Prefix Renumbering - Useful in transparent renumbering of hosts in the network when an organization changes its service provider.
Note: As an alternative to stateless auto-configuration, network hosts can obtain their IPv6 addresses using Dynamic Host Control Protocol (DHCP) servers via stateful auto-configuration.

Note: FTOS provides the flexibility to add prefixes to advertise responses to RS messages. By default the RA response messages are not sent when an RS message is received. Enable the RA response messages with the ipv6 nd prefix default command in INTERFACE mode.

FTOS manipulation of IPv6 stateless auto-configuration supports the router side only. Neighbor Discovery (ND) messages are advertised so the neighbor can use this information to auto-configure its address. However, received Neighbor Discovery (ND) messages are not used to create an IPv6 address.

362

IPv6 Basics

The router redistribution functionality in Neighbor Discovery Protocol (NDP) is similar to IPv4 router redirect messages. Neighbor Discovery Protocol (NDP) uses ICMPv6 redirect messages (Type 137) to inform nodes that a better router exists on the link.

IPv6 Headers
The IPv6 header has a fixed length of 40 bytes. This provides 16 bytes each for Source and Destination information, and 8 bytes for general header information. The IPv6 header includes the following fields: Version (4 bits) Traffic Class (8 bits) Flow Label (20 bits) Payload Length (16 bits) Next Header (8 bits) Hop Limit (8 bits) Source Address (128 bits) Destination Address (128 bits)

IPv6 provides for Extension Headers. Extension Headers are used only if necessary. There can be no extension headers, one extension header or more than one extension header in an IPv6 packet. Extension Headers are defined in the Next Header field of the preceding IPv6 header. IPv6 header fields The 40 bytes of the IPv6 header are ordered as show in Figure 17-1.
Figure 17-1. IPv6 Header Fields

0 Version

12

16

20

24

28 Hop Limit

32

Traffic Class Payload Length

Flow Label Next Header

64 128 192

Source Address

Destination Address
320

Version (4 bits)
The Version field always contains the number 6, referring to the packets IP version.

IPv6 Basics | 363

www.dell.com | support.dell.com

Traffic Class (8 bits)

The Traffic Class field deals with any data that needs special handling. These bits define the packet priority and are defined by the packet Source. Sending and forwarding routers use this field to identify different IPv6 classes and priorities. Routers understand the priority settings and handle them appropriately during conditions of congestion.

Flow Label (20 bits)

The Flow Label field identifies packets requiring special treatment in order to manage real-time data traffic. The sending router can label sequences of IPv6 packets so that forwarding routers can process packets within the same flow without needing to reprocess each packets head separately.
Note: All packets in the flow must have the same source and destination addresses.

Payload Length (16 bits)

The Payload Length field specifies the packet payload. This is the length of the data following the IPv6 header. IPv6 Payload Length only includes the data following the header, not the header itself. The Payload Length limit of 2 bytes requires that the maximum packet payload be 64 KB. However, the Jumbogram option type Extension header supports larger packet sizes when required.

Next Header (8 bits)

The Next Header field identifies the next headers type. If an Extension header is used, this field contains the type of Extension header (Table 17-1). If the next header is a TCP or UDP header, the value in this field is the same as for IPv4. The Extension header is located between the IP header and the TCP or UDP header.
Table 17-1. Value 0 4 6 8 41 43 44 50 51 Next Header field values Description Hop-by Hop option header following IPv4 TCP Exterior Gateway Protocol (EGP) IPv6 Routing header Fragmentation header Encrypted Security Authentication header

364

IPv6 Basics

Table 17-1. Value 59 60

Next Header field values (continued) Description No Next Header Destinations option header

Note: This is not a comprehensive table of Next Header field values. Refer to the Internet Assigned Numbers Authority (IANA) web page http://www.iana.org/assignments/protocol-numbers for a complete and current listing.

Hop Limit (8 bits)

The Hop Limit field shows the number of hops remaining for packet processing. In IPv4, this is known as the Time to Live (TTL) field and uses seconds rather than hops. Each time the packet moves through a forwarding router, this field decrements by 1. If a router receives a packet with a Hop Limit of 1, it decrements it to 0 (zero). The router discards the packet and sends an ICMPv6 message back to the sending router indicating that the Hop Limit was exceeded in transit.

Source Address (128 bits)

The Source Address field contains the IP address for the packet originator.

Destination Address (128 bits)

The Destination Address field contains the intended recipients IP address. This can be either the ultimate destination or the address of the next hop router.

Extension Header fields

Extension headers are used only when necessary. Due to the streamlined nature of the IPv6 header, adding extension headers do not severely impact performance. Each Extension headerss lengths vary, but they are always a multiple of 8 bytes. Each extension header is identified by the Next Header field in the IPv6 header that precedes it. Extension headers are viewed only by the destination router identified in the Destination Address field. If the Destination Address is a multicast address, the Extension headers are examined by all the routers in that multicast group. However, if the Destination Address is a Hop-by-Hop options header, the Extension header is examined bye every forwarding router along the packets route. The Hop-by-Hop options header must immediately follow the IPv6 header, and is noted by the value 0 (zero) in the Next Header field (Table 17-1). Extension headers are processed in the order in which they appear in the packet header.

IPv6 Basics | 365

www.dell.com | support.dell.com

Hop-by-Hop Options header

The Hop-by-Hop options header contains information that is examined by every router along the packets path. It follows the IPv6 header and is designated by the Next Header value 0 (zero) (Table 17-1). When a Hop-by-Hop Options header is not included, the router knows that it does not have to process any router specific information and immediately processes the packet to its final destination. When a Hop-by-Hop Options header is present, the router only needs this extension header and does not need to take the time to view further into the packet. The Hop-by-Hop Options header contains: Next Header (1 byte) This field identifies the type of header following the Hop-by-Hop Options header and uses the same values shown in Table 17-1. Header Extension Length (1 byte) This field identifies the length of the Hop-by-Hop Options header in 8-byte units, but does not include the first 8 bytes. Consequently, if the header is less than 8 bytes, the value is 0 (zero). Options (size varies) This field can contain 1 or more options. The first byte if the field identifies the Option type, and directs the router how to handle the option. Skip and continue processing Discard the packet. Discard the packet and send an ICMP Parameter Problem Code 2 message to the packets Source IP Address identifying the unknown option type Discard the packet and send an ICMP Parameter Problem, Code 2 message to the packets Source IP Address only if the Destination IP Address is not a multicast address. The second byte contains the Option Data Length. The third byte specifies whether the information can change en route to the destination. The value is 1 if it can change; the value is 0 if it cannot change.

00 01 10 11

Addressing
IPv6 addresses are normally written as eight groups of four hexadecimal digits, where each group is separated by a colon (:). For example, 2001:0db8:0000:0000:0000:0000:1428:57ab is a valid IPv6 address. If one or more four-digit group(s) is 0000, the zeros may be omitted and replaced with two colons(::). For example, 2001:0db8:0000:0000:0000:0000:1428:57ab can be shortened to 2001:0db8::1428:57ab. Only one set of double colons is supported in a single address. Any number of consecutive 0000 groups may be reduced to two colons, as long as there is only one double colon used in an address. Leading zeros in a group can also be omitted (as in ::1 for localhost). All the addresses in the following list are all valid and equivalent.

366

IPv6 Basics

2001:0db8:0000:0000:0000:0000:1428:57ab 2001:0db8:0000:0000:0000::1428:57ab 2001:0db8:0:0:0:0:1428:57ab 2001:0db8:0:0::1428:57ab 2001:0db8::1428:57ab 2001:db8::1428:57ab

IPv6 networks are written using Classless Inter-Domain Routing (CIDR) notation. An IPv6 network (or subnet) is a contiguous group of IPv6 addresses the size of which must be a power of two; the initial bits of addresses, which are identical for all hosts in the network, are called the network's prefix. A network is denoted by the first address in the network and the size in bits of the prefix (in decimal), separated with a slash. Since a single host is seen as a network with a 128-bit prefix, host addresses may be written with a following /128. For example, 2001:0db8:1234::/48 stands for the network with addresses 2001:0db8:1234:0000:0000:0000:0000:0000 through 2001:0db8:1234:ffff:ffff:ffff:ffff:ffff

Link-local Addresses
Link-local addresses, starting with fe80:, are assigned only in the local link area. The addresses are generated usually automatically by the operating system's IP layer for each network interface. This provides instant automatic network connectivity for any IPv6 host and means that if several hosts connect to a common hub or switch, they have an instant communication path via their link-local IPv6 address. . Link-local addresses cannot be routed to the public Internet.

Static and Dynamic Addressing

Static IP addresses are manually assigned to a computer by an administrator. Dynamic IP addresses are assigned either randomly or by a server using Dynamic Host Configuration Protocol (DHCP). Even though IP addresses assigned using DHCP may stay the same for long periods of time, they can change. In some cases, a network administrator may implement dynamically assigned static IP addresses. In this case, a DHCP server is used, but it is specifically configured to always assign the same IP address to a particular computer, and never to assign that IP address to another computer. This allows static IP addresses to be configured in one place, without having to specifically configure each computer on the network in a different way. In IPv6, every interface, whether using static or dynamic address assignments, also receives a local-link address automatically in the fe80::/64 subnet.

IPv6 Basics | 367

www.dell.com | support.dell.com

Implementing IPv6 with FTOS

FTOS supports both IPv4 and IPv6, and both may be used simultaneously in your system.
Note: Dell Networking recommends that you use FTOS version 7.6.1.0 or later when implementing IPv6 functionality on an E-Series system.

Table 17-2 lists the FTOS Version in which an IPv6 feature became available for each platform. The sections following the table give some greater detail about the feature. Specific platform support for each feature or functionality is designated by the c e s symbols.
Table 17-2. Feature and/or Functionality FTOS and IPv6 Feature Support

FTOS Release Introduction E-Series E-Series TeraScale ExaScale C-Series S-Series 7.8.1 7.8.1

Documentation and Chapter Location

Basic IPv6 Commands IPv6 Basic Addressing IPv6 address types: Unicast

7.4.1

8.2.1

IPv6 Basic Commands in the FTOS Command Line Interface Reference Guide

7.4.1

8.2.1 8.2.1 8.2.1 8.2.1

7.8.1 7.8.1 7.8.1 7.8.1

7.8.1 7.8.1 7.8.1 7.8.1

Extended Address Space in this chapter IPv6 Neighbor Discovery in this chapter Stateless Autoconfiguration in this chapter Path MTU Discovery in this chapter

IPv6 neighbor discovery 7.4.1 IPv6 stateless autoconfiguration IPv6 MTU path discovery IPv6 Routing Static routing Route redistribution Multiprotocol BGP extensions for IPv6 IPv6 BGP MD5 Authentication IS-IS for IPv6 7.4.1 7.4.1 7.4.1 8.2.1.0 7.5.1 7.4.1 7.4.1

8.2.1 8.2.1 8.2.1 8.2.1.0 8.2.1

7.8.1 7.8.1 7.8.1 8.2.1.0

7.8.1

Assign a Static IPv6 Route in this chapter OSPF, IS-IS, and IPv6 BGP chapters in the FTOS Command Line Reference Guide IPv6 BGP in the FTOS Command Line Reference Guide IPv6 BGP in the FTOS Command Line Reference Guide Intermediate System to Intermediate System (IS-IS) in the FTOS Configuration Guide IPv6 IS-IS in the FTOS Command Line Reference Guide

368

IPv6 Basics

Table 17-2. IS-IS for IPv6 support for redistribution

FTOS and IPv6 Feature Support (continued) 7.6.1 8.2.1 Intermediate System to Intermediate System (IS-IS) in the FTOS Configuration Guide IPv6 IS-IS in the FTOS Command Line Reference Guide

ISIS for IPv6 support for 7.6.1 distribute lists and administrative distance

8.2.1

Intermediate System to Intermediate System (IS-IS) in the FTOS Configuration Guide IPv6 IS-IS in the FTOS Command Line Reference Guide

OSPF for IPv6 (OSPFv3) 7.4.1 Equal Cost Multipath for 7.4.1 IPv6 IPv6 Services and Management Telnet client over IPv6 (outbound Telnet) 7.5.1

8.2.1 8.2.1

7.8.1 7.8.1 7.8.1

OSPFv3 in the FTOS Command Line Reference Guide

8.2.1

7.8.1

7.8.1

Telnet with IPv6 in this chapter Control and Monitoring in the FTOS Command Line Reference Guide

Telnet server over IPv6 (inbound Telnet)

7.4.1

8.2.1

7.8.1

7.8.1

Telnet with IPv6 in this chapter Control and Monitoring in the FTOS Command Line Reference Guide

Secure Shell (SSH) client 7.5.1 support over IPv6 (outbound SSH) Layer 3 only 7.4.1 Secure Shell (SSH) server support over IPv6 (inbound SSH) Layer 3 only IPv6 Access Control Lists IPv6 Multicast PIM-SM for IPv6 7.4.1 7.4.1

8.2.1

7.8.1

7.8.1

SSH over an IPv6 Transport in this chapter

8.2.1

7.8.1

7.8.1

SSH over an IPv6 Transport in this chapter

8.2.1

7.8.1

8.2.1.0

IPv6 Access Control Lists in the FTOS Command Line Reference Guide

8.2.1

IPv6 Multicast in this chapter; IPv6 PIM in the FTOS Command Line Reference Guide

PIM-SSM for IPv6

7.5.1

8.2.1

IPv6 Multicast in this chapter IPv6 PIM in the FTOS Command Line Reference Guide

MLDv1/v2

7.4.1

8.2.1

IPv6 Multicast in this chapter Multicast IPv6 in the FTOS Command Line Reference Guide

IPv6 Basics | 369

www.dell.com | support.dell.com

Table 17-2. MLDv1 Snooping

FTOS and IPv6 Feature Support (continued) 7.4.1 8.2.1 IPv6 Multicast in this chapter Multicast IPv6 in the FTOS Command Line Reference Guide

MLDv2 Snooping

8.3.1.0

8.3.1.0

IPv6 Multicast in this chapter Multicast IPv6 in the FTOS Command Line Reference Guide

IPv6 QoS trust DSCP values 7.4.1 8.2.1 QoS for IPv6 in this chapter

Path MTU Discovery

IPv6 MTU Discovery is supported on platforms

ces

Path MTU (Maximum Transmission Unit) defines the largest packet size that can traverse a transmission path without suffering fragmentation. Path MTU for IPv6 uses ICMPv6 Type-2 messages to discover the largest MTU along the path from source to destination and avoid the need to fragment the packet. The recommended MTU for IPv6 is 1280. Greater MTU settings increase processing efficiency because each packet carries more data while protocol overheads (headers, for example) or underlying per-packet delays remain fixed.

370

IPv6 Basics

Figure 17-2. MTU Discovery Path

Source Router A MTU = 1600 Packet (MTU = 1600) ICMPv6 (Type 2) Use MTU = 1400 MTU = 1400 Router B MTU = 1200 Destination

Packet (MTU = 1400) ICMPv6 (Type 2) Use MTU = 1200 Packet (MTU = 1200) Packet Received

IPv6 Neighbor Discovery

IPv6 NDP is supported on platforms

ces

Neighbor Discovery Protocol (NDP) is a top-level protocol for neighbor discovery on an IPv6 network. In lieu of ARP, NDP uses "Neighbor Solicitation" and "Neighbor Advertisement" ICMPv6 messages for determining relationships between neighboring nodes. Using these messages, an IPv6 device learns the link-layer addresses for neighbors known to reside on attached links, quickly purging cached values that become invalid. With ARP, each node broadcasts ARP requests on the entire link. This approach causes unnecessary processing by uninterested nodes. With NDP, each node sends a request only to the intended destination via a multicast address with the unicast address used as the last 24 bits. Other hosts on the link do not participate in the process, greatly increasing network bandwidth efficiency.

IPv6 Basics | 371

www.dell.com | support.dell.com

Figure 17-3. NDP Router Redistribution

Router C Send a Packet to Network 2001:db8::1428:57ab Router A Local Link Packet Destination (2001:db8::1428:57ab) ICMPv6 Redirect (Data: Use Router C) Packet Destination (Destination 2001:db8::1428:57ab)

Network 2001:db8::1428:57ab

Router B

IPv6 Neighbor Discovery of MTU packets

With FTOS 8.3.1.0, you can set the MTU advertised through the RA packets to incoming routers, without altering the actual MTU setting on the interface. The ip nd mtu command sets the value advertised to routers. It does not set the actual MTU rate. For example, if ip nd mtu is set to 1280, the interface will still pass 1500-byte packets, if that is what is set with the mtu command.

QoS for IPv6

IPv6 QoS is supported on platforms

FTOS IPv6 supports quality of service based on DSCP field. You can configure FTOS to honor the DSCP value on incoming routed traffic and forward the packets with the same value.

IPv6 Multicast
IPv6 Multicast is supported only on platform

FTOS supports the following protocols to implement IPv6 multicast routing:

372

IPv6 Basics

Multicast Listener Discovery Protocol (MLD). MLD on a multicast router sends out periodic general MLD queries that the switch forwards through all ports in the VLAN. There are two versions of MLD: MLD version 1 is based on version 2 of the Internet Group Management Protocol (IGMP) for IPv4, and MLD version 2 is based on version 3 of the IGMP for IPv4. IPv6 multicast for FTOS supports versions 1 and 2 PIM-SM. Protocol-Independent Multicast-Sparse Mode (PIM-SM) is a multicast protocol in which multicast receivers explicitly join to receive multicast traffic. The protocol uses a router as the root or Rendezvous Point (RP) of the share tree distribution tree to distribute multicast traffic to a multicast group. Messages to join the multicast group (Join messages) are sent towards the RP and data is sent from senders to the RP so receivers can discover who are the senders and begin receiving traffic destined to the multicast group. PIM in Source Specific Multicast (PIM-SSM). PIM-SSM protocol is based on the source specific model for forwarding Multicast traffic across multiple domains on the Internet. It is restricted to shortest path trees (SPTs) to specific sources described by hosts using MLD. PIM-SSM is essentially a subset of PIM-SM protocol, which has the capability to join SPTs. The only difference being register states and shared tree states for Multicast groups in SSM range are not maintained. End-hosts use MLD to register their interest in a particular source-group (S,G) pair. PIM-SSM protocol interacts with MLD to construct the multicast forwarding tree rooted at the source S.

Refer to FTOS Command Line Interface Reference document chapters Multicast IPv6, and Protocol Independent Multicast (IPv6) for configuration details.

SSH over an IPv6 Transport

IPv6 SSH is supported on platforms

ces

FTOS supports both inbound and outbound SSH sessions using IPv6 addressing. Inbound SSH supports accessing the system through the management interface as well as through a physical Layer 3 interface. Refer to the Security Commands chapter in the FTOS Command Line Interface Reference document for SSH configuration details.

Configuration Task List for IPv6

This section contains information regarding the following: Change your CAM-Profile on an E-Series system (mandatory) Adjust your CAM-Profile on an C-Series or S-Series Assign an IPv6 Address to an Interface Assign a Static IPv6 Route Telnet with IPv6 SNMP over IPv6 Show IPv6 Information

IPv6 Basics | 373

www.dell.com | support.dell.com

Clear IPv6 Routes

Change your CAM-Profile on an E-Series system

The cam-profile command is supported only on platform

Change your CAM profile to the CAM ipv6-extacl before doing any further IPv6 configuration. Once the CAM profile is changed, save the configuration and reboot your router.
Command Syntax cam-profile ipv6-extacl microcode ipv6-extacl chassis | linecard slot Command Mode EXEC Privileged Purpose Enable the CAM profile with IPv6 extended ACLs on the entire chassis or on a specific linecard chassis changes the CAM profile for all linecards in the chassis linecard slot/port changes the CAM profile only for the specified slot

Figure 17-4 displays the IPv6 CAM profile summary for a chassis that already has IPv6 CAM profile configured. Figure 17-5 shows the full IPv6 CAM profiles. Refer to Chapter 9, Content Addressable Memory, on page 203 for complete information regarding CAM configuration.
Figure 17-4. Command Example: show cam-profile summary (E-Series)
FTOS#show cam-profile summary -- Chassis CAM Profile -: Current Settings : Next Boot Profile Name : IPV6-ExtACL : IPV6-ExtACL MicroCode Name : IPv6-ExtACL : IPv6-ExtACL -- Line card 1 -Profile Name MicroCode Name FTOS# : Current Settings : Next Boot : IPV6-ExtACL : IPV6-ExtACL : IPv6-ExtACL : IPv6-ExtACL

374

IPv6 Basics

Figure 17-5. Command Example: show cam profile (E-Series)

FTOS#show cam-profile -- Chassis CAM Profile -CamSize Profile Name L2FIB L2ACL IPv4FIB IPv4ACL IPv4Flow EgL2ACL EgIPv4ACL Reserved IPv6FIB IPv6ACL IPv6Flow EgIPv6ACL MicroCode Name : : : : : : : : : : : : : : : : 18-Meg Current Settings IPV6-ExtACL 32K entries 1K entries 192K entries 12K entries 8K entries 1K entries 1K entries 2K entries 6K entries 3K entries 4K entries 1K entries IPv6-ExtACL

: : : : : : : : : : : : : : :

Next Boot IPV6-ExtACL 32K entries 1K entries 192K entries 12K entries 8K entries 1K entries 1K entries 2K entries 6K entries 3K entries 4K entries 1K entries IPv6-ExtACL

-- Line card 1 -CamSize : 18-Meg : Current Settings : Next Boot --More--

Adjust your CAM-Profile on an C-Series or S-Series

The cam-acl command is supported on platforms

cs

Although this is not a mandatory step, if you plan to implement IPv6 ACLs, you must adjust your CAM settings. The CAM space is allotted in FP blocks. The total space allocated must equal 13 FP blocks. Note that there are 16 FP blocks, but the System Flow requires 3 blocks that cannot be reallocated. The ipv6acl allocation must be entered as a factor of 2 (2, 4, 6, 8, 10). All other profile allocations can use either even or odd numbered ranges. The default option sets the CAM Profile as follows: L3 ACL (ipv4acl): 6 L2 ACL(l2acl) : 5 IPv6 L3 ACL (ipv6acl): 0 L3 QoS (ipv4qos): 1 L2 QoS (l2qos): 1

IPv6 Basics | 375

www.dell.com | support.dell.com

Save the new CAM settings to the startup-config (write-mem or copy run start) then reload the system for the new settings to take effect.
Command Syntax
cam-acl { ipv6acl }

Command Mode CONFIGURATION

Purpose Allocate space for IPV6 ACLs. Enter the CAM profile name followed by the amount to be allotted. When not selecting the default option, you must enter all of the profiles listed and a range for each. The total space allocated must equal 13. The ipv6acl range must be a factor of 2.

show cam-acl

EXEC EXEC Privilege

Show the current CAM settings.

Assign an IPv6 Address to an Interface

IPv6 Addresses are supported on platforms

ces

Essentially IPv6 is enabled in FTOS simply by assigning IPv6 addresses to individual router interfaces. IPv6 and IPv4 can be used together on a system, but be sure to differentiate that usage carefully. Use the ipv6 address command to assign an IPv6 address to an interface.
Command Syntax ipv6 address ipv6 address/ mask Command Mode CONFIG-INTERFACE Purpose Enter the IPv6 Address for the device. ipv6 address : x:x:x:x::x mask : prefix length 0-128

IPv6 addresses are normally written as eight groups of four hexadecimal digits, where each group is separated by a colon (:). Omitting zeros is accepted as described in Addressing earlier in this chapter.

376

IPv6 Basics

Assign a Static IPv6 Route

IPv6 Static Routes are supported on platforms

ces
Purpose Set up IPv6 static routes prefix: IPv6 route prefix type {slot/port}: interface type and slot/port forwarding router: forwarding routers address tag: route tag Enter the keyword interface followed by the type of interface and slot/port information: For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a loopback interface, enter the keyword loopback followed by the loopback number For a linecard interface, enter the keyword linecard followed by the slot number For a port-channel interface, enter the keyword port-channel followed by the port-channel number For a VLAN interface, enter the keyword vlan followed by the VLAN ID For a Null interface, enter the keyword null followed by the Null interface number

Use the ipv6 route command to configure IPv6 static routes.

Command Syntax ipv6 route prefix type {slot/ port} forwarding router tag Command Mode CONFIGURATION

Telnet with IPv6

IPv6 Telnet is supported on platforms

ces

The Telnet client and server in FTOS support IPv6 connections. You can establish a Telnet session directly to the router using an IPv6 Telnet client, or an IPv6 Telnet connection can be initiated from the router.
Note: Telnet to link local addresses is not supported.

IPv6 Basics | 377

www.dell.com | support.dell.com

Command Syntax telnet ipv6 address

Command Mode EXEC or EXEC Privileged

Purpose Enter the IPv6 Address for the device. ipv6 address : x:x:x:x::x mask : prefix length 0-128

IPv6 addresses are normally written as eight groups of four hexadecimal digits, where each group is separated by a colon (:). Omitting zeros is accepted as described in Addressing earlier in this chapter.

SNMP over IPv6

SNMP is supported on platforms

ces

Simple Network Management Protocol (SNMP) can be configured over IPv6 transport so that an IPv6 host can perform SNMP queries and receive SNMP notifications from a device running FTOS IPv6. The FTOS SNMP-server commands for IPv6 have been extended to support IPv6. Refer to the SNMP and SYSLOG chapter in the FTOS Command Line Interface Reference for more information regarding SNMP commands. snmp-server host snmp-server user ipv6 snmp-server community ipv6 snmp-server community access-list-name ipv6 snmp-server group ipv6 snmp-server group access-list-name ipv6

Show IPv6 Information

All of the following show commands are supported on platforms View specific IPv6 configuration with the following commands.

ces

Command Syntax show ipv6 ?

Command Mode EXEC or EXEC Privileged

Purpose List the IPv6 show options

378

IPv6 Basics

Command Syntax

Command Mode

Purpose

FTOS#show ipv6 ? accounting IPv6 accounting information cam linecard IPv6 CAM Entries for Line Card fib linecard IPv6 FIB Entries for Line Card interface IPv6 interface information mbgproutes MBGP routing table mld MLD information mroute IPv6 multicast-routing table neighbors IPv6 neighbor information ospf OSPF information pim PIM V6 information prefix-list List IPv6 prefix lists route IPv6 routing information rpf RPF table FTOS#

Show an IPv6 Interface

View the IPv6 configuration for a specific interface with the following command.
Command Syntax show ipv6 interface type {slot/ port} Command Mode EXEC Purpose Show the currently running configuration for the specified interface Enter the keyword interface followed by the type of interface and slot/port information: For all brief summary of IPv6 status and configuration , enter the keyword brief. For all IPv6 configured interfaces, enter the keyword configured. For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a loopback interface, enter the keyword loopback followed by the loopback number For a linecard interface, enter the keyword linecard followed by the slot number For a port-channel interface, enter the keyword port-channel followed by the port-channel number For a VLAN interface, enter the keyword vlan followed by the VLAN ID

Figure 17-6 illustrates the show ipv6 interface command output.

IPv6 Basics | 379

www.dell.com | support.dell.com

Figure 17-6. Command Example: show ipv6 interface

FTOS#show ipv6 interface gi 2/2 GigabitEthernet 2/2 is down, line protocol is down IPV6 is enabled Link Local address: fe80::201:e8ff:fe06:95a3 Global Unicast address(es): 3:4:5:6::8, subnet is 3::/24 Global Anycast address(es): Joined Group address(es): ff02::1 ff02::2 ff02::1:ff00:8 ff02::1:ff06:95a3 MTU is 1500 ICMP redirects are not sent DAD is enabled, number of DAD attempts: 1 ND reachable time is 30 seconds ND advertised reachable time is 30 seconds ND advertised retransmit interval is 30 seconds ND router advertisements are sent every 200 seconds ND router advertisements live for 1800 seconds

Show IPv6 Routes

View the global IPv6 routing information with the following command.
Command Syntax show ipv6 route type Command Mode EXEC Purpose Show IPv6 routing information for the specified route type. Enter the keyword: To display information about a network, enter the ipv6 address (X:X:X:X::X). To display information about a host, enter the hostname. To display information about all IPv6 routes (including non-active routes), enter all. To display information about all connected IPv6 routes, enter connected. To display information about brief summary of all IPv6 routes, enter summary. To display information about Border Gateway Protocol (BGP) routes, enter bgp. To display information about ISO IS-IS routes, enter isis. To display information about Open Shortest Path First (OSPF) routes, enter ospf. To display information about Routing Information Protocol (RIP), enter rip. To display information about static IPv6 routes, enter static. To display information about an IPv6 Prefix lists, enter list and the prefix-list name.

380

IPv6 Basics

Figure 17-7 illustrates the show ipv6 route command output.

Figure 17-7. Command Example: show ipv6 route
FTOS#show ipv6 route Codes: C - connected, L - local, S - static, R - RIP, B - BGP, IN - internal BGP, EX - external BGP,LO - Locally Originated, O - OSPF, IA - OSPF inter area, N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2, E1 - OSPF external type 1, E2 - OSPF external type 2, i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, IA - IS-IS inter area, * - candidate default, Gateway of last resort is not set Destination Dist/Metric, Gateway, Last Change ----------------------------------------------------2001::/64 [0/0] Direct, Gi 1/1, 00:28:49 2002::/120 [0/0] Direct, Gi 1/1, 00:28:49 2003::/120 [0/0] Direct, Gi 1/1, 00:28:49

C C C

Figure 17-8 illustrates the show ipv6 route summary command output.
Figure 17-8. Command Example: show ipv6 route summary
FTOS#show ipv6 route summary Route Source connected static Total Active Routes 5 0 5 Non-active Routes 0 0 0

Figure 17-9 illustrates the show ipv6 route static command output.
Figure 17-9. Command Example: show ipv6 route static
FTOS#show ipv6 route static Destination Dist/Metric, Gateway, Last Change ----------------------------------------------------S 8888:9999:5555:6666:1111:2222::/96 [1/0] via 2222:2222:3333:3333::1, Gi 9/1, 00:03:16 S 9999:9999:9999:9999::/64 [1/0] via 8888:9999:5555:6666:1111:2222:3333:4444, 00:03:16

IPv6 Basics | 381

www.dell.com | support.dell.com

Show the Running-Configuration for an Interface

View the configuration for any interface with the following command.
Command Syntax show running-config interface type {slot/port} Command Mode EXEC Purpose Show the currently running configuration for the specified interface Enter the keyword interface followed by the type of interface and slot/port information: For a 10/100/1000 Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/ port information. For the Management interface on the RPM, enter the keyword ManagementEthernet followed by the slot/port information. For a 10 Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information.

Figure 17-10 illustrates the show running-config command output. Note the IPv6 address listed.
Figure 17-10. Command Example: show running-config interface

FTOS#show run int gi 2/2 ! interface GigabitEthernet 2/2 no ip address ipv6 address 3:4:5:6::8/24 shutdown FTOS#

Clear IPv6 Routes

Use the clear IPv6 route command to clear routes from the IPv6 routing table.

Command Syntax clear ipv6 route {* | ipv6 address

prefix-length}

Command Mode EXEC

Purpose Clear (refresh) all or a specific routes from the IPv6 routing table. * : all routes ipv6 address : x:x:x:x::x mask : prefix length 0-128

382

IPv6 Basics

Command Syntax

Command Mode

Purpose

IPv6 addresses are normally written as eight groups of four hexadecimal digits, where each group is separated by a colon (:). Omitting zeros is accepted as described in Addressing earlier in this chapter.

IPv6 Basics | 383

384 www.dell.com | support.dell.com

IPv6 Basics

18
iSCSI Optimization
iSCSI Optimization is supported on platforms This chapter describes how to configure internet small computer system interface (iSCSI) optimization, which enables quality-of-service (QoS) treatment for iSCSI traffic. The topics covered in this chapter include: iSCSI Optimization Overview Default iSCSI Optimization Values iSCSI Optimization Prerequisites Configuring iSCSI Optimization Displaying iSCSI Optimization Information

iSCSI Optimization Overview

iSCSI is a TCP/IP-based protocol for establishing and managing connections between IP-based storage devices and initiators in a storage area network (SAN). iSCSI optimization functions as follows on the , except as noted.

Auto-detection of EqualLogic storage arrays The switch detects any active EqualLogic array directly attached to its ports. Manual configuration to detect Compellent storage arrays where auto-detection is not supported. (This feature is available on platforms ) Automatic configuration of switch ports after detection of storage arrays.

Figure 18-1 shows iSCSI optimization between servers and a storage array in which a stack of three switches connect installed servers (iSCSI initiators) to a storage array (iSCSI targets) in a SAN network. iSCSI optimization running on the master switch is configured to use dot1p priority-queue assignments to ensure that iSCSI traffic in these sessions receives priority treatment when forwarded on stacked switch hardware.

iSCSI Optimization | 385

www.dell.com | support.dell.com

Figure 18-1. iSCSI Optimization Example

Detection and Auto-configuration for Dell EqualLogic Arrays

The iSCSI optimization feature includes auto-provisioning support with the ability to detect directly connected Dell EqualLogic storage arrays and automatically reconfigure the switch to enhance storage traffic flows. The switch uses the link layer discovery protocol (LLDP) to discover Dell EqualLogic devices on the network. If you are using reload-type jump-start command you must enable the LLDP protocol manually. If you delete the startup configuration and reload, the LLDP protocol may already be enabled. If it is not, you must enable it manually. For more information about LLDP, refer to Chapter 21, Link Layer Discovery Protocol.

386

iSCSI Optimization

The following message is displayed the first time a Dell EqualLogic array is detected and describes the configuration changes that are automatically performed:
%STKUNIT0-M:CP %IFMGR-5-IFM_ISCSI_AUTO_CONFIG: This switch is being configured for optimal conditions to support iSCSI traffic which will cause some automatic configuration to occur including jumbo frames and flow-control on all ports; no storm control and spanning-tree port fast to be enabled on the port of detection.

The following syslog message is generated the first time an EqualLogic array is detected:
%STKUNIT0-M:CP %LLDP-5-LLDP_EQL_DETECTED: EqualLogic Storage Array detected on interface Te 1/ 43

At the first detection of an EqualLogic array, an MTU of 12000 for the S4810 and 9252 for S60 is enabled on all ports and port-channels (if it is has not already been enabled). Spanning-tree portfast is enabled on the interface identified by LLDP. Unicast storm control is disabled on the interface identified by LLDP.

Detection and Port Configuration for Dell Compellent Arrays

This feature is available on platforms Switches support the iscsi profile-compellent command to configure a port connected to a Dell Compellent storage array. The command configures a port for the best iSCSI traffic conditions and must be entered in INTERFACE Configuration mode. The following message is displayed the first time you use the iscsi profile-compellent command to configure a port connected to a Dell Compellent storage array and describes the configuration changes that are automatically performed:
%STKUNIT0-M:CP %IFMGR-5-IFM_ISCSI_AUTO_CONFIG: This switch is being configured for optimal conditions to support iSCSI traffic which will cause some automatic configuration to occur including jumbo frames and flow-control on all ports; no storm control and spanning-tree port fast to be enabled on the port of detection.

After you execute the iscsi profile-compellent command, the following actions occur: Jumbo frame size is set to 12000 for the S4810 and 9252 for S55 and S60 on all ports and port-channels, if it is not already enabled. Spanning-tree portfast is enabled on the interface. Unicast storm control is disabled on the interface.

You must enter the iscsi profile-compellent command in INTERFACE configuration mode. For example:
FTOS(conf-if-te-o/50# iscsi profile-compellent)

iSCSI Optimization | 387

www.dell.com | support.dell.com

Auto-detection of Dell Compellent

To auto-detect iSCSI optimization on a switch connected to a Dell Compellent array, follow these steps:
Step 1 Task Configure the auto-detection of Dell Compellent arrays on a port. Default: Dell Compellent disk arrays are not detected. Command
[no] iscsi profile-compellent

Command Mode INTERFACE

Enabling and Disabling iSCSI Optimization

This feature is available on platforms

Note: iSCSI auto-configuration and auto-detection is disabled by default.

If iSCSI is enabled, flow control will be automatically enabled on all interfaces. To disable the flow control on all interfaces, enter the command no flow control rx on tx off and save the configuration. To disable iSCSI optimization, which can turn on flow control again on reboot, enter the command no iscsi enable and save the configuration. When you enable iSCSI on the switch, the following actions occur: Link-level flow control is globally enabled, if it is not already enabled, and PFC is disabled. iSCSI session snooping is enabled. iSCSI LLDP monitoring starts to automatically detect EqualLogic arrays.

The following message is displayed when you enable iSCSI on a switch and describes the configuration changes that are automatically performed:
%STKUNIT0-M:CP %IFMGR-5-IFM_ISCSI_ENABLE: iSCSI has been enabled causing flow control to be enabled on all interfaces. EQL detection and enabling iscsi profile-compellent on an interface may cause some automatic configurations to occur like jumbo frames on all ports and no storm control and spanning tree port-fast on the port of detection.

You can reconfigure any of the auto-provisioned configuration settings that result when you enable iSCSI on a switch. When you disable the iSCSI feature, iSCSI resources are released and the detection of EqualLogic arrays using LLDP is disabled. Disabling iSCSI does not remove the MTU, flow control, portfast, or storm control configuration applied as a result of enabling iSCSI.

388

iSCSI Optimization

Default iSCSI Optimization Values

Table 18-1 shows the default values for the iSCSI optimization feature.
Table 18-1. iSCSI Optimization: Default Parameters Default Value Disabled

Parameter iSCSI Optimization global setting

iSCSI Optimization Prerequisites

iSCSI optimization requires LLDP on the switch. LLDP is disabled by default (refer to Chapter 21, Link Layer Discovery Protocol).

Configuring iSCSI Optimization

This feature is available on platforms To configure iSCSI optimization on a switch, follow these steps:
Note: This is recommended for Dell EqualLogic customers who deploy the S60 and S4810 platforms in their iSCSI environment.

Note: Not following the commands in this sequence may result in partial iSCSI optimizations parameters being enabled, which is not recommended.

Step 1 2 3

Task

Command

Command Mode

(Pre-requisite) For all 1G interfaces on the S60 and S4810, configure them as switchports. For more information on configuring switchports, refer to Chapter 15, Interfaces. (Pre-requisite) Enable LLDP protocol on the ports on the S60 and S4810 platforms to be connected to the Dell EqualLogic storage array. For more information, refer to Chapter 21, Link Layer Discovery Protocol. Enable iscsi optimization on the ports to be connected to the Dell EqualLogic storage array.
iscsi enable

CONFIGURATION

Displaying iSCSI Optimization Information

This feature is available on platforms:

iSCSI Optimization | 389

www.dell.com | support.dell.com

Use the show commands in Table 18-2 to display information on iSCSI optimization
Table 18-2. Command
show run iscsi

Displaying iSCSI Optimization Information Output Displays all globally-configured non-default iSCSI settings in the current FTOS session.

390

iSCSI Optimization

19
Link Aggregation Control Protocol
Link Aggregation Control Protocol is supported on platforms The major sections in the chapter are: Introduction to Dynamic LAGs and LACP LACP Configuration Tasks Shared LAG State Tracking Configure LACP as Hitless LACP Basic Configuration Example

ce s

Introduction to Dynamic LAGs and LACP

A Link Aggregation Group (LAG), referred to as a port channel by FTOS, can provide both load-sharing and port redundancy across line cards. LAGs can be enabled as static or dynamic. The benefits and constraints are basically the same, as described in Port Channel Interfaces in Chapter 15, Interfaces. The unique benefit of a dynamic LAG is that its ports can toggle between participating in the LAG or acting as dedicated ports, whereas ports in a static LAG must be specifically removed from the LAG in order to act alone. FTOS uses LACP to create dynamic LAGs. LACP provides a standardized means of exchanging information between two systems (also called Partner Systems) and automatically establishes the LAG between the systems. LACP permits the exchange of messages on a link to allow their LACP instances to: Reach agreement on the identity of the LAG to which the link belongs. Move the link to that LAG. Enable the transmission and reception functions in an orderly manner.

The FTOS implementation of LACP is based on the standards specified in the IEEE 802.3: Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications. LACP functions by constantly exchanging custom MAC PDUs across LAN Ethernet links. The protocol packets are only exchanged between ports that are configured as LACP capable.

Link Aggregation Control Protocol | 391

www.dell.com | support.dell.com

Important Points to Remember

In FTOS software version 8.3.3.9, the default startup configuration is all 1G ports in L2 switch mode, admin up. LACP enables you to add members to a port channel (LAG) as long as it has no static members. Conversely, if the LAG already contains a statically defined member (channel-member command), the port-channel mode command is not permitted. A static LAG cannot be created if a dynamic LAG using the selected number already exists. No dual membership in static and dynamic LAGs: If a physical interface is a part of a static LAG, then the command port-channel-protocol lacp will be rejected on that interface. If a physical interface is a part of a dynamic LAG, it cannot be added as a member of a static LAG. The command channel-member gigabitethernet x/y will be rejected in the static LAG interface for that physical interface. A dynamic LAG can be created with any type of configuration. There is a difference between the shutdown and no interface port-channel: The shutdown command on LAG xyz disables the LAG and retains the user commands. However, the system does not allow the channel number xyz to be statically created. The command no interface port-channel channel-number deletes the specified LAG, including a dynamically created LAG. This command causes all LACP-specific commands on the member interfaces to be removed. The interfaces are restored to a state that is ready to be configured. Note: There will be no configuration on the interface since that condition is required for an interface to be part of a LAG. Link dampening can be configured on individual members of a LAG. See Link Debounce Timer for more information. LACP cannot add an interface to a LAG if one of the LAG members is shut down on the remote interface. If a remote LAG member is shut down, Message 1 appears on the local system when you attempt to add a member. In this case, enable all members of the LAG on the remote system, and then add any new members on the local system.

Message 1 LACP Remote Port Down Error Message

% Error: This port property does not match with other LAG member.

LACP modes
FTOS provides the following three modes for configuration of LACP: OffIn this state, an interface is not capable of being part of a dynamic LAG. LACP does not run on any port that is configured to be in this state. ActiveIn this state, the interface is said to be in the active negotiating state. LACP runs on any link that is configured to be in this state. A port in Active state also automatically initiates negotiations with other ports by initiating LACP packets.

392

Link Aggregation Control Protocol

PassiveIn this state, the interface is not in an active negotiating state, but LACP will run on the link. A port in Passive state also responds to negotiation requests (from ports in Active state). Ports in Passive state respond to LACP packets.

FTOS supports LAGs in the following cases: A port in Active state can set up a port channel (LAG) with another port in Active state. A port in Active state can set up a LAG with another port in Passive state. A port in Passive state cannot set up a LAG with another port in Passive state.

LACP Configuration Commands

If aggregated ports are configured with compatible LACP modes (Off, Active, Passive), LACP can automatically link them, as defined in IEEE 802.3, Section 43. The following commands configure LACP:
Command Syntax [no] lacp system-priority priority-value Command Mode CONFIGURATION Purpose Configure the system priority. Range: 1 65535 (the higher the number, the lower the priority) Default: 32768 Enable or disable LACP on any LAN port: Default is LACP disabled This command creates a new context. Configure LACP mode. Default is LACP active number cannot statically contain any links Configure port priority. Ranges: 1 65535 (the higher the number, the lower the priority) Default: 32768

[no] port-channel-protocol lacp

INTERFACE

[no] port-channel number mode [active | passive | off] [no] lacp port-priority priority-value

LACP

LACP

LACP Configuration Tasks

The tasks covered in this section are: Create a LAG Configure the LAG interfaces as dynamic Set the LACP long timeout Monitor and Debugging LACP Configure Shared LAG State Tracking

Link Aggregation Control Protocol | 393

www.dell.com | support.dell.com

Create a LAG
To create a dynamic port channel (LAG), define the LAG and then the LAG interfaces. Use the interface port-channel and switchport commands, as shown in Figure 19-1, which uses the example of LAG 32:
Figure 19-1. Placing a LAG into the Default VLAN
FTOS(conf)#interface port-channel 32 FTOS(conf-if-po-32)#no shutdown FTOS(conf-if-po-32)#switchport

The LAG is in the default VLAN. To place the LAG into a non-default VLAN, use the tagged command on the LAG (Figure 19-2):
Figure 19-2. Placing a LAG into a Non-default VLAN
FTOS(conf)#interface vlan 10 FTOS(conf-if-vl-10)#tagged port-channel 32

Port-channel interfaces support link aggregation. They provide redundancy by aggregating physical interfaces into a single logical interface. Port-channel interfaces can be used to create large-capacity interfaces by utilizing a group of low-speed links. For more information, refer to the Port Channel Interfaces section in the Interfaces chapter.

Table 19-1. Platform E-Series C-Series

Number of Port-channels per Platform Port-channels 255 128 52 128 Members/Channel 16 8 8 8

S-Series: S50 and S25 S-Series: S55, S60 and S4810

Configure the LAG interfaces as dynamic

After creating a LAG, configure the dynamic LAG interfaces. Figure 19-3 shows ports 3/15, 3/16, 4/15, and 4/16 added to LAG 32 in LACP mode with the command port-channel-protocol lacp.

394

Link Aggregation Control Protocol

Figure 19-3. Creating a Dynamic LAG Example

FTOS(conf)#interface Gigabitethernet 3/15 FTOS(conf-if-gi-3/15)#no shutdown FTOS(conf-if-gi-3/15)#port-channel-protocol lacp FTOS(conf-if-gi-3/15-lacp)#port-channel 32 mode active ... FTOS(conf)#interface Gigabitethernet 3/16 FTOS(conf-if-gi-3/16)#no shutdown FTOS(conf-if-gi-3/16)#port-channel-protocol lacp FTOS(conf-if-gi-3/16-lacp)#port-channel 32 mode active ... FTOS(conf)#interface Gigabitethernet 4/15 FTOS(conf-if-gi-4/15)#no shutdown FTOS(conf-if-gi-4/15)#port-channel-protocol lacp FTOS(conf-if-gi-4/15-lacp)#port-channel 32 mode active ... FTOS(conf)#interface Gigabitethernet 4/16 FTOS(conf-if-gi-4/16)#no shutdown FTOS(conf-if-gi-4/16)#port-channel-protocol lacp FTOS(conf-if-gi-4/16-lacp)#port-channel 32 mode active

The port-channel 32 mode active command shown above may be successfully issued as long as there is no existing static channel-member configuration in LAG 32.

Set the LACP long timeout

PDUs are exchanged between port channel (LAG) interfaces to maintain LACP sessions. PDUs are transmitted at either a slow or fast transmission rate, depending upon the LACP timeout value. The timeout value is the amount of time that a LAG interface waits for a PDU from the remote system before bringing the LACP session down. The default timeout value is 1 second; it can be configured to be 30 seconds. Invoking the longer timeout might prevent the LAG from flapping if the remote system is up but temporarily unable to transmit PDUs due to a system interruption.
Note: The 30-second timeout is available for dynamic LAG interfaces only. The lacp long-timeout command can be entered for static LAGs, but it has no effect.

To configure the LACP long timeout (Figure 196):

Step 1 Task Set the LACP timeout value to 30 seconds. Command Syntax
lacp long-timeout

Command Mode CONFIG-INT-PO

Link Aggregation Control Protocol | 395

www.dell.com | support.dell.com

Figure 19-4. Invoking the LACP Long Timeout

FTOS(conf)# interface port-channel 32 FTOS(conf-if-po-32)#no shutdown FTOS(conf-if-po-32)#switchport FTOS(conf-if-po-32)#lacp long-timeout FTOS(conf-if-po-32)#end FTOS# show lacp 32 Port-channel 32 admin up, oper up, mode lacp Actor System ID: Priority 32768, Address 0001.e800.a12b Partner System ID: Priority 32768, Address 0001.e801.45a5 Actor Admin Key 1, Oper Key 1, Partner Oper Key 1 LACP LAG 1 is an aggregatable link A - Active LACP, B - Passive LACP, C - Short Timeout, D - Long Timeout E - Aggregatable Link, F - Individual Link, G - IN_SYNC, H - OUT_OF_SYNC I - Collection enabled, J - Collection disabled, K - Distribution enabled L - Distribution disabled, M - Partner Defaulted, N - Partner Non-defaulted, O - Receiver is in expired state, P - Receiver is not in expired state Port Gi 10/6 is enabled, LACP is enabled and mode is lacp Actor Admin: State ADEHJLMP Key 1 Priority 128

Note: View PDU exchanges and the timeout value using the command debug lacp. See Monitor and Debugging LACP.

Monitor and Debugging LACP

The system log (syslog) records faulty LACP actions. To debug LACP, use the following command:
Command Syntax [no] debug lacp [config | events | pdu [in | out | [interface [in | out]]]] Command Mode EXEC Purpose Debug LACP, including configuration and events.

Shared LAG State Tracking

Shared LAG State Tracking provides the flexibility to bring down a port channel (LAG) based on the operational state of another LAG. At any time, only two LAGs can be a part of a group such that the fate (status) of one LAG depends on the other LAG. In Figure 19-5, line-rate traffic from R1 destined for R4 follows the lowest-cost route via R2, as shown. Traffic is equally distributed between LAGs 1 and 2. If LAG 1 fails, all traffic from R1 to R4 flows across LAG 2 only. This condition over-subscribes the link, and packets are dropped.

396

Link Aggregation Control Protocol

Figure 19-5. LAGs using ECMP without Shared LAG State Tracking
R4
Po 1 failure

Po

R1

Po

2
Po 2 over-subscribed

R2

R3 fnC0049mp

To avoid packet loss, traffic must be re-directed through the next lowest-cost link (R3 to R4). FTOS has the ability to bring LAG 2 down in the event that LAG 1 fails, so that traffic can be re-directed, as described. This is what is meant by Shared LAG State Tracking. To achieve this functionality, you must group LAG 1 and LAG 2 into a single entity, called a failover group.

Configure Shared LAG State Tracking

To configure Shared LAG State Tracking, you configure a failover group:
Step 1 2 Task Enter port-channel failover group mode. Create a failover group and specify the two port-channels that will be members of the group. Command
port-channel failover-group group number port-channel number port-channel number

Command Mode CONFIGURATION CONFIG-PO-FAILOVER-GRP

In Figure 19-6, LAGs 1 and 2 have been placed into to the same failover group.
Figure 19-6. Configuring Shared LAG State Tracking
R2#config R2(conf)#port-channel failover-group R2(conf-po-failover-grp)#group 1 port-channel 1 port-channel 2

View the failover group configuration using the show running-configuration po-failover-group command, as shown in Figure 19-7.
Figure 19-7. Viewing Shared LAG State Tracking in the Running-configuration
R2#show running-config po-failover-group ! port-channel failover-group group 1 port-channel 1 port-channel 2

Link Aggregation Control Protocol | 397

www.dell.com | support.dell.com

In Figure 19-8, LAGs 1 and 2 are members of a failover group. LAG 1 fails and LAG 2 is brought down upon the failure. This effect is logged by Message 2, in which a console message declares both LAGs down at the same time.
Figure 19-8. Shared LAG State Tracking
R2(conf)# port-channel failover-group R2(conf-po-failover-grp)# group 1 port-channel 1 port-channel 2

R4

Po

R1

R2

Po

R3 fnC0049mp

Message 2 Shared LAG State Tracking Console Message

2d1h45m: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Po 1 2d1h45m: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Po 2

View the status of a failover group member using the command show interface port-channel, as shown in Figure 19-9.
Figure 19-9. Viewing Status of a Failover Group Member
R2#show interface Port-channel 2 Port-channel 2 is up, line protocol is down (Failover-group 1 is down) Hardware address is 00:01:e8:05:e8:4c, Current address is 00:01:e8:05:e8:4c Interface index is 1107755010 Minimum number of links to bring Port-channel up is 1 Port-channel is part of failover-group 1 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 1000 Mbit Members in this channel: Gi 1/17(U) ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:01:28 Queueing strategy: fifo

Note: The set of console messages shown in Message 2 appear only if Shared LAG State Tracking is configured on that router (the feature can be configured on one or both sides of a link). For example, in Figure 19-8, if Shared LAG State Tracking is configured on R2 only, then no messages appear on R4 regarding the state of LAGs in a failover group.

Important Points about Shared LAG State Tracking

This feature is available for static and dynamic LAGs.

398

Link Aggregation Control Protocol

Only a LAG can be a member of a failover group. Shared LAG State Tracking can be configured on one side of a link or on both sides. If a LAG that is part of a failover group is deleted, the failover group is deleted. If a LAG moves to the down state due to this feature, its members may still be in the up state.

Configure LACP as Hitless

Configure LACP as Hitless is supported only on platforms:

ce

LACP on Dell Networking systems can be configured to be hitless. When configured as hitless, there is no noticeable impact on dynamic LAG state upon an RPM failover. Critical LACP state information is synchronized between the two RPMs. See Hitless Behavior in the High Available chapter of the C-Series or E-Series FTOS Configuration Guide.
Configure LACP to be hitless using the command redundancy protocol lacp from CONFIGURATION mode, as shown in Figure 19-10.
Figure 19-10. Enabling Hitless LACP

FTOS(conf)#redundancy protocol lacp

FTOS#show running-config redundancy ! redundancy protocol lacp FTOS# FTOS#show running-config interface gigabitethernet 0/12 ! interface GigabitEthernet 0/12 no ip address ! port-channel-protocol LACP port-channel 200 mode active no shutdown

LACP Basic Configuration Example

The screenshots in this section are based on the example topology shown in Figure 19-11. Two routers are named ALPHA and BRAVO, and their hostname prompts reflect those names. The sections are: Configuring a LAG on ALPHA Summary of the configuration on ALPHA Summary of the configuration on BRAVO

Link Aggregation Control Protocol | 399

www.dell.com | support.dell.com

Figure 19-11.

LACP Sample Topology

Port Channel 10 ALPHA Gig 2/31 Gig 2/32 Gig 3/22 BRAVO Gig 3/21

Gig 2/33

Gig 3/23

Configuring a LAG on ALPHA

Figure 19-12. Creating a LAG on ALPHA
Alpha(conf)#interface port-channel 10 Alpha(conf-if-po-10)#no ip address Alpha(conf-if-po-10)#switchport Alpha(conf-if-po-10)#no shutdown Alpha(conf-if-po-10)#show config ! interface Port-channel 10 no ip address switchport no shutdown ! Alpha(conf-if-po-10)#

400

Link Aggregation Control Protocol

Figure 19-13.

Inspecting a LAG Port Configuration on ALPHA

Alpha#sh int gig 2/31 GigabitEthernet 2/31 is up, line protocol is up Port is part of Port-channel 10 Hardware is Force10Eth, address is 00:01:e8:06:95:c0 Current address is 00:01:e8:06:95:c0 Interface index is 109101113 Port will not be disabled on partial SFM failure Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes Shows the speed of this physical interface. LineSpeed 1000 Mbit, Mode full duplex, Slave Also shows it is the slave of the GigE link. Flowcontrol rx on tx on ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:02:11 Queueing strategy: fifo Input Statistics: 132 packets, 16368 bytes 0 Vlans 0 64-byte pkts, 12 over 64-byte pkts, 120 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 132 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 136 packets, 16718 bytes, 0 underruns 0 64-byte pkts, 15 over 64-byte pkts, 121 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 136 Multicasts, 0 Broadcasts, 0 Unicasts 0 Vlans, 0 throttles, 0 discarded, 0 collisions, 0 wreddrops Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Time since last interface status change: 00:02:14

Link Aggregation Control Protocol | 401

www.dell.com | support.dell.com

Figure 19-14.

Inspecting Configuration of LAG 10 on ALPHA

Indicates the MAC address assigned to the LAG. This does NOT match any of the physical interface MAC addresses.

Alpha#show int port-channel 10 Port-channel 10 is up, line protocol is up Created by LACP protocol Hardware address is 00:01:e8:06:96:63, Current address is 00:01:e8:06:96:63 Interface index is 1107755018 Confirms the number of links to bring up Minimum number of links to bring Port-channel up is 1 the LAG and that this is a switch Internet address is not set port instead of a router port. MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 3000 Mbit Members in this channel: Gi 2/31(U) Gi 2/32(U) Gi 2/33(U) ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:04:09 Confirms the total bandwidth for this Queueing strategy: fifo LAG and which interfaces are active. Input Statistics: 621 packets, 78732 bytes 0 Vlans 0 64-byte pkts, 18 over 64-byte pkts, 603 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 621 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 630 packets, 79284 bytes, 0 underruns 0 64-byte pkts, 30 over 64-byte pkts, 600 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 630 Multicasts, 0 Broadcasts, 0 Unicasts 0 Vlans, 0 throttles, 0 discarded, 0 collisions, 0 wreddrops Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 2 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 2 packets/sec, 0.00% of line-rate Time since last interface status change: 00:03:38

402

Link Aggregation Control Protocol

Figure 19-15.

Using the show lacp Command to Verify LAG 10 Status on ALPHA

Alpha#sho lacp 10 Port-channel 10 admin up, oper up, mode lacp Actor System ID: Priority 32768, Address 0001.e806.953e Partner System ID: Priority 32768, Address 0001.e809.c24a Actor Admin Key 10, Oper Key 10, Partner Oper Key 10 LACP LAG 10 is an aggregatable link

Shows LAG status

A - Active LACP, B - Passive LACP, C - Short Timeout, D - Long Timeout E - Aggregatable Link, F - Individual Link, G - IN_SYNC, H - OUT_OF_SYNC I - Collection enabled, J - Collection disabled, K - Distribution enabled L - Distribution disabled, M - Partner Defaulted, N - Partner Non-defaulted, O - Receiver is in expired state, P - Receiver is not in expired state Port Gi 2/31 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 Port Gi 2/32 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 Port Gi 2/33 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 Alpha# Interfaces participating in the LAG are included here.

Link Aggregation Control Protocol | 403

www.dell.com | support.dell.com

Summary of the configuration on ALPHA

Figure 19-16. Summary of the configuration on ALPHA
Alpha(conf-if-po-10)#int gig 2/31 Alpha(conf-if-gi-2/31)#no ip address Alpha(conf-if-gi-2/31)#no switchport Alpha(conf-if-gi-2/31)#shutdown Alpha(conf-if-gi-2/31)#port-channel-protocol lacp Alpha(conf-if-gi-2/31-lacp)#port-channel 10 mode active Alpha(conf-if-gi-2/31-lacp)#no shut Alpha(conf-if-gi-2/31)#show config ! interface GigabitEthernet 2/31 no ip address ! port-channel-protocol LACP port-channel 10 mode active no shutdown ! Alpha(conf-if-gi-2/31)# interface Port-channel 10 no ip address switchport no shutdown interface GigabitEthernet 2/31 no ip address no switchport switchport port-channel-protocol LACP port-channel 10 mode active no shutdown

404

Link Aggregation Control Protocol

Summary of the configuration on BRAVO

Figure 19-17. Summary of the configuration on BRAVO
Bravo(conf-if-gi-3/21)#int port-channel 10 Bravo(conf-if-po-10)#no ip add Bravo(conf-if-po-10)#switch Bravo(conf-if-po-10)#no shut Bravo(conf-if-po-10)#show config ! interface Port-channel 10 no ip address switchport no shutdown ! Bravo(conf-if-po-10)#exit Bravo(conf)#int gig 3/21 Bravo(conf)#no ip address Bravo(conf)#no switchport Bravo(conf)#shutdown Bravo(conf-if-gi-3/21)#port-channel-protocol lacp Bravo(conf-if-gi-3/21-lacp)#port-channel 10 mode active Bravo(conf-if-gi-3/21-lacp)#no shut Bravo(conf-if-gi-3/21)#end ! interface GigabitEthernet 3/21 no ip address ! port-channel-protocol LACP port-channel 10 mode active no shutdown Bravo(conf-if-gi-3/21)#end int port-channel 10 no ip address switchport no shutdown show config int gig 3/21 no ip address no switchport shutdown port-channel-protocol lacp port-channel 10 mode active no shut show config end

Link Aggregation Control Protocol | 405

www.dell.com | support.dell.com

Figure 19-18.

Using the show interface Command to Inspect a LAG Port on BRAVO

Shows the status of this nterface. Also shows it is part of LAG 10.

Bravo#show int gig 3/21 GigabitEthernet 3/21 is up, line protocol is up Port is part of Port-channel 10 Hardware is Force10Eth, address is 00:01:e8:09:c3:82 Current address is 00:01:e8:09:c3:82 Shows that this is a Layer 2 port. Interface index is 140034106 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 1000 Mbit, Mode full duplex, Master Shows the speed of this physical interface. Flowcontrol rx on tx on Also shows it is the Master of the GigE link. ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:15:05 Queueing strategy: fifo Input Statistics: 708 packets, 89934 bytes 0 Vlans 0 64-byte pkts, 15 over 64-byte pkts, 693 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 708 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 705 packets, 89712 bytes, 0 underruns 0 64-byte pkts, 12 over 64-byte pkts, 693 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 705 Multicasts, 0 Broadcasts, 0 Unicasts 0 Vlans, 0 throttles, 0 discarded, 0 collisions, 0 wreddrops Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Time since last interface status change: 00:12:39

406

Link Aggregation Control Protocol

Figure 19-19.

Using the show interfaces port-channel Command to Inspect LAG 10

Indicates the MAC address assigned to the LAG. This does NOT match any of the physical interface MAC addresses.

TOS#sh int port 10 ort-channel 10 is up, line protocol is up Created by LACP protocol Hardware address is 00:01:e8:09:c4:ef, Current address is 00:01:e8:09:c4:ef nterface index is 1107755018 Confirms the number of links to bring up Minimum number of links to bring Port-channel up is 1 the LAG and that this is a switch nternet address is not set port instead of a router port. MTU 1554 bytes, IP MTU 1500 bytes ineSpeed 3000 Mbit Confirms the total bandwidth for this Members in this channel: Gi 3/21(U) Gi 3/22(U) Gi 3/23(U) LAG and which interfaces are active. ARP type: ARPA, ARP Timeout 04:00:00 ast clearing of "show interface" counters 00:13:07 Queueing strategy: fifo nput Statistics: 2189 packets, 278744 bytes 0 Vlans 0 64-byte pkts, 32 over 64-byte pkts, 2157 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 2189 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 2173 packets, 277350 bytes, 0 underruns 0 64-byte pkts, 19 over 64-byte pkts, 2154 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 2173 Multicasts, 0 Broadcasts, 0 Unicasts 0 Vlans, 0 throttles, 0 discarded, 0 collisions, 0 wreddrops Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 2 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 2 packets/sec, 0.00% of line-rate ime since last interface status change: 00:13:00

TOS#

Link Aggregation Control Protocol | 407

www.dell.com | support.dell.com

Figure 19-20.

Using the show lacp Command to Inspect LAG Status

FTOS#show lacp 10 Port-channel 10 admin up, oper up, mode lacp Actor System ID: Priority 32768, Address 0001.e809.c24a Partner System ID: Priority 32768, Address 0001.e806.953e Actor Admin Key 10, Oper Key 10, Partner Oper Key 10 LACP LAG 10 is an aggregatable link

Shows LAG status

A - Active LACP, B - Passive LACP, C - Short Timeout, D - Long Timeout E - Aggregatable Link, F - Individual Link, G - IN_SYNC, H - OUT_OF_SYNC I - Collection enabled, J - Collection disabled, K - Distribution enabled L - Distribution disabled, M - Partner Defaulted, N - Partner Non-defaulted, O - Receiver is in expired state, P - Receiver is not in expired state Port Gi 3/21 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 Port Gi 3/22 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 Port Gi 3/23 is enabled, LACP is enabled and mode is lacp Actor Admin: State ACEHJLMP Key 10 Priority 32768 Oper: State ACEGIKNP Key 10 Priority 32768 Partner Admin: State BDFHJLMP Key 0 Priority 0 Oper: State ACEGIKNP Key 10 Priority 32768 FTOS# Interfaces participating in the LAG are included here.

PPP is a connection-oriented protocol that enables layer two links over a variety of different physical layer connections. It is supported on both synchronous and asynchronous lines, and can operate in half-duplex or full-duplex mode. It was designed to carry IP traffic but is general enough to allow any type of network layer datagram to be sent over a PPP connection. As its name implies, it is for point-to-point connections between exactly two devices, and assumes that frames are sent and received in the same order.

408

Link Aggregation Control Protocol

20
Layer 2
Layer 2 features are supported on platforms:

ces

This chapter describes the following Layer 2 features: Managing the MAC Address Table MAC Learning Limit NIC Teaming Microsoft Clustering Configuring Redundant Pairs Restricting Layer 2 Flooding Far-end Failure Detection

Managing the MAC Address Table

FTOS provides the following management activities for the MAC address table: Clear the MAC Address Table Set the Aging Time for Dynamic Entries Configure a Static MAC Address Display the MAC Address Table

Clear the MAC Address Table

You may clear the MAC address table of dynamic entries:
Task Clear a MAC address table of dynamic entries. address deletes the specified entry all deletes all dynamic entries interface deletes all entries for the specified interface vlan deletes all entries for the specified VLAN Command Syntax
clear mac-address-table dynamic {address | all | interface | vlan}

Command Mode EXEC Privilege

Layer 2 | 409

www.dell.com | support.dell.com

Set the Aging Time for Dynamic Entries

Learned MAC addresses are entered in the table as dynamic entries, which means that they are subject to aging. For any dynamic entry, if no packet arrives on the switch with the MAC address as the source or destination address within the timer period, the address is removed from the table. The default aging time is 1800 seconds.
Task Disable MAC address aging for all dynamic entries. Specify an aging time. Command Syntax
mac-address-table aging-time 0

Command Mode CONFIGURATION CONFIGURATION

mac-address-table aging-time seconds

Range: 10-1000000

Set the Aging Time for Dynamic Entries on a VLAN

Set the Aging Time for Dynamic Entries on a VLAN is available only on platform:
Task Specify an aging time. Command Syntax
mac-address-table aging-time seconds

e
Command Mode INTERFACE VLAN

Range: 1-1000000

FTOS Behavior: The time elapsed before the configured MAC aging time expires is not precisely as configured. For example, the VLAN configuration mac-address-table aging-time 1, does not remove dynamic entries from the CAM after precisely 1 second. The actual minimum aging time for entries is approximately 5 seconds because this is the default MAC address table scanning interval. Therefore, MAC aging configurations of less than 5 seconds, as in this example, might be ineffective. Configuring mac-address-table station-move time-interval 500, solves this limitation. Reducing the scanning interval to the minimum, 500 milliseconds, increases the detection speed, which results in FTOS clearing entries closer to the actual desired aging time.

Configure a Static MAC Address

A static entry is one that is not subject to aging. Static entries must be entered manually:
Task Create a static MAC address entry in the MAC address table. Command Syntax
mac-address-table static

Command Mode CONFIGURATION

410

Layer 2

Display the MAC Address Table

To display the contents of the MAC address table:
Task Display the contents of the MAC address table. address displays the specified entry. aging-time displays the configured aging-time. count displays the number of dynamic and static entries for all VLANs, and the total number of entries. dynamic displays only dynamic entries interface displays only entries for the specified interface. static displays only static entries. vlan displays only entries for the specified VLAN. Command Syntax
show mac-address-table [address | aging-time [vlan vlan-id]| count | dynamic | interface | static | vlan]

Command Mode EXEC Privilege

MAC Learning Limit

This section has the following sub-sections: mac learning-limit dynamic mac learning-limit station-move mac learning-limit no-station-move Learning Limit Violation Actions Station Move Violation Actions Recovering from Learning Limit and Station Move Violations Per-VLAN MAC Learning Limit

MAC Address Learning Limit is a method of port security on Layer 2 port-channel and physical interfaces, and VLANs. It enables you to set an upper limit on the number of MAC addresses that learned on an interface/VLAN. After the limit is reached, the system drops all traffic from a device with an unlearned MAC address.
FTOS Behavior: When configuring MAC Learning Limit on a port or VLAN the configuration is accepted (becomes part of running-config and show mac learning-limit interface) before the system verifies that sufficient CAM space exists. If the CAM check fails, the a message is displayed: %E90MH:5 %ACL_AGENT-2-ACL_AGENT_LIST_ERROR: Unable to apply access-list Mac-Limit on GigabitEthernet 5/84 In this case, the configuration is still present in the running-config and show output. Remove the configuration before re-applying a MAC learning limit with lower value. Also, ensure that Syslog messages can be viewed on your session. Note: The CAM-check failure message beginning in FTOS version 8.3.1.0 is different from versions 8.2.1.1 and earlier, which read: % Error: ACL returned error % Error: Remove existing limit configuration if it was configured before

Layer 2 | 411

www.dell.com | support.dell.com

To set a MAC learning limit on an interface:

Task Specify the number of MAC addresses that the system can learn off a Layer 2 interface. Command Syntax
mac learning-limit address_limit

Command Mode INTERFACE

Three options are available with the mac learning-limit command: dynamic, no-station-move, and station-move.

Note: An SNMP trap is available for mac learning-limit station-move. No other SNMP traps are available for MAC Learning Limit, including limit violations.

mac learning-limit dynamic

The MAC address table is stored on the Layer 2 FIB region of the CAM (and the Layer 2 ACL region on the E-Series). On the C-Series and S-Series the Layer 2 FIB region allocates space for static MAC address entries and dynamic MAC address entries (all MAC address entries on the E-Series are dynamic). When MAC Learning Limit is enabled, entries created on this port are static by default. When you configure the dynamic option, learned MAC addresses are stored in the dynamic region and are subject to aging. Entries created before this option is set are not affected.
FTOS Behavior: If you do not configure the dynamic option, the C-Series and S-Series do not detect station moves in which a MAC address learned off of a MAC-limited port is learned on another port on same line card. Therefore, FTOS does not take any configured station-move violation action. When a MAC address is relearned on any other linecard (any line card except the one to which the original MAC-limited port belongs), the station-move is detected, and the system takes the configured the violation action.

mac learning-limit station-move

mac learning-limit station-move is available only on platforms:

cs

The station-move option, allows a MAC address already in the table to be learned off of another interface. For example, if you disconnect a network device from one interface and reconnect it to another interface, the MAC address is learned on the new interface. When the system detects this station move, the system clears the entry learned on the original interface, and installs a new entry on the new interface.

412

Layer 2

mac learning-limit no-station-move

The no-station-move option, also known as sticky MAC, provides additional port security by preventing a station move. When this option is configured, the first entry in the table is maintained instead of creating a new entry on the new interface. no-station-move is the default behavior. Entries created before this option is set are not affected.
FTOS Behavior: The C-Series and S-Series do not generate a station-move violation log entry for physical interfaces or port-channels when you configure mac learning-limit or when you configure mac learning-limit station-move-violation log. FTOS detects a station-move violation only when you configure mac learning-limit dynamic, and logs the violation only when you configure the mac learning-limit station-move-violation log, as shown below: FTOS(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 no ip address switchport mac learning-limit 1 dynamic no-station-move mac learning-limit station-move-violation log no shutdown

To display a list of all interfaces with a MAC learning limit:

Task Display a list of all interfaces with a MAC learning limit. Command Syntax
show mac learning-limit

Command Mode EXEC Privilege

Learning Limit Violation Actions

Learning Limit Violation Actions are supported only on platform:

You can configure the system to take an action when the MAC learning limit is reached on an interface and a new address is received using one of the following options with the mac learning-limit command:
Task Generate a system log message when the MAC learning limit is exceeded. Shut down the interface and generate a system log message when the MAC learning limit is exceeded. Command Syntax
learn-limit-violation log

Command Mode INTERFACE INTERFACE

learn-limit-violation shutdown

Station Move Violation Actions

Station Move Violation Actions are supported only on platforms: s , , and

Layer 2 | 413

www.dell.com | support.dell.com

is the default behavior (see mac learning-limit no-station-move). You can configure the system to take an action if a station move occurs using one the following options with the mac learning-limit command:.
no-station-move

Task Generate a system log message indicating a station move. Shut down the first port to learn the MAC address. Shut down the second port to learn the MAC address. Shut down both the first and second port to learn the MAC address.

Command Syntax
station-move-violation log

Command Mode INTERFACE INTERFACE INTERFACE INTERFACE

station-move-violation shutdown-original

station-move-violation shutdown-offending

station-move-violation shutdown-both

To display a list of interfaces configured with MAC learning limit or station move violation actions:
Task Display a list of all of the interfaces configured with MAC learning limit or station move violation. Command Syntax
show mac learning-limit violate-action

Command Mode CONFIGURATION

Recovering from Learning Limit and Station Move Violations

After a learning-limit or station-move violation shuts down an interface, you must manually reset it:
Task Reset interfaces in ERR_Disabled state caused by a learning limit violation or station move violation. Reset interfaces in ERR_Disabled state caused by a learning limit violation. Reset interfaces in ERR_Disabled state caused by a station move violation. Command Syntax
mac learning-limit reset

Command Mode CONFIGURATION

mac learning-limit reset learn-limit-violation [interface | all] mac learning-limit reset station-move-violation [interface | all]

CONFIGURATION CONFIGURATION

Note: Alternatively, you can reset the interface by shutting it down using the shutdown command, and then reenabling it using the command no shutdown.

Per-VLAN MAC Learning Limit

Per-VLAN MAC Learning Limit is available only on platform:

414

Layer 2

An individual MAC learning limit can be configured for each VLAN using Per-VLAN MAC Learning Limit. One application of Per-VLAN MAC Learning Limit is on access ports. In Figure 20-1, an Internet Exchange Point (IXP) connects multiple Internet Service Provider (ISP). An IXP can provide several types of services to its customers including public an private peering. Public peering means that all customers are connected to one VLAN, and if one ISP wants to peer with another ISP, it establishes a BGP peering session over this VLAN. Private Peering means that the IXP sets up a separate VLAN between two customers that want to peer privately; only the ports of these two ISPs would belong to this VLAN, and they would peer via BGP. In Figure 20-1, Per-VLAN MAC Learning Limit is used on the access ports for the ISPs that have subscribed to private and public peering since these access ports are members of multiple VLANs.
Figure 20-1. Per-VLAN MAC Learning Limit
Internet Exchange Point

802.1QTagged

interface GigabitEthernet 1/1 ... mac learning-limit 1 vlan 10 mac learning-limit 1 vlan 20

ISP A ISP B ISP C

ISP A, B, and C are all public peers through VLAN 10. In addition, ISP A and C are private peers on a separate VLAN, VLAN 20. Since the access ports for ISP A and C are members of multiple VLANs, Per-VLAN MAC Learning Limit can be applied to those ports.

Task Configure a MAC learning limit on a VLAN. Display the MAC learning limit counters for a VLAN.

Command Syntax
mac learning-limit limit vlan vlan-id

Command Mode INTERFACE

show mac learning-limit [interface slot/port [vlan vlan-id]]

EXEC Privilege

Layer 2 | 415

www.dell.com | support.dell.com

Task

Command Syntax
Unknown SA Drops 0 0 0 0 Unknown SA Drops 0 0 Unknown SA Drops 0 0 0 0 0 0 0 0

Command Mode

FTOS#show mac learning-limit Interface Vlan Learning Dynamic Static Slot/port Id Limit MAC count MAC count Gi 5/84 2 2 0 Gi 5/84 * 5 0 Gi 5/85 3 3 0 Gi 5/85 * 10 0 FTOS#show mac learning-limit interface gig 5/84 Interface Vlan Learning Dynamic Static Slot/port Id Limit MAC count MAC count Gi 5/84 2 2 0 Gi 5/84 * 5 0 FTOS#show mac learning-limit interface gig 5/84 vlan 2 Interface Vlan Learning Dynamic Static Slot/port Id Limit MAC count MAC count Gi 5/84 2 2 0

NIC Teaming
NIC teaming is a feature that allows multiple network interface cards in a server to be represented by one MAC address and one IP address in order to provide transparent redundancy, balancing, and to fully utilize network adapter resources. Figure 20-2 shows a topology where two NICs have been teamed together. In this case, if the primary NIC fails, traffic switches to the secondary NIC, since they are represented by the same set of addresses.
Figure 20-2. Redundant NICs with NIC Teaming

X
MAC: A:B:C:D A:B IP: 1.1.1.1

Port 0/1

k Active Lin Port 0/5

fnC0025mp

416

Layer 2

When NIC teaming is employed, consider that the server MAC address is originally learned on Port 0/1 of the switch (Figure 20-3). When the NIC fails, the same MAC address is learned on Port 0/5 of the switch. The MAC address must be disassociated with the one port and re-associated with another in the ARP table; in other words, the ARP entry must be moved. To ensure that this happens, you must configure the command mac-address-table station-move refresh-arp on the Dell Networking switch at the time that NIC teaming is being configured on the server.
Note: If this command is not configured, traffic continues to be forwarded to the failed NIC until the ARP entry on the switch times out.

Figure 20-3. Configuring mac-address-table station-move refresh-arp Command

X
MAC: A:B:C:D A:B IP: 1.1.1.1 Move MAC address

Port 0/1

k Active Lin Port 0/5

fnC0026mp

mac-address-table station-move refresh-arp configured at time of NIC teaming

MAC Move Optimization

MAC Move Optimization is supported only on platform:

Station-move detection takes 5000ms because this is the interval at which the detection algorithm runs. On the E-Series, you can reduce detection time to as little as 500ms using the command mac-address-table station-move threshold time-interval (though at the expense of CPU resources). is the number of times a station move must be detected in a single interval in order to trigger a system log message. For example, if you configure mac-address-table station-move threshold 2 time-interval 5000, and 4 station moves occur in 5000ms, then two log messages are generated.
threshold

Microsoft Clustering
Microsoft Clustering is supported on the E, S55, and S60 platforms:

Layer 2 | 417

www.dell.com | support.dell.com

Microsoft Clustering allows multiple servers using Microsoft Windows to be represented by one MAC address and IP address in order to provide transparent failover or balancing. FTOS does not recognize server clusters by default; it must be configured to do so.

Default Behavior
When an ARP request is sent to a server cluster, either the active server or all of the servers send a reply, depending on the cluster configuration. If the active server sends a reply, the Dell Networking switch learns the active servers MAC address. If all servers reply, the switch registers only the last received ARP reply, and the switch learns one servers actual MAC address (Figure 20-4); the virtual MAC address is never learned. Since the virtual MAC address is never learned, traffic is forwarded to only one server rather than the entire cluster, and failover and balancing are not preserved (Figure 20-5).
Figure 20-4. Server Clustering: Multiple ARP Replies
Source MAC: MACS1 Destination MAC: MACClient Type: 0x0806 Source IP: IPS1 Source MAC: MACCluster Pad CRC

Server1:

IPS1 MACS1

Ethernet Frame Header

ARP Reply Client

La
Server2: IPS2 MACS2 Microsoft Server Cluster: IPCluster MACCluster Server3: IPS3 MACS3

st A

VLAN 1

RP

Re

ply

Server4:

IPS4 MACS4 fnC0027mp

Figure 20-5. Server Clustering: Failover and Balancing Not Preserved

Server1: IPS1 MACS1 VLAN 1 Server2: IPS2 MACS2 Client

Microsoft Server Cluster: IPCluster MACCluster

Server3:

IPS3 MACS3

Data

Server4:

IPS4 MACS4 fnC0028mp

418

Layer 2

Configuring the Switch for Microsoft Server Clustering

To preserve failover and balancing, the Dell Networking switch must learn the clusters virtual MAC address, and it must forward traffic destined for the server cluster out all member ports in the VLAN connected to the cluster. To ensure that this happens, you must configure the command vlan-flooding on the Dell Networking switch at the time that the Microsoft cluster is configured (Figure 20-6). As shown in Figure 20-6, the server MAC address is given in the Ethernet frame header of the ARP reply, while the virtual MAC address representing the cluster is given in the payload. The vlan-flooding command directs the system to discover that there are different MAC addresses in an ARP reply and associate the virtual MAC address with the VLAN connected to the cluster. Then, all traffic destined for the cluster is flooded out of all member ports. Since all of the servers in the cluster receive traffic, failover and balancing are preserved.
Figure 20-6. Server Cluster: Failover and Balancing Preserved with the vlan-flooding Command
Source MAC: MACS1 Destination MAC: MACClient Type: 0x0806 Source IP: IPS1 Source MAC: MACCluster Pad CRC

Server1:

IPS1 MACS1

Ethernet Frame Header VLAN 1

ARP Reply Client

Server2: IPS2 MACS2 Microsoft Server Cluster: IPCluster MACCluster Server3: IPS3 MACS3 Data

Server4:

IPS4 MACS4

vlan-flooding configured

fnC0029mp

Enable and Disable VLAN Flooding

ARP entries already resolved through the VLAN are deleted when the feature is enabled. This ensures that ARP entries across the VLAN are consistent. All ARP entries learned after the feature is enabled are deleted when the feature is disabled, and RP2 triggers ARP resolution. The feature is disabled with the command no vlan-flooding. When a port is added to the VLAN, the port automatically receives traffic if the feature is enabled. Old ARP entries are not deleted or updated. When a member port is deleted, its ARP entries are also deleted from the CAM. Port channels in the VLAN also receive traffic. There is no impact on the configuration from saving the configuration. The feature is not reflected in the output of the show arp command but is reflected in the output of the command show ipf fib.

Layer 2 | 419

www.dell.com | support.dell.com

The ARP entries exist in the secondary RPM CAM, so failover has no effect on the feature.

Configuring Redundant Pairs

Configuring Redundant Pairs is supported only on platform:

ecs

Networks that employ switches that do not support Spanning Tree (STP)for example, networks with Digital Subscriber Line Access Mutiplexers (DSLAM)cannot have redundant links between switches because they create switching loops (Figure 20-7). The Redundant Pairs feature enables you to create redundant links in networks that do not use STP by configuring backup interfaces for the interfaces on either side of the primary link.
Note: For details on STP, see Chapter 38, Spanning Tree Protocol, on page 723.

Assign a backup interface to an interface using the command switchport backup. The backup interface remains in down state until the primary fails, at which point it transitions to up state. If the primary interface fails, and later comes up, it becomes the backup interface for the redundant pair. FTOS supports Gigabit and 10-Gigabit interfaces as backup interfaces. You must apply all other configurations to each interface in the redundant pair such that their configurations are identical, so that transition to the backup interface in the event of a failure is transparent to rest of the network.
Figure 20-7. Configuring Redundant Layer 2 Pairs without Spanning Tree
Redundant links create a switching loop. Without STP broadcast storms occurs.

Use backup interfaces to create redundant links in networks without STP

FTOS(conf-if-gi-3/41)#switchport FTOS(conf-if-gi-3/41)#switchport backup gi 3/42 FTOS(conf-if-gi-3/41)#no shutdown FTOS(conf-if-gi-4/31)#switchport FTOS(conf-if-gi-4/31)#no shutdown

R3 3/41
Backup Link

R4 4/31

3/42
FTOS(conf-if-gi-3/42)#switchport FTOS(conf-if-gi-3/42)#no shutdown

4/32
FTOS(conf-if-gi-4/32)#switchport FTOS(conf-if-gi-4/32)#no shutdown

fnC0067mp

Important Points about Configuring Redundant Pairs

You may not configure any interface to be a backup for more than one interface, no interface can have more than one backup, and a backup interface may not have a backup interface. Neither the active nor the backup interface may be a member of a LAG. The active and standby do not have to be of the same type (1G, 10G, etc).

420

Layer 2

You may not enable any Layer 2 protocol on any interface of a redundant pair or to ports connected to them.

In Figure 20-8, interface 3/41 is a backup interface for 3/42, and 3/42 is in the down state, as shown in message Message 1. If 3/41 fails, 3/42 transitions to the up state, which makes the backup link active. A message similar to Message 1 appears whenever you configure a backup port.
Message 1 Configuring a Backup Layer 2 Port
02:28:04: %RPM0-P:CP %IFMGR-5-L2BKUP_WARN: Do not run any Layer2 protocols on Gi 3/41 and Gi 3/42 02:28:04: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 3/42 02:28:04: %RPM0-P:CP %IFMGR-5-STATE_ACT_STBY: Changed interface state to standby: Gi 3/42

Figure 20-8. CLI for Configuring Redundant Layer 2 Pairs without Spanning Tree
FTOS(conf-if-range-gi-3/41-42)#switchport backup interface GigabitEthernet 3/42 FTOS(conf-if-range-gi-3/41-42)#show config ! interface GigabitEthernet 3/41 no ip address switchport switchport backup interface GigabitEthernet 3/42 no shutdown ! interface GigabitEthernet 3/42 no ip address switchport no shutdown FTOS(conf-if-range-gi-3/41-42)# FTOS(conf-if-range-gi-3/41-42)#do show ip int brief | find 3/41 GigabitEthernet 3/41 unassigned YES Manual up up GigabitEthernet 3/42 unassigned NO Manual up down [output omitted] FTOS(conf-if-range-gi-3/41-42)#interface gig 3/41 FTOS(conf-if-gi-3/41)#shutdown 00:24:53: %RPM0-P:CP %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Gi 3/41 FTOS(conf-if-gi-3/41)#00:24:55: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 3/41 00:24:55: %RPM0-P:CP %IFMGR-5-INACTIVE: Changed Vlan interface state to inactive: Vl 1 00:24:55: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Gi 3/42 00:24:55: %RPM0-P:CP %IFMGR-5-ACTIVE: Changed Vlan interface state to active: Vl 1 00:24:55: %RPM0-P:CP %IFMGR-5-STATE_STBY_ACT: Changed interface state from standby to active: Gi 3/42 FTOS(conf-if-gi-3/41)#do show ip int brief | find 3/41 GigabitEthernet 3/41 unassigned NO Manual administratively down down GigabitEthernet 3/42 unassigned YES Manual up up [output omitted]

Layer 2 | 421

www.dell.com | support.dell.com

Restricting Layer 2 Flooding

Restricting Layer 2 Flooding is supported only on platform:

When Layer 2 multicast traffic must be forwarded on a VLAN that has multiple ports with different speeds on the same port-pipe, forwarding is limited to the speed of the slowest port. Restricted Layer 2 Flooding prevents slower ports from lowering the throughput of multicast traffic on faster ports by restricting flooding to ports with a speed equal to or above a link speed you specify. For example, if a VLAN that has an (auto-negotiated) 100M port and a 1G port on the same port-pipe, and you enable Restricted Layer 2 Flooding with a minimum speed of 1G, multicast traffic is only flooded on the 1G port. Enable Restricted Layer 2 Flooding using the command restrict-flooding from INTERFACE VLAN mode. In combination with restrict-flooding, you can use the command mac-flood-list from CONFIGURATION mode, without the min-speed option, to allow some specific multicast traffic (identified using a MAC address range you specify) to be flooded on all ports regardless of the restrict-flooding configuration. Conversely, if you want all multicast traffic to be flooded on all ports, but some specific traffic to be restricted, use mac-flood-list with the min-speed option, but without restrict-flooding configured. This configuration restricts flooding only for traffic with destination multicast MAC addresses within the multicast MAC address range you specify. In Figure 20-9, flooding of unknown multicast traffic is restricted to 1G ports on VLAN100 using the command restrict-flooding. However, the command mac-flood-list allows traffic with MAC addresses 01:01:e8:00:00:00 to 01:01:e8:ff:ff:ff to be flooded on all ports regardless of link speed.
Figure 20-9. Restricting Layer 2 Multicast Flooding over Low Speed Ports
FTOS(conf)#$1:01:e8:00:00:00 ff:ff:ff:00:00:00 vlan 100-200,300 FTOS#show run | find mac-flood-list mac-flood-list 01:01:e8:00:00:00 ff:ff:ff:00:00:00 vlan 100-200,300 [output omitted] FTOS(conf)#interface vlan 100 FTOS(conf-if-vl-100)#restrict-flooding multicast min-speed 1000 FTOS(conf-if-vl-100)#show config ! interface Vlan 100 restrict-flooding multicast min-speed 1000 no shutdown FTOS(conf-if-vl-100)#

Far-end Failure Detection

Far-end Failure Detection is supported only on platform:

Far-end Failure Detection (FEFD) is a protocol that senses remote data link errors in a network. It responds by sending a unidirectional report that triggers an echoed response after a specified time interval.

422

Layer 2

Figure 20-10.

Configuring Far-end Failure Detection

FTOS(conf-if-gi-4/0)#show config ! interface GigabitEthernet 4/0 no ip address switchport fefd FTOS(conf-if-gi-1/0)#show config no shutdown ! interface GigabitEthernet 1/0 no ip address switchport fefd R2 no shutdown Keep-alive

R1

Interval

2w0d4h : FEFD packet sent via interface Gi 1/0 Sender state -- Bi-directional Sender info -- Mgmt Mac(00:01:e8:14:89:25), Slot-Port(Gi 1/0) Peer info -- Mgmt Mac (00:01:e8:14:89:25), Slot-Port(Gi 4/0) Sender hold time -- 3 (second) R1 Echo R2

2w0d4h : FEFD packet sent via interface Gi 4/0 Sender state -- Bi-directional Sender info -- Mgmt Mac(00:01:e8:14:89:25), Slot-Port(Gi 4/0) Peer info -- Mgmt Mac (00:01:e8:14:89:25), Slot-Port(Gi 1/0) Sender hold time -- 3 (second) Layer2 001

The report consists of several packets in SNAP format that are sent to the nearest known MAC address. In the event of a far-end failure, the device stops receiving frames, and after the specified time interval assumes that the far-end is not available. The connecting line protocol is brought down so that upper layer protocols can detect the neighbor unavailability faster.

FEFD state changes

FEFD enabled systems (comprised of one or more interfaces) will automatically switch between four different modes: Idle, Unknown, Bi-directional, and Err-disabled. 1. An interface on which FEFD is not configured is in Idle state. 2. Once FEFD is enabled on an interface, it transitions to the Unknown state, and sends an FEFD packet to the remote end of the link. 3. When the local interface receives the echoed packet from the remote end, the local interface transitions to the Bi-directional state. 4. If the FEFD enabled system is configured to use FEFD in Normal mode and neighboring echoes are not received after three intervals, (each interval can be set between 3 and 300 seconds by the user) the state changes to unknown. 5. If the FEFD system has been set to Aggressive mode and neighboring echoes are not received after three intervals, the state changes to Err-disabled. All interfaces in the Err-disabled state must be manually reset using the fefd reset [interface] command in EXEC privilege mode (it can be done globally or one interface at a time) before the FEFD enabled system can become operational again.

Layer 2 | 423

www.dell.com | support.dell.com

Table 20-1.

State Changes When Configuring FEFD Local Admin Status Down Up Up Up Up Up Up Up Local Protocol Status Down Down Up Up Down Down Down Down Remote Admin Status Up Up Up Up Up Up Up Up Remote Protocol Status Down Down Up Up Down Down Down Down

Local Event Shutdown Shutdown FEFD enable FEFD enable FEFD + FEFD disable FEFD + FEFD disable Link Failure Link Failure

Mode Normal

Local State Admin Shutdown

Remote State Unknown Err-disabled Bi-directional Bi-directional Unknown Err-disabled Unknown Err-disabled

Aggressive Admin Shutdown Normal Aggressive Normal Aggressive Normal Aggressive Bi-directional Bi-directional Locally disabled Locally disabled Unknown Err-disabled

Important Points to Remember

FEFD enabled ports are subject to an 8 to 10 second delay during an RPM failover before becoming operational. FEFD can be enabled globally or on a per interface basis, interface FEFD configurations override global FEFD configurations. FTOS supports FEFD on physical Ethernet interfaces only, excluding the management interface.

Configuring FEFD
You can configure FEFD for all interfaces from CONFIGURATION mode, or on individual interfaces from INTERFACE mode.

Enable FEFD Globally

To enable FEFD globally on all interfaces enter the command fefd-global in CONFIGURATION mode. Report interval frequency and mode adjustments can be made by supplementing this command as well.
Step 1 Task Setup two or more connected interfaces for Layer 2 or Layer 3 use Activate the necessary ports administratively Command Syntax
ip address ip address, switchport no shutdown

Command Mode INTERFACE

INTERFACE

424

Layer 2

Step 3

Task Enable fefd globally

Command Syntax
fefd {interval | mode}

Command Mode CONFIGURATION

Entering the show fefd command in EXEC privilege mode displays information about the state of each interface.
Figure 20-11. Show FEFD global outputs

FTOS#show fefd FEFD is globally 'ON', interval is 3 seconds, mode is 'Normal'. INTERFACE Gi Gi Gi Gi 1/0 1/1 1/2 1/3 MODE Normal Normal Normal Normal INTERVAL (second) 3 3 3 3 STATE Bi-directional Admin Shutdown Admin Shutdown Admin Shutdown

FTOS#show run fefd ! fefd-global mode normal fefd-global interval 3

Enable FEFD on an Interface

Entering the command fefd in INTERFACE mode enables FEFD on a per interface basis. To change the FEFD mode, supplement the fefd command in INTERFACE mode by entering the command fefd [mode {aggressive | normal}]. To disable FEFD protocol on one interface, enter the command fefd disable in INTERFACE mode. Disabling an interface will shut down all protocols working on that interfaces connected line, and will not delete your previous FEFD configuration which can be enabled again at any time.
Step 1 Task Setup two or more connected interfaces for Layer 2 or Layer 3 use Activate the necessary ports administratively Enable FEFD on each interface Command Syntax
ip address ip address, switchport no shutdown

Command Mode INTERFACE

INTERFACE

fefd {disable | interval | mode}

INTERFACE

Layer 2 | 425

www.dell.com | support.dell.com

Figure 20-12.

FEFD enabled interface configuration

FTOS(conf-if-gi-1/0)#show config ! interface GigabitEthernet 1/0 no ip address switchport fefd mode normal no shutdown FTOS(conf-if-gi-1/0)#do show fefd | grep 1/0 Gi 1/0 Normal 3 Unknown

Debugging FEFD
By entering the command debug fefd events in EXEC privilege mode, output is displayed whenever events occur that initiate or disrupt an FEFD enabled connection.
Figure 20-13. Debug FEFD events display

FTOS#debug fefd events FTOS#config FTOS(conf)#int gi 1/0 FTOS(conf-if-gi-1/0)#shutdown 2w1d22h: %RPM0-P:CP %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Gi 1/0 FTOS(conf-if-gi-1/0)#2w1d22h : FEFD state on Gi 1/0 changed from ANY to Unknown 2w1d22h: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 1/0 2w1d22h: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 4/0 2w1d22h: %RPM0-P:CP %IFMGR-5-INACTIVE: Changed Vlan interface state to inactive: Vl 1 2w1d22h : FEFD state on Gi 4/0 changed from Bi-directional to Unknown

Entering the command debug fefd packets in EXEC privilege mode will provide output for each packet transmission over the FEFD enabled connection.
Figure 20-14. Debug FEFD packets display

FTOS#debug fefd packets FTOS#2w1d22h : FEFD packet sent via interface Gi 1/0 Sender state -- Bi-directional Sender info -- Mgmt Mac(00:01:e8:14:89:25), Slot-Port(Gi 1/0) Peer info -- Mgmt Mac (00:01:e8:14:89:25), Slot-Port(Gi 4/0) Sender hold time -- 3 (second) 2w1d22h : FEFD packet received on interface Gi 4/0 Sender state -- Bi-directional Sender info -- Mgmt Mac(00:01:e8:14:89:25), Slot-Port(Gi 1/0) Peer info -- Mgmt Mac (00:01:e8:14:89:25), Slot-Port(Gi 4/0) Sender hold time -- 3 (second)

426

Layer 2

During an RPM Failover

In the event that an RPM failover occurs, FEFD will become operationally down on all enabled ports for approximately 8-10 seconds before automatically becoming operational again.
Figure 20-15. FEFD state change during an RPM failover

02-05-2009 12:40:38 Local7.Debug 10.16.151.12 Feb 5 07:06:09: %RPM1-S:CP %RAM-6-FAILOVER_REQ: RPM failover request from active peer: User request. 02-05-2009 12:40:38 Local7.Debug 10.16.151.12 Feb 5 07:06:19: %RPM1-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Gi 0/45 02-05-2009 12:40:38 Local7.Debug 10.16.151.12 Feb 5 07:06:19: %RPM1-P:CP %FEFD-5-FEFD-BIDIRECTION-LINK-DETECTED: Interface Gi 0/45 has bidirectional link with its peer

Layer 2 | 427

428 www.dell.com | support.dell.com

Layer 2

21
Link Layer Discovery Protocol
Link Layer Discovery Protocol is supported only on platforms: This chapter contains the following sections: 802.1AB (LLDP) Overview TIA-1057 (LLDP-MED) Overview Configuring LLDP

ces

802.1AB (LLDP) Overview

Link Layer Discovery Protocol (LLDP)defined by IEEE 802.1ABis a protocol that enables a LAN device to advertise its configuration and receive configuration information from adjacent LLDP-enabled LAN infrastructure devices. The collected information is stored in a management information base (MIB) on each device, and is accessible via SNMP.

Protocol Data Units

Configuration information is exchanged in the form of Type, Length, Value (TLV) segments. Figure 21-1 shows the Chassis ID TLV. TypeThe kind of information included in the TLV LengthThe value, in octets, of the TLV after the Length field ValueThe configuration information that the agent is advertising
Type, Length, Value (TLV) Segment

Figure 21-1.
TLV Header

TLV Type (1-127)

TLV Length

Value

7 bits

9 bits

0-511 octets

Chassis ID Sub-type

Chassis ID fnC0057mp

1 octet

1- 255 octets

Link Layer Discovery Protocol | 429

www.dell.com | support.dell.com

TLVs are encapsulated in a frame called an LLDP Data Unit (LLDPDU) (Figure 21-2), which is transmitted from one LLDP-enabled device to its LLDP-enabled neighbors. LLDP is a one-way protocol. LLDP-enabled devices (LLDP agents) can transmit and/or receive advertisements, but they cannot solicit and do not respond to advertisements. There are five types of TLVs. All types are mandatory in the construction of an LLDPDU except Optional TLVs. The inclusion of individual Optional TLVs is user configurable.
Table 21-1. Type 0 1 2 3 TLV End of LLDPDU Chassis ID Port ID Time to Live Optional Type, Length, Value (TLV) Types Description Marks the end of an LLDPDU An administratively assigned name that identifies the LLDP agent An administratively assigned name that identifies a port through which TLVs are sent and received A value that tells the receiving agent how long the information contained in the TLV Value field is valid Includes sub-types of TLVs that advertise specific configuration information. These sub-types are Management TLVs, IEEE 802.1, IEEE 802.3, and TIA-1057 Organizationally Specific TLVs.

Figure 21-2. LLDPDU Frame

Preamble Start Frame Delimiter Destination MAC (01:80:C2:00:00:0E) Source MAC Ethernet Type (0x88CC) LLDPDU Padding FCS

TLV 1 Chassis ID

TLV 2 Port ID

TLV 3 Port Description

TLV 4 System Name

TLV 6 TLV 7 TLV 5 System Description System Capabilities Management Addr

TLV 127 Organizationally Specific

TLV 0 End of LLDPDU

fnC0047mp

Optional TLVs
FTOS supports the following optional TLVs: Management TLVs IEEE 802.1 and 802.3 Organizationally Specific TLVs TIA-1057 Organizationally Specific TLVs

430

Link Layer Discovery Protocol

Management TLVs
A Management TLV is an Optional TLVs sub-type. This kind of TLV contains essential management information about the sender. The five types are described in Table 21-2.

Organizationally Specific TLVs

Organizationally specific TLVs can be defined by a professional organization or a vendor. They have two mandatory fields (Figure 21-3) in addition to the basic TLV fields (Figure 21-1): Organizationally Unique Identifier (OUI)a unique number assigned by the IEEE to an organization or vendor. OUI Sub-typeThese sub-types indicate the kind of information in the following data field. The sub-types are determined by the owner of the OUI.

Figure 21-3. Organizationally Specific TLV

TLV Type (127) TLV Length Organizationally Unique ID (OUI) 3 octets Organizationally Defined Sub-type Organizationally Specific Data fnC0052mp 7 bits 9 bits 1 octet 0 - 507 octets

IEEE Organizationally Specific TLVs

Eight TLV types have been defined by the IEEE 802.1 and 802.3 working groups (Table 21-2) as a basic part of LLDP; the IEEE OUI is 00-80-C2. You can configure the Dell Networking system to advertise any or all of these TLVs.
Table 21-2. Type TLV Optional TLVs 4 5 6 7 Port description System name System description System capabilities A user-defined alphanumeric string that describes the port. FTOS does not currently support this TLV. A user-defined alphanumeric string that identifies the system. A user-defined alphanumeric string that describes the system Identifies the chassis as one or more of the following: repeater, bridge, WLAN Access Point, Router, Telephone, DOCSIS cable device, end station only, or other Indicates the network address of the management interface. FTOS does not currently support this TLV. Optional TLV Types Description

Management address

IEEE 802.1 Organizationally Specific TLVs 127 Port-VLAN ID On Dell Networking systems, indicates the untagged VLAN to which a port belongs

Link Layer Discovery Protocol | 431

www.dell.com | support.dell.com

Table 21-2. Type TLV 127

Optional TLV Types Description On Dell Networking systems, indicates the tagged VLAN to which a port belongs (and the untagged VLAN to which a port belongs if the port is in hybrid mode) Indicates the user-defined alphanumeric string that identifies the VLAN. This TLV is supported on C-Series only. Indicates the protocols that the port can process. FTOS does not currently support this TLV.

Port and Protocol VLAN ID

127 127

VLAN Name Protocol Identity

IEEE 802.3 Organizationally Specific TLVs 127 MAC/PHY Configuration/Status Indicates the capability and current setting of the duplex status and bit rate, and whether the current settings are the result of auto-negotiation. This TLV is not available in the FTOS implementation of LLDP, but is available and mandatory (non-configurable) in the LLDP-MED implementation. Dell Networking supports LLDP-MED protocol, which recommends that Power via MDI TLV be not implemented, and therefore Dell Networking implements Extended Power via MDI TLV only. Indicates whether the link is capable of being aggregated, whether it is currently in a LAG, and the port identification of the LAG. FTOS does not currently support this TLV. Indicates the maximum frame size capability of the MAC and PHY

127

Power via MDI

127

Link Aggregation

127

Maximum Frame Size

TIA-1057 (LLDP-MED) Overview

Link Layer Discovery ProtocolMedia Endpoint Discovery (LLDP-MED)as defined by ANSI/ TIA-1057 provides additional organizationally specific TLVs so that endpoint devices and network connectivity devices can advertise their characteristics and configuration information; the OUI for the Telecommunications Industry Association (TIA) is 00-12-BB. LLDP-MED Endpoint Deviceany device that is on an IEEE 802 LAN network edge can communicate using IP and uses the LLDP-MED framework. LLDP-MED Network Connectivity Deviceany device that provides access to an IEEE 802 LAN to an LLDP-MED endpoint device and supports IEEE 802.1AB (LLDP) and TIA-1057 (LLDP-MED). The Dell Networking system is an LLDP-MED network connectivity device.

With regard to connected endpoint devices, LLDP-MED provides network connectivity devices with the ability to: manage inventory manage Power over Ethernet (PoE) identify physical location identify network policy

LLDP-MED is designed for, but not limited to, VoIP endpoints.

432

Link Layer Discovery Protocol

TIA Organizationally Specific TLVs

The Dell Networking system is an LLDP-MED Network Connectivity Device (Device Type 4). Network connectivity devices are responsible for: transmitting an LLDP-MED capabilities TLV to endpoint devices storing the information that endpoint devices advertise

Table 21-3 describes the five types of TIA-1057 Organizationally Specific TLVs.
Table 21-3. Type 127 TIA-1057 (LLDP-MED) Organizationally Specific TLVs TLV LLDP-MED Capabilities Description Indicates: whether the transmitting device supports LLDP-MED what LLDP-MED TLVs it supports LLDP device class Indicates the application type, VLAN ID, Layer 2 Priority, and DSCP value Indicates the physical location of the device expressed in one of three possible formats: Coordinate Based LCI Civic Address LCI Emergency Call Services ELIN Indicates power requirements, priority, and power status Implementation of this set of TLVs is optional in LLDP-MED devices. None or all TLVs must be supported. FTOS does not currently support these TLVs. Indicates the hardware revision of the LLDP-MED device Indicates the firmware revision of the LLDP-MED device Indicates the software revision of the LLDP-MED device Indicates the device serial number of the LLDP-MED device Indicates the manufacturer of the LLDP-MED device Indicates the model of the LLDP-MED device Indicates a user specified device number to manage inventory

Sub-type 1

127 127

2 3

Network Policy Location Identification

127

Extended Power via MDI

Inventory Management TLVs 127 127 127 127 127 127 127 127 5 6 7 8 9 10 11 12-255 Inventory - Hardware Revision Inventory - Firmware Revision Inventory - Software Revision Inventory - Serial Number Inventory - Manufacturer Name Inventory - Model Name Inventory - Asset ID Reserved

Link Layer Discovery Protocol | 433

www.dell.com | support.dell.com

LLDP-MED Capabilities TLV

The LLDP-MED Capabilities TLV communicates the types of TLVs that the endpoint device and the network connectivity device support. LLDP-MED network connectivity devices must transmit the Network Policies TLV. The value of the LLDP-MED Capabilities field in the TLV is a 2 octet bitmap (Figure 21-4), each bit represents an LLDP-MED capability (Table 21-4). The possible values of the LLDP-MED Device Type is listed in Table 21-5. The Dell Networking system is a Network Connectivity device, which is Type 4.

When you enable LLDP-MED in FTOS (using the command advertise med) the system begins transmitting this TLV.
Figure 21-4. LLDP-MED Capabilities TLV
TLV Type (127) TLV Length (7) Organizationally Organizationally Unique ID Defined Sub-type (00-12-BB) (1) 3 octets 1 octet LLDP-MED Capabilites (00000000 00001111) 2 octets LLDP-MED Device Type (4) fnC0053mp 7 bits 9 bits 1 octet

Table 21-4. Bit Position 0 1 2 3 4 5 6-15 Table 21-5. Value 0 1 2 3 4 5-255

FTOS LLDP-MED Capabilities TLV LLDP-MED Capabilities Network Policy Location Identification Extended Power via MDI-PSE Extended Power via MDI-PD Inventory reserved LLDP-MED Device Types Device Type Type Not Defined Endpoint Class 1 Endpoint Class 2 Endpoint Class 3 Network Connectivity Reserved FTOS Support Yes Yes Yes Yes No No No

434

Link Layer Discovery Protocol

LLDP-MED Network Policies TLV

A network policy in the context of LLDP-MED is a devices VLAN configuration and associated Layer 2 and Layer 3 configurations, specifically: VLAN ID VLAN tagged or untagged status Layer 2 priority DSCP value

The application type is a represented by an integer (the Type integer in Table 21-6), which indicates a device function for which a unique network policy is defined. An individual LLDP-MED Network Policy TLV is generated for each application type that you specify with the FTOS CLI (Advertising TLVs).
Note: With regard to Table 21-6, signaling is a series of control packets that are exchanged between an endpoint device and a network connectivity device to establish and maintain a connection. These signal packets might require a different network policy than the media packets for which a connection is made. In this case, configure the signaling application.

Table 21-6. Type 0 1 2 3

Network Policy Applications Description Specify this application type for dedicated IP telephony handsets and other appliances supporting interactive voice services. Specify this application type only if voice control packets use a separate network policy than voice data. Specify this application type for a separate limited voice service for guest users with their own IP telephony handsets and other appliances supporting interactive voice services. Specify this application type only if guest voice control packets use a separate network policy than voice data. Softphone is a computer program that enables IP telephony on a computer, rather than using a phone. Specify this application type for this type of endpoint device. Specify this application type for dedicated video conferencing and other similar appliances supporting real-time interactive video. Specify this application type for broadcast or multicast based video content distribution and other similar applications supporting streaming video services. This does not include video applications relying on TCP with buffering. Specify this application type only if video control packets use a separate network policy than video data.

Application Reserved Voice Voice Signaling Guest Voice

4 5 6 7

Guest Voice Signaling Softphone Voice Video Conferencing Streaming Video

Video Signaling

9-255 Reserved

Link Layer Discovery Protocol | 435

www.dell.com | support.dell.com

Figure 21-5. LLDP-MED Policies TLV

TLV Type (127) TLV Length (8) Organizationally Organizationally Unique ID Defined Sub-type (00-12-BB) (2) 3 octets 1 octet Application Type (0-255) 1 octet U T X (0) VLAN ID (0-4095) 12 bits L2 Priority (0-7) 3 bits DSCP Value (0-63) 6 bits

7 bits

9 bits

3 bits

Extended Power via MDI TLV

The Extended Power via MDI TLV enables advanced PoE management between LLDP-MED endpoints and network connectivity devices. Advertise the Extended Power via MDI on all ports that are connected to an 802.3af powered, LLDP-MED endpoint device. Power Typethere are two possible power types: Power Sourcing Entity (PSE) or Power Device (PD). The Dell Networking system is a PSE, which corresponds to a value of 0, based on the TIA-1057 specification. Power Sourcethere are two possible power sources: Primary and Backup. The Dell Networking system is a Primary Power Source, which corresponds to a value of 1, based on the TIA-1057 specification. Power Prioritythere are three possible priorities: Low, High, and Critical. On Dell Networking systems, the default power priority is High, which corresponds to a value of 2 based on the TIA-1057 specification. You can configure a different power priority through the CLI, Dell Networking also honors the power priority value sent by the powered device. However, the CLI configuration takes precedence. Power ValueDell Networking advertises the maximum amount of power that can be supplied on the port. By default it is 15.4W, which corresponds to a Power Value of 130, based on the TIA-1057 specification. You can advertise a different Power Value using the max-milliwatts option with the power inline auto | static command. Dell Networking also honors the power value (power requirement) sent by the powered device when the port is configured for power inline auto.

Figure 21-6. Extended Power via MDI TLV

TLV Type (127) TLV Length (7) Organizationally Organizationally Unique ID Defined Sub-type (00-12-BB) (4) 3 octets 1 octet Power Type (0) Power Source (1) Power Priority (2) Power Value (130) fnC0056mp 7 bits 9 bits 2 bits 2 bits 4 bits 2 octets

Configuring LLDP
Configuring LLDP is a two-step process: 1. Enable LLDP globally. See page 438. 2. Advertise TLVs out of an interface. See page 438.

Related Configuration Tasks

Viewing the LLDP Configuration

436

Link Layer Discovery Protocol

Viewing Information Advertised by Adjacent LLDP Agents Configuring LLDPDU Intervals Configuring Transmit and Receive Mode Configuring a Time to Live Debugging LLDP

Important Points to Remember

LLDP is disabled by default. Dell Networking systems support up to 8 neighbors per interface. Dell Networking systems support a maximum of 8000 total neighbors per system. If the number of interfaces multiplied by 8 exceeds the maximum, the system will not configure more than 8000. INTERFACE level configurations override all CONFIGURATION level configurations. LLDP is not hitless.

LLDP Compatibility
Spanning Tree and Force10 Ring Protocol blocked ports allow LLDPDUs. 802.1X controlled ports do not allow LLDPDUs until the connected device is authenticated.

CONFIGURATION versus INTERFACE Configurations

All LLDP configuration commands are available in PROTOCOL LLDP mode, which is a sub-mode of CONFIGURATION mode and INTERFACE mode. Configurations made at CONFIGURATION level are global, that is, they affect all interfaces on the system. Configurations made at INTERFACE level affect only the specific interface, and they override CONFIGURATION level configurations.

Link Layer Discovery Protocol | 437

www.dell.com | support.dell.com

Figure 21-7. Configuration and Interface mode LLDP Commands

R1(conf)#protocol lldp R1(conf-lldp)#? advertise Advertise TLVs disable Disable LLDP protocol globally end Exit from configuration mode exit Exit from LLDP configuration mode hello LLDP hello configuration mode LLDP mode configuration (default = rx and tx) multiplier LLDP multiplier configuration no Negate a command or set its defaults show Show LLDP configuration R1(conf-lldp)#exit R1(conf)#interface gigabitethernet 1/31 R1(conf-if-gi-1/31)#protocol lldp R1(conf-if-gi-1/31-lldp)#? advertise Advertise TLVs disable Disable LLDP protocol on this interface end Exit from configuration mode exit Exit from LLDP configuration mode hello LLDP hello configuration mode LLDP mode configuration (default = rx and tx) multiplier LLDP multiplier configuration no Negate a command or set its defaults show Show LLDP configuration R1(conf-if-gi-1/31-lldp)#

Enabling LLDP
LLDP is disabled by default. LLDP can be enabled and disabled globally or per interface. If LLDP is enabled globally, all up interfaces send periodic LLDPDUs. To enable LLDP:
Step 1 2 Task Enter Protocol LLDP mode. Enable LLDP. Command
protocol lldp no disable

Command Mode CONFIGURATION or INTERFACE PROTOCOL LLDP

Disabling and Undoing LLDP

Disable LLDP globally or for an interface using the command disable. Undo an LLDP configuration by preceding the relevant command with the keyword no.

Advertising TLVs
You can configure the system to advertise TLVs out of all interfaces or out of specific interfaces. If you configure the system globally, all interfaces will send LLDPDUs with the specified TLVs.

438

Link Layer Discovery Protocol

If you configure an interface, only the interface will send LLDPDUs with the specified TLVs.

If LLDP is configured both globally and at interface level, the interface level configuration overrides the global configuration. To advertise TLVs:
Command Mode CONFIGURATI ON or INTERFACE PROTOCOL LLDP

Step 1

Task Enter LLDP mode.

Command
protocol lldp

Advertise one or more TLVs. Include the keyword for each TLV you want to advertise. For management TLVs: system-capabilities,
system-description

advertise {management-tlv | dot1-tlv | dot3-tlv | med}

For 802.1 TLVs: port-protocol-vlan-id,

port-vlan-id, vlan-name

For 802.3 TLVs: max-frame-size For TIA-1057 TLVs:

guest-voice guest-voice-signaling location-identification power-via-mdi softphone-voice streaming-video video-conferencing video-signaling voice voice-signaling

Note: vlan-name is supported on C-Series and S-Series only.

In Figure 21-8, LLDP is enabled globally. R1 and R2 are transmitting periodic LLDPDUs that contain management, 802.1, and 802.3 TLVs.
Figure 21-8. Configuring LLDP
R2(conf)#protocol lldp R2(conf-lldp)#no disable R2(conf-lldp)#advertise management-tlv system-capabilities system-description R2(conf-lldp)#ad dot1-tlv vlan-name R2(conf-lldp)#max-frame-size R1(conf)#protocol lldp R1(conf-lldp)#no disable R1(conf-lldp)#advertise management-tlv system-capabilities system-description R1(conf-lldp)#ad dot1-tlv vlan-name R1(conf-lldp)#max-frame-size

R2
2/11
LLDPDU

R1
1/21
fnC0074mp

R2(conf)#int gig 2/11 R2(conf-if-gi-2/11)# switchport R2(conf-if-gi-2/11)#no shut R1(conf)#int gig 1/21 R1(conf-if-gi-1/21)# switchport R1(conf-if-gi-1/21)#no shut

Link Layer Discovery Protocol | 439

www.dell.com | support.dell.com

Viewing the LLDP Configuration

Display the LLDP configuration using the command show config in either CONFIGURATION or INTERFACE mode, as shown in Figure 21-9 and Figure 21-10, respectively
Figure 21-9. Viewing LLDP Global Configurations
R1(conf)#protocol lldp R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description hello 10 no disable R1(conf-lldp)#

Figure 21-10.

Viewing LLDP Interface Configurations

R1(conf-lldp)#exit R1(conf)#interface gigabitethernet 1/31 R1(conf-if-gi-1/31)#show config ! interface GigabitEthernet 1/31 no ip address switchport no shutdown R1(conf-if-gi-1/31)#protocol lldp R1(conf-if-gi-1/31-lldp)#show config ! protocol lldp R1(conf-if-gi-1/31-lldp)#

Viewing Information Advertised by Adjacent LLDP Agents

Display brief information about adjacent devices using the command show lldp neighbors, as shown in Figure 21-11. Display all of the information that neighbors are advertising using the command show lldp neighbors detail, as shown in Figure 21-12.
Figure 21-11. Viewing Brief Information Advertised by Adjacent LLDP Agents

R1(conf-if-gi-1/31-lldp)#end R1(conf-if-gi-1/31)#do show lldp neighbors Loc PortID Rem Host Name Rem Port Id Rem Chassis Id ------------------------------------------------------------------------Gi 1/21 Gi 1/31 GigabitEthernet 2/11 GigabitEthernet 3/11 00:01:e8:06:95:3e 00:01:e8:09:c2:4a

440

Link Layer Discovery Protocol

Figure 21-12.

Viewing All Information Advertised by Adjacent LLDP Agent

R1#show lldp neighbors detail ======================================================================== Local Interface Gi 1/21 has 1 neighbor Total Frames Out: 6547 Total Frames In: 4136 Total Neighbor information Age outs: 0 Total Frames Discarded: 0 Total In Error Frames: 0 Total Unrecognized TLVs: 0 Total TLVs Discarded: 0 Next packet will be sent after 7 seconds The neighbors are given below: ----------------------------------------------------------------------Remote Chassis ID Subtype: Mac address (4) Remote Chassis ID: 00:01:e8:06:95:3e Remote Port Subtype: Interface name (5) Remote Port ID: GigabitEthernet 2/11 Local Port ID: GigabitEthernet 1/21 Locally assigned remote Neighbor Index: 4 Remote TTL: 120 Information valid for next 120 seconds Time since last information change of this neighbor: 01:50:16 Remote MTU: 1554 Remote System Desc: Dell Force10 Networks Real Time Operating System Software Dell Force10 Operating System Version: 1.0. Dell Force10 Application Software Version: 8.3.12.0 Copyright(c) 1999-2011 by Dell Inc. Build Time: Thu Mar 22 20:27:34 PDT 2012 Existing System Capabilities: Repeater Bridge Router Enabled System Capabilities: Repeater Bridge Router Remote Port Vlan ID: 1 Port and Protocol Vlan ID: 1, Capability: Supported, Status: Enabled --------------------------------------------------------------------------========================================================================

Configuring LLDPDU Intervals

LLDPDUs are transmitted periodically; the default interval is 30 seconds. You can configure a non-default transmit intervalat CONFIGURATION level or INTERFACE levelusing the command hello (Figure 21-13).

Link Layer Discovery Protocol | 441

www.dell.com | support.dell.com

Figure 21-13.

Configuring LLDPDU Transmit and Receive Mode

R1(conf)#protocol lldp R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#mode ? rx Rx only tx Tx only R1(conf-lldp)#mode tx R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description mode tx no disable R1(conf-lldp)#no mode R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#

Configuring Transmit and Receive Mode

Once LLDP is enabled, Dell Networking systems transmit and receive LLDPDUs by default. You can configure the systemat CONFIGURATION level or INTERFACE levelto transmit only by executing the command mode tx, or receive only by executing the command mode rx. Return to the default with the no mode command (Figure 21-14).

442

Link Layer Discovery Protocol

Figure 21-14.

Configuring LLDPDU Transmit and Receive Mode

R1(conf)#protocol lldp R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#mode ? rx Rx only tx Tx only R1(conf-lldp)#mode tx R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description mode tx no disable R1(conf-lldp)#no mode R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#

Configuring a Time to Live

The information received from a neighbor expires after a specific amount of time (measured in seconds) called a Time to Live (TTL). The TTL is the product of the LLDPDU transmit interval (hello) and an integer called a multiplier. The default multiplier is 4, which results in a default TTL of 120 seconds. Adjust the TTL valueat CONFIGURATION level or INTERFACE levelusing the command multiplier. Return to the default multiplier value using the command no multiplier (Figure 21-15).

Link Layer Discovery Protocol | 443

www.dell.com | support.dell.com

Figure 21-15.

Configuring LLDPDU Time to Live

R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#multiplier ? <2-10> Multiplier (default=4) R1(conf-lldp)#multiplier 5 R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description multiplier 5 no disable R1(conf-lldp)#no multiplier R1(conf-lldp)#show config ! protocol lldp advertise dot1-tlv port-protocol-vlan-id port-vlan-id advertise dot3-tlv max-frame-size advertise management-tlv system-capabilities system-description no disable R1(conf-lldp)#

Debugging LLDP
The command debug lldp enables you to view the TLVs that your system is sending and receiving. Use the debug lldp brief command to view a readable version of the TLVs. Use the debug lldp detail command to view a readable version of the TLVs plus a hexadecimal version of the entire LLDPDU.

444

Link Layer Discovery Protocol

Figure 21-16.

debug lldp detailLLDPDU Packet Dissection

FTOS# debug lldp interface gigabitethernet 1/2 packet detail tx FTOS#1w1d19h : Transmit timer blew off for local interface Gi 1/2 1w1d19h : Forming LLDP pkt to send out of interface Gi 1/2 1w1d19h : TLV: Chassis ID, Len: 7, Subtype: Mac address (4), Value: 00:01:e8:0d:b6:d6 1w1d19h : TLV: Port ID, Len: 20, Subtype: Interface name (5), Value: GigabitEthernet 1/2 1w1d19h : TLV: TTL, Len: 2, Value: 120 1w1d19h : TLV: SYS_DESC, Len: 207, Value:Dell Force10 Networks Real Time Operating System Software. Dell Force10 Operating System Version: 1.0. Dell Force10 Application Software Version: 8.3.11.4. Copyright (c)1999-2011 Dell Inc. Time: Fri Oct 26 12:22:22 PDT 2007 1w1d19h : TLV: SYSTEM CAPAB, Len: 4, Value: Existing: Repeater Bridge Router, Enabled: Repeater Bridge Router 1w1d19h : TLV: ENDOFPDU, Len: 0 1w1d19h : Sending LLDP pkt out of Gi 1/2 of length 270 Source Address (LLDP Multicast) 1w1d19h : Packet dump: Dell Force10 System Chassis ID 1w1d19h : 01 80 c2 00 00 0e 00 01 e8 0d b7 3b 81 00 00 00 802.1Q Header 1w1d19h : 88 cc 02 07 04 00 01 e8 0d b6 d6 04 14 05 47 69 1w1d19h : 67 61 62 69 74 45 74 68 65 72 6e 65 74 20 31 2f 1w1d19h : 32 06 02 00 78 0c cf 46 6f 72 63 65 31 30 20 4e 1w1d19h : 65 74 77 6f 72 6b 73 20 52 65 61 6c 20 54 69 6d 1w1d19h : 65 20 4f 70 65 72 61 74 69 6e 67 20 53 79 73 74 1w1d19h : 65 6d 20 53 6f 66 74 77 61 72 65 2e 20 46 6f 72 1w1d19h : 63 65 31 30 20 4f 70 65 72 61 74 69 6e 67 20 53 1w1d19h : 79 73 74 65 6d 20 56 65 72 73 69 6f 6e 3a 20 31 1w1d19h : 2e 30 2e 20 46 6f 72 63 65 31 30 20 41 70 70 6c 1w1d19h : 69 63 61 74 69 6f 6e 20 53 6f 66 74 77 61 72 65 1w1d19h : 20 56 65 72 73 69 6f 6e 3a 20 45 5f 4d 41 49 4e 1w1d19h : 34 2e 37 2e 35 2e 32 37 36 2e 20 43 6f 70 79 72 1w1d19h : 69 67 68 74 20 28 63 29 20 31 39 39 39 2d 42 75 1w1d19h : 69 6c 64 20 54 69 6d 65 3a 20 46 72 69 20 4f 63 1w1d19h : 74 20 32 36 20 31 32 3a 32 32 3a 32 32 20 50 44 1w1d19h : 54 20 32 30 30 37 0e 04 00 16 00 16 00 00 1w1d19h : LLDP frame sent out successfully of Gi 1/2 1w1d19h : Started Transmit timer for Loc interface Gi 1/2 for time 30 sec fnC0051mp

Relevant Management Objects

FTOS supports all IEEE 802.1AB MIB objects. Table 21-7 lists the objects associated with received and transmitted TLVs. Table 21-8 lists the objects associated with the LLDP configuration on the local agent. Table 21-9 lists the objects associated with IEEE 802.1AB Organizationally Specific TLVs. Table 21-10 lists the objects associated with received and transmitted LLDP-MED TLVs.

Link Layer Discovery Protocol | 445

www.dell.com | support.dell.com

Table 21-7.

LLDP Configuration MIB Objects LLDP Variable adminStatus msgTxHold msgTxInterval rxInfoTTL txInfoTTL LLDP MIB Object lldpPortConfigAdminStatus Description Whether the local LLDP agent is enabled for transmit, receive, or both

MIB Object Category LLDP Configuration

lldpMessageTxHoldMultiplier Multiplier value lldpMessageTxInterval lldpRxInfoTTL lldpTxInfoTTL Transmit Interval value Time to Live for received TLVs Time to Live for transmitted TLVs

Basic TLV Selection

mibBasicTLVsTxEnable lldpPortConfigTLVsTxEnable Indicates which management TLVs are enabled for system ports mibMgmtAddrInstanceT lldpManAddrPortsTxEnable xEnable The management addresses defined for the system and and the ports through which they are enabled for transmission Total number of times that a neighbors information is deleted on the local system due to an rxInfoTTL timer expiration Total number of LLDP frames received then discarded Total number of LLDP frames received on a port with errors Total number of LLDP frames received through the port Total number of LLDP frames transmitted through the port

LLDP Statistics

statsAgeoutsTotal

lldpStatsRxPortAgeoutsTotal

statsFramesDiscardedTot lldpStatsRxPortFramesDiscar al dedTotal statsFramesInErrorsTotal lldpStatsRxPortFramesErrors statsFramesInTotal statsFramesOutTotal lldpStatsRxPortFramesTotal lldpStatsTxPortFramesTotal

statsTLVsDiscardedTotal lldpStatsRxPortTLVsDiscarde Total number of TLVs received then dTotal discarded statsTLVsUnrecognizedT lldpStatsRxPortTLVsUnrecog Total number of all TLVs the local otal nizedTotal agent does not recognize

446

Link Layer Discovery Protocol

Table 21-8. TLV Type 1

LLDP System MIB Objects TLV Name Chassis ID TLV Variable chassis ID subtype System Local Remote chassid ID Local Remote LLDP MIB Object lldpLocChassisIdSubtype lldpRemChassisIdSubtype lldpLocChassisId lldpRemChassisId lldpLocPortIdSubtype lldpRemPortIdSubtype lldpLocPortId lldpRemPortId lldpLocPortDesc lldpRemPortDesc lldpLocSysName lldpRemSysName lldpLocSysDesc lldpRemSysDesc lldpLocSysCapSupported lldpRemSysCapSupported lldpLocSysCapEnabled lldpRemSysCapEnabled lldpLocManAddrLen lldpRemManAddrLen lldpLocManAddrSubtype lldpRemManAddrSubtype lldpLocManAddr lldpRemManAddr lldpLocManAddrIfSubtype lldpRemManAddrIfSubtype lldpLocManAddrIfId lldpRemManAddrIfId lldpLocManAddrOID lldpRemManAddrOID

Port ID

port subtype

Local Remote

port ID

Local Remote

Port Description

port description

Local Remote

System Name

system name

Local Remote

System Description

system description

Local Remote

System Capabilities

system capabilities

Local Remote

Management Address

enabled capabilities

Local Remote

management address length

Local Remote

management address subtype

Local Remote

management address

Local Remote

interface numbering subtype

Local Remote

interface number

Local Remote

OID

Local Remote

Link Layer Discovery Protocol | 447

www.dell.com | support.dell.com

Table 21-9. TLV Type 127

LLDP 802.1 Organizationally Specific TLV MIB Objects TLV Name Port-VLAN ID TLV Variable PVID System Local Remote LLDP MIB Object lldpXdot1LocPortVlanId lldpXdot1RemPortVlanId lldpXdot1LocProtoVlanSupported lldpXdot1RemProtoVlanSupported lldpXdot1LocProtoVlanEnabled lldpXdot1RemProtoVlanEnabled lldpXdot1LocProtoVlanId lldpXdot1RemProtoVlanId lldpXdot1LocVlanId lldpXdot1RemVlanId lldpXdot1LocVlanName lldpXdot1RemVlanName lldpXdot1LocVlanName lldpXdot1RemVlanName

127

Port and Protocol VLAN ID

port and protocol VLAN supported Local Remote port and protocol VLAN enabled Local Remote PPVID Local Remote

127

VLAN Name

VID

Local Remote

VLAN name length

Local Remote

VLAN name

Local Remote

448

Link Layer Discovery Protocol

Table 21-10.

LLDP-MED System MIB Objects TLV Variable LLDP-MED Capabilities System Local LLDP-MED MIB Object lldpXMedPortCapSupported lldpXMedPortConfigTLVsTx Enable lldpXMedRemCapSupported, lldpXMedRemConfigTLVsTx Enable lldpXMedLocDeviceClass lldpXMedRemDeviceClass lldpXMedLocMediaPolicyApp Type lldpXMedRemMediaPolicyAp pType lldpXMedLocMediaPolicyUnk nown lldpXMedLocMediaPolicyUnk nown lldpXMedLocMediaPolicyTag ged lldpXMedLocMediaPolicyTag ged lldpXMedLocMediaPolicyVla nID lldpXMedRemMediaPolicyVl anID lldpXMedLocMediaPolicyPrio rity lldpXMedRemMediaPolicyPri ority lldpXMedLocMediaPolicyDsc p lldpXMedRemMediaPolicyDs cp lldpXMedLocLocationSubtype lldpXMedRemLocationSubtyp e lldpXMedLocLocationInfo lldpXMedRemLocationInfo

TLV Sub-Type TLV Name 1 LLDP-MED Capabilities

Remote

LLDP-MED Class Type

Local Remote

Network Policy

Application Type

Local Remote

Unknown Policy Flag

Local Remote

Tagged Flag

Local Remote

VLAN ID

Local Remote

L2 Priority

Local Remote

DSCP Value

Local Remote

Location Identifier

Location Data Format

Local Remote

Location ID Data

Local Remote

Link Layer Discovery Protocol | 449

www.dell.com | support.dell.com

Table 21-10.

LLDP-MED System MIB Objects (continued) TLV Variable Power Device Type System Local Remote Power Source Local LLDP-MED MIB Object lldpXMedLocXPoEDeviceTyp e lldpXMedRemXPoEDeviceTy pe lldpXMedLocXPoEPSEPower Source, lldpXMedLocXPoEPDPowerS ource lldpXMedRemXPoEPSEPowe rSource, lldpXMedRemXPoEPDPower Source lldpXMedLocXPoEPDPowerP riority, lldpXMedLocXPoEPSEPortP DPriority lldpXMedRemXPoEPSEPowe rPriority, lldpXMedRemXPoEPDPower Priority lldpXMedLocXPoEPSEPortPo werAv, lldpXMedLocXPoEPDPower Req lldpXMedRemXPoEPSEPowe rAv, lldpXMedRemXPoEPDPower Req

TLV Sub-Type TLV Name 4 Extended Power via MDI

Remote

Power Priority

Local

Remote

Power Value

Local

Remote

450

Link Layer Discovery Protocol

22
Multiple Spanning Tree Protocol
Multiple Spanning Tree Protocol is supported on platforms:

ces

Protocol Overview
Multiple Spanning Tree Protocol (MSTP)specified in IEEE 802.1Q-2003is an RSTP-based spanning tree variation that improves on PVST+. MSTP allows multiple spanning tree instances and allows you to map many VLANs to one spanning tree instance to reduce the total number of required instances. In contrast, PVST+ allows a spanning tree instance for each VLAN. This 1:1 approach is not suitable if you have many VLANs, because each spanning tree instance costs bandwidth and processing resources. In Figure 22-1, three VLANs are mapped to two Multiple Spanning Tree instances (MSTI). VLAN 100 traffic takes a different path than VLAN 200 and 300 traffic. The behavior in Figure 22-1 demonstrates how you can use MSTP to achieve load balancing.
Figure 22-1. MSTP with Three VLANs Mapped to Two Spanning Tree Instances

R1

MSTI 1: VLAN 100 MSTI 2: VLAN 200, VLAN 300

MSTI 1 root R2

1/21 1/31
Forwarding Blocking

2/11 2/31

3/11 3/21

MSTI 2 root

R3

Multiple Spanning Tree Protocol | 451

www.dell.com | support.dell.com

FTOS supports three other variations of Spanning Tree, as shown in Table 22-1.
Table 22-1. FTOS Supported Spanning Tree Protocols IEEE Specification 802.1d 802.1w 802.1s Third Party

Dell Networking Term Spanning Tree Protocol Rapid Spanning Tree Protocol Multiple Spanning Tree Protocol Per-VLAN Spanning Tree Plus

Implementation Information
The FTOS MSTP implementation is based on IEEE 802.1Q-2003, and interoperates only with bridges that also use this standard implementation. MSTP is compatible with STP and RSTP. FTOS supports only one MSTP region. When you enable MSTP, all ports in Layer 2 mode participate in MSTP. On the C-Series and S-Series, you can configure 64 MSTIs including the default instance 0 (CIST).

Configure Multiple Spanning Tree Protocol

Configuring Multiple Spanning Tree is a four-step process: 1. Configure interfaces for Layer 2. See page 725. 2. Place the interfaces in VLANs. 3. Enable Multiple Spanning Tree Protocol. See page 453. 4. Create Multiple Spanning Tree Instances, and map VLANs to them. See page 453.

Related Configuration Tasks

Create Multiple Spanning Tree Instances Add and Remove Interfaces Influence MSTP Root Selection Interoperate with Non-FTOS Bridges Modify Global Parameters Modify Interface Parameters Configure an EdgePort Flush MAC Addresses after a Topology Change Debugging and Verifying MSTP Configuration

452

Multiple Spanning Tree Protocol

Preventing Network Disruptions with BPDU Guard SNMP Traps for Root Elections and Topology Changes Configuring Spanning Trees as Hitless

Enable Multiple Spanning Tree Globally

MSTP is not enabled by default. To enable MSTP:
Step 1 2 Task Enter PROTOCOL MSTP mode. Enable MSTP. Command Syntax
protocol spanning-tree mstp no disable

Command Mode CONFIGURATION PROTOCOL MSTP

Verify that MSTP is enabled using the show config command from PROTOCOL MSTP mode, as shown in Figure 22-2.
Figure 22-2. Verifying MSTP is Enabled
FTOS(conf)#protocol spanning-tree mstp FTOS(config-mstp)#show config ! protocol spanning-tree mstp no disable FTOS#

When you enable MSTP, all physical, VLAN, and port-channel interfaces that are enabled and in Layer 2 mode are automatically part of the MSTI 0. Within an MSTI, only one path from any bridge to any other bridge is enabled. Bridges block a redundant path by disabling one of the link ports.

Add and Remove Interfaces

To add an interface to the MSTP topology, configure it for Layer 2 and add it to a VLAN. If you previously disabled MSTP on the interface using the command no spanning-tree 0, re-enable it using the command spanning-tree 0. Remove an interface from the MSTP topology using the command no spanning-tree 0 command. See also Removing an Interface from the Spanning Tree Group for BPDU Filtering behavior.

Create Multiple Spanning Tree Instances

A single MSTI provides no more benefit than RSTP. To take full advantage of MSTP you must create multiple MSTIs and map VLANs to them.

Multiple Spanning Tree Protocol | 453

www.dell.com | support.dell.com

Create an MSTI using the command msti from PROTOCOL MSTP mode. Specify the keyword vlan followed by the VLANs that you want to participate in the MSTI, as shown in Figure 22-3.
Figure 22-3. Mapping VLANs to MSTI Instances
FTOS(conf)#protocol spanning-tree mstp FTOS(conf-mstp)#msti 1 vlan 100 FTOS(conf-mstp)#msti 2 vlan 200-300 FTOS(conf-mstp)#show config ! protocol spanning-tree mstp no disable MSTI 1 VLAN 100 MSTI 2 VLAN 200-300

All bridges in the MSTP region must have the same VLAN-to-instance mapping. View to which instance a VLAN is mapped using the command show spanning-tree mst vlan from EXEC Privilege mode, as shown in Figure 22-6. View the forwarding/discarding state of the ports participating in an MSTI using the command show spanning-tree msti from EXEC Privilege mode, as shown in Figure 22-4.
Figure 22-4. Viewing MSTP Port States
FTOS#show spanning-tree msti 1 MSTI 1 VLANs mapped 100 Root Identifier has priority 32768, Address 0001.e806.953e Root Bridge hello time 2, max age 20, forward delay 15, max hops 19 Bridge Identifier has priority 32768, Address 0001.e80d.b6d6 Configured hello time 2, max age 20, forward delay 15, max hops 20 Current root has priority 32768, Address 0001.e806.953e Number of topology changes 2, last change occured 1d2h ago on Gi 1/21 Port 374 (GigabitEthernet 1/21) is root Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.374 Designated root has priority 32768, address 0001.e806.953e Designated bridge has priority 32768, address 0001.e806.953e Designated port id is 128.374, designated path cost 20000 Number of transitions to forwarding state 1 BPDU (MRecords): sent 93671, received 46843 The port is not in the Edge port mode Port 384 (GigabitEthernet 1/31) is alternate Discarding Port path cost 20000, Port priority 128, Port Identifier 128.384 Designated root has priority 32768, address 0001.e806.953e Designated bridge has priority 32768, address 0001.e809.c24a Designated port id is 128.384, designated path cost 20000 Number of transitions to forwarding state 1 BPDU (MRecords): sent 39291, received 7547 The port is not in the Edge port mode

454

Multiple Spanning Tree Protocol

Influence MSTP Root Selection

MSTP determines the root bridge, but you can assign one bridge a lower priority to increase the probability that it will become the root bridge. To change the bridge priority:
Task Assign a number as the bridge priority. A lower number increases the probability that the bridge becomes the root bridge. Range: 0-61440, in increments of 4096 Default: 32768 Command Syntax
msti instance bridge-priority priority

Command Mode PROTOCOL MSTP

The simple configuration Figure 22-1 by default yields the same forwarding path for both MSTIs. Figure 22-5, shows how R3 is assigned bridge priority 0 for MSTI 2, which elects a different root bridge than MSTI 2. View the bridge priority using the command show config from PROTOCOL MSTP mode, also shown in Figure 22-5.
Figure 22-5. Changing the Bridge Priority
R3(conf-mstp)#msti 2 bridge-priority 0 1d2h51m: %RPM0-P:RP2 %SPANMGR-5-STP_ROOT_CHANGE: MSTP root changed for instance 2. My Bridge ID: 0:0001.e809.c24a Old Root: 32768:0001.e806.953e New Root: 0:0001.e809.c24a R3(conf-mstp)#show config ! protocol spanning-tree mstp no disable MSTI 1 VLAN 100 MSTI 2 VLAN 200,300 MSTI 2 bridge-priority 0

Interoperate with Non-FTOS Bridges

FTOS supports only one MSTP region. A region is a combination of three unique qualities: Name is a mnemonic string you assign to the region. The default region name on FTOS is null. Revision is a two-byte number. The default revision number on FTOS is 0. VLAN-to-instance mapping is the placement of a VLAN in an MSTI.

For a bridge to be in the same MSTP region as another, all three of these qualities must match exactly. The default values for name and revision will match on all Dell Networking FTOS equipment. If you have non-FTOS equipment that will participate in MSTP, ensure these values to match on all the equipment.
Note: Some non-FTOS equipment may implement a non-null default region name. SFTOS, for example, uses the Bridge ID, while others may use a MAC address.

Multiple Spanning Tree Protocol | 455

www.dell.com | support.dell.com

To change the region name or revision:

Task Change the region name. Change the region revision number. Range: 0 to 65535 Default: 0 Command Syntax
name name revision number

Command Mode PROTOCOL MSTP PROTOCOL MSTP

View the current region name and revision using the command show spanning-tree mst configuration from EXEC Privilege mode, as shown in Figure 22-6.
Figure 22-6. Viewing the MSTP Region Name and Revision
FTOS(conf-mstp)#name my-mstp-region FTOS(conf-mstp)#exit FTOS(conf)#do show spanning-tree mst config MST region name: my-mstp-region Revision: 0 MSTI VID 1 100 2 200-300

Modify Global Parameters

The root bridge sets the values for forward-delay, hello-time, max-age, and max-hops and overwrites the values set on other MSTP bridges. Forward-delay is the amount of time an interface waits in the Listening State and the Learning State before it transitions to the Forwarding State. Hello-time is the time interval in which the bridge sends MSTP Bridge Protocol Data Units (BPDUs). Max-age is the length of time the bridge maintains configuration information before it refreshes that information by recomputing the MST topology. Max-hops is the maximum number of hops a BPDU can travel before a receiving switch discards it.

Note: Dell Networking recommends that only experienced network administrators change MSTP parameters. Poorly planned modification of MSTP parameters can negatively impact network performance.

To change MSTP parameters, use the following commands on the root bridge:
Task Change the forward-delay parameter. Range: 4 to 30 Default: 15 seconds Command Syntax
forward-delay seconds

Command Mode PROTOCOL MSTP

456

Multiple Spanning Tree Protocol

Task Change the hello-time parameter. Note: With large configurations (especially those with more ports) Dell Networking recommends that you increase the hello-time. Range: 1 to 10 Default: 2 seconds Change the max-age parameter. Range: 6 to 40 Default: 20 seconds Change the max-hops parameter. Range: 1 to 40 Default: 20

Command Syntax
hello-time seconds

Command Mode PROTOCOL MSTP

max-age seconds

PROTOCOL MSTP

max-hops number

PROTOCOL MSTP

View the current values for MSTP parameters using the show running-config spanning-tree mstp command from EXEC privilege mode.
Figure 22-7. Viewing the Current Values for MSTP Parameters
FTOS(conf-mstp)#forward-delay 16 FTOS(conf-mstp)#exit FTOS(conf)#do show running-config spanning-tree mstp ! protocol spanning-tree mstp no disable name my-mstp-region MSTI 1 VLAN 100 MSTI 2 VLAN 200-300 forward-delay 16 MSTI 2 bridge-priority 4096 FTOS(conf)#

Modify Interface Parameters

You can adjust two interface parameters to increase or decrease the probability that a port becomes a forwarding port: Port cost is a value that is based on the interface type. The greater the port cost, the less likely the port will be selected to be a forwarding port. Port priority influences the likelihood that a port will be selected to be a forwarding port in case that several ports have the same port cost.

Multiple Spanning Tree Protocol | 457

www.dell.com | support.dell.com

Table 22-2 lists the default values for port cost by interface.
Table 22-2. Port Cost 100-Mb/s Ethernet interfaces 1-Gigabit Ethernet interfaces 10-Gigabit Ethernet interfaces Port Channel with 100 Mb/s Ethernet interfaces Port Channel with 1-Gigabit Ethernet interfaces Port Channel with 10-Gigabit Ethernet interfaces MSTP Default Port Cost Values Default Value 200000 20000 2000 180000 18000 1800

To change the port cost or priority of an interface:

Task Change the port cost of an interface. Range: 0 to 200000 Default: see Table 22-2. Change the port priority of an interface. Range: 0 to 240, in increments of 16 Default: 128 Command Syntax
spanning-tree msti number cost cost

Command Mode INTERFACE

spanning-tree msti number priority priority

INTERFACE

View the current values for these interface parameters using the command show config from INTERFACE mode. See Figure 22-8.

Configure an EdgePort
The EdgePort feature enables interfaces to begin forwarding traffic approximately 30 seconds sooner. In this mode an interface forwards frames by default until it receives a BPDU that indicates that it should behave otherwise; it does not go through the Learning and Listening states. The bpduguard shutdown-on-violation option causes the interface hardware to be shutdown when it receives a BPDU. When only bpduguard is implemented, although the interface is placed in an Error Disabled state when receiving the BPDU, the physical interface remains up and spanning-tree will drop packets in the hardware after a BPDU violation. BPDUs are dropped in the software after receiving the BPDU violation. This feature is the same as PortFast mode in Spanning Tree. Caution: Configure EdgePort only on links connecting to an end station. EdgePort can cause loops if it is enabled on an interface connected to a network.

458

Multiple Spanning Tree Protocol

To enable EdgePort on an interface, use the following command:

Task Enable EdgePort on an interface. Command Syntax
spanning-tree mstp edge-port [bpduguard | shutdown-on-violation]

Command Mode INTERFACE

Verify that EdgePort is enabled on a port using the command show config from the INTERFACE mode, as shown in Figure 22-8.
FTOS Behavior: Regarding bpduguard shutdown-on-violation behavior: 1If the interface to be shutdown is a port channel then all the member ports are disabled in the hardware. 2When a physical port is added to a port channel already in error disable state, the new member port will also be disabled in the hardware. 3When a physical port is removed from a port channel in error disable state, the error disabled state is cleared on this physical port (the physical port will be enabled in the hardware). 4The reset linecard command does not clear the error disabled state of the port or the hardware disabled state. The interface continues to be disables in the hardware. The error disabled state can be cleared with any of the following methods: Perform an shutdown command on the interface. Disable the shutdown-on-violation command on the interface ( no spanning-tree stp-id portfast [bpduguard | [shutdown-on-violation]] ). Disable spanning tree on the interface (no spanning-tree in INTERFACE mode). Disabling global spanning tree (no spanning-tree in CONFIGURATION mode). Figure 22-8. Configuring EdgePort
FTOS(conf-if-gi-3/41)#spanning-tree mstp edge-port FTOS(conf-if-gi-3/41)#show config ! interface GigabitEthernet 3/41 no ip address switchport spanning-tree mstp edge-port spanning-tree MSTI 1 priority 144 no shutdown FTOS(conf-if-gi-3/41)#

Flush MAC Addresses after a Topology Change

FTOS has an optimized MAC address flush mechanism for RSTP, MSTP, and PVST+ that flushes addresses only when necessary, which allows for faster convergence during topology changes. However, you may activate the flushing mechanism defined by 802.1Q-2003 using the command tc-flush-standard, which flushes MAC addresses upon every topology change notification. View the enable status of this feature using the command show running-config spanning-tree mstp from EXEC Privilege mode.

Multiple Spanning Tree Protocol | 459

www.dell.com | support.dell.com

MSTP Sample Configurations

The running-configurations in Figure 22-10, Figure 22-11, and Figure 22-11 support the topology shown in Figure 22-9. The configurations are from FTOS systems. An S50 system using SFTOS, configured as shown Figure 22-13, could be substituted for an FTOS router in this sample following topology and MSTP would function as designed.
Figure 22-9. MSTP with Three VLANs Mapped to Two Spanning Tree Instances

root R1
1/2 1/3
Blocking Forwarding

R2
2/1 2/3

3/1

3/2

R3

460

Multiple Spanning Tree Protocol

Figure 22-10.

Router 1 Running-configuration
Enable MSTP globally Set Region Name and Revision Map MSTP Instances to VLANs

protocol spanning-tree mstp no disable name Tahiti revision 123 MSTI 1 VLAN 100 MSTI 2 VLAN 200,300 ! interface GigabitEthernet 1/21 no ip address switchport no shutdown ! interface GigabitEthernet 1/31 no ip address switchport no shutdown ! interface Vlan 100 no ip address tagged GigabitEthernet 1/21,31 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 1/21,31 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 1/21,31 no shutdown

Assign Layer-2 interfaces to MSTP topology

Create VLANs mapped to MSTP Instances Tag interfaces to VLANs

Multiple Spanning Tree Protocol | 461

www.dell.com | support.dell.com

Figure 22-11.

Router 2 Running-configuration

protocol spanning-tree mstp no disable name Tahiti revision 123 MSTI 1 VLAN 100 MSTI 2 VLAN 200,300 ! interface GigabitEthernet 2/11 no ip address switchport no shutdown ! interface GigabitEthernet 2/31 no ip address switchport no shutdown ! interface Vlan 100 no ip address tagged GigabitEthernet 2/11,31 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 2/11,31 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 2/11,31 no shutdown

Enable MSTP globally Set Region Name and Revision Map MSTP Instances to VLANs

Assign Layer-2 interfaces to MSTP topology

Create VLANs mapped to MSTP Instances Tag interfaces to VLANs

462

Multiple Spanning Tree Protocol

Figure 22-12.

Router 3 Running-configuration

protocol spanning-tree mstp no disable name Tahiti revision 123 MSTI 1 VLAN 100 MSTI 2 VLAN 200,300 ! interface GigabitEthernet 3/11 no ip address switchport no shutdown ! interface GigabitEthernet 3/21 no ip address switchport no shutdown ! interface Vlan 100 no ip address tagged GigabitEthernet 3/11,21 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 3/11,21 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 3/11,21 no shutdown

Enable MSTP globally Set Region Name and Revision Map MSTP Instances to VLANs

Assign Layer-2 interfaces to MSTP topology

Create VLANs mapped to MSTP Instances Tag interfaces to VLANs

Multiple Spanning Tree Protocol | 463

www.dell.com | support.dell.com

Figure 22-13.

SFTOS Example Running-Configuration

configuration name Tahiti configuration revision 123 MSTi instance 1 MSTi vlan 1 100 MSTi instance 2 MSTi vlan 2 200 MSTi vlan 2 300

spanning-tree spanning-tree spanning-tree spanning-tree spanning-tree spanning-tree spanning-tree spanning-tree

Enable MSTP globally Set Region Name and Revision Map MSTP Instances to VLANs

interface 1/0/31 no shutdown spanning-tree port mode enable switchport protected 0 exit interface 1/0/32 no shutdown spanning-tree port mode enable switchport protected 0 exit interface vlan tagged 1/0/31 tagged 1/0/32 exit interface vlan tagged 1/0/31 tagged 1/0/32 exit interface vlan tagged 1/0/31 tagged 1/0/32 exit 100

Assign Layer-2 interfaces to MSTP topology

200 Create VLANs mapped to MSTP Instances Tag interfaces to VLANs

300

Debugging and Verifying MSTP Configuration

Display BPDUs using the command debug spanning-tree mstp bpdu from EXEC Privilege mode. Display MSTP-triggered topology change messages debug spanning-tree mstp events.

464

Multiple Spanning Tree Protocol

Figure 22-14.

Displaying BPDUs and Events

FTOS#debug spanning-tree mstp bpdu 1w1d17h : MSTP: Sending BPDU on Gi 1/31 : ProtId: 0, Ver: 3, Bpdu Type: MSTP, Flags 0x68 CIST Root Bridge Id: 32768:0001.e806.953e, Ext Path Cost: 20000 Regional Bridge Id: 32768:0001.e809.c24a, CIST Port Id: 128:384 Msg Age: 2, Max Age: 20, Hello: 2, Fwd Delay: 15, Ver1 Len: 0, Ver3 Len: 96 Name: my-mstp-region, Rev: 0, Int Root Path Cost: 20000 Rem Hops: 19, Bridge Id: 32768:0001.e80d.b6d6 E1200#1w1d17h : INST 1: Flags: 0x28, Reg Root: 32768:0001.e809.c24a, Int Root Co Brg/Port Prio: 32768/128, Rem Hops: 19 INST 2: Flags: 0x68, Reg Root: 4096:0001.e809.c24a, Int Root Cost: 20000 Brg/Port Prio: 32768/128, Rem Hops: 19 [output omitted] FTOS#debug spanning-tree mstp events 1w1d17h : MSTP: TC flag set in the incoming BPDU on port Gi 1/31 for instance 0 1w1d17h : MSTP: TC flag set in the incoming BPDU on port Gi 1/31 for instance 0 1w1d17h : MSTP: TC flag set in the incoming BPDU on port Gi 1/31 for instance 0

Examine your individual routers to ensure all the necessary parameters match. 1. Region Name 2. Region Version 3. VLAN to Instance mapping The show spanning-tree mst commands will show various portions of the MSTP configuration. To view the overall MSTP configuration on the router, use the show running-configuration spanning-tree mstp in the EXEC Privilege mode (output sample shown in Figure 22-15). Use the debug spanning-tree mstp bpdu command to monitor and verify that the MSTP configuration is connected and communicating as desired (output sample shown in Figure 22-16). Key items to look for in the debug report: MSTP flags indicate communication received from the same region. In Figure 22-16, the output shows that the MSTP routers are located in the same region. Does the debug log indicate that packets are coming from a Different Region (Figure 22-17)? If so, one of the key parameters is not matching. MSTP Region Name and Revision The configured name and revisions must be identical among all the routers. Is the Region name blank? That may mean that a name was configured on one router and but was not configured or was configured differently on another router (spelling and capitalization counts). MSTP Instances. Use the show commands to verify the VLAN to MSTP Instance mapping. Are there extra MSTP Instances in the Sending or Received logs? That may mean that an additional MSTP Instance was configured on one router but not the others.

Multiple Spanning Tree Protocol | 465

www.dell.com | support.dell.com

Figure 22-15.

Sample Output for show running-configuration spanning-tree mstp command

FTOS#show run spanning-tree mstp ! protocol spanning-tree mstp name Tahiti revision 123 MSTI 1 VLAN 100 MSTI 2 VLAN 200,300

Figure 22-16.

Displaying BPDUs and Events - Debug Log of Successful MSTP Configuration

FTOS#debug spanning-tree mstp bpdu MSTP debug bpdu is ON FTOS# 4w0d4h : MSTP: Sending BPDU on Gi 2/21 : ProtId: 0, Ver: 3, Bpdu Type: MSTP, Flags 0x6e CIST Root Bridge Id: 32768:0001.e806.953e, Ext Path Cost: 0 Regional Bridge Id: 32768:0001.e806.953e, CIST Port Id: 128:470 Msg Age: 0, Max Age: 20, Hello: 2, Fwd Delay: 15, Ver1 Len: 0, Ver3 Len: 96 Name: Tahiti, Rev: 123, Int Root Path Cost: 0 Rem Hops: 20, Bridge Id: 32768:0001.e806.953e 4w0d4h : INST 1: Flags: 0x6e, Reg Root: 32768:0001.e806.953e, Int Root Cost: 0 Brg/Port Prio: 32768/128, Rem Hops: 20 INST 2: Flags: 0x6e, Reg Root: 32768:0001.e806.953e, Int Root Cost: 0 Brg/Port Prio: 32768/128, Rem Hops: 20 4w0d4h : MSTP: Received BPDU on Gi 2/21 : Indicates MSTP ProtId: 0, Ver: 3, Bpdu Type: MSTP, Flags 0x78Same Region routers are in the CIST Root Bridge Id: 32768:0001.e806.953e, Ext Path Cost: 0 (single) region Regional Bridge Id: 32768:0001.e806.953e, CIST Port Id: 128:470 Msg Age: 0, Max Age: 20, Hello: 2, Fwd Delay: 15, Ver1 Len: 0, Ver3 Len: 96 Name: Tahiti, Rev: 123, Int Root Path Cost: 0 Rem Hops: 19, Bridge Id: 32768:0001.e8d5.cbbd 4w0d4h : INST 1: Flags: 0x78, Reg Root: 32768:0001.e806.953e, Int Root Cost: 0 Brg/Port Prio: 32768/128, Rem Hops: 19 INST 2: Flags: 0x78, Reg Root: 32768:0001.e806.953e, Int Root Cost: 0 Brg/Port Prio: 32768/128, Rem Hops: 19 MSTP Instance

MSTP Region name and revision

Figure 22-17.

Displaying BPDUs and Events - Debug Log of Unsuccessful MSTP Configuration

4w0d4h : MSTP: Received BPDU on Gi 2/21 : ProtId: 0, Ver: 3, Bpdu Type: MSTP, Flags 0x78Different Region CIST Root Bridge Id: 32768:0001.e806.953e, Ext Path Cost: 0 Regional Bridge Id: 32768:0001.e806.953e, CIST Port Id: 128:470 Msg Age: 0, Max Age: 20, Hello: 2, Fwd Delay: 15, Ver1 Len: 0, Ver Name: Tahiti, Rev: 123, Int Root Path Cost: 0 Rem Hops: 20, Bridge Id: 32768:0001.e8d5.cbbd 4w0d4h : INST 1: Flags: 0x70, Reg Root: 32768:0001.e8d5.cbbd, Int Brg/Port Prio: 32768/128, Rem Hops: 20 INST 2: Flags: 0x70, Reg Root: 32768:0001.e8d5.cbbd, Int Root Cost Brg/Port Prio: 32768/128, Rem Hops: 20

Indicates MSTP routers are in different regions and are not communicating with each other

466

Multiple Spanning Tree Protocol

23
Multicast Features
Multicast Features are supported on platforms: This chapter contains the following sections: Enable IP Multicast Multicast with ECMP First Packet Forwarding for Lossless Multicast Multicast Policies Multicast Traceroute Multicast Quality of Service Optimize the E-Series for Multicast Traffic Tune the Central Scheduler for Multicast

ces

FTOS supports the following multicast protocols: PIM Sparse-Mode PIM Source-Specific Mode Internet Group Management Protocol

Implementation Information
Multicast is not supported on secondary IP addresses.

Enable IP Multicast
Enable IP Multicast is supported on platforms

ces

Multicast Features | 467

www.dell.com | support.dell.com

Prior to enabling any multicast protocols, you must enable multicast routing.
Task Enable multicast routing. Command Syntax
ip multicast-routing

Command Mode CONFIGURATION

Multicast with ECMP

Dell Networking multicast uses Equal-cost Multi-path (ECMP) routing to load-balance multiple streams across equal cost links. When creating the shared-tree Protocol Independent Multicast (PIM) uses routes from all configured routing protocols to select the best route to the rendezvous point (RP). If there are multiple, equal-cost paths, the PIM selects the route with the least number of currently running multicast streams. If multiple routes have the same number of streams, PIM selects the first equal-cost route returned by the Route Table Manager (RTM). In Figure 23-1, the receiver joins three groups. The last-hop DR initially has two equal-cost routes to the RP with no streams, so it non-deterministically selects Route 1 for the Group 1 IGMP Join message. Route 1 then has one stream associated with it, so the last-hop DR sends the Group 2 Join by Route 2. It then non-deterministically selects Route 2 for the Group 3 Join since both routes already have one multicast stream.
Figure 23-1. Multicast with ECMP

te Rou

2
P Joi IGM
n: G3

Gig X

Gig Z

Gig B

IG M P J

2 oin: G

RP
IGMP Join:

Gig A Gig W
G1

Gig Y Source IGMP Group Table Group Address Interface Group 1 GigabitEthernet Y Group 2 GigabitEthernet X Group 3 GigabitEthernet X

Receiver IGMP Group Table Group Address Interface Group 1 GigabitEthernet A Group 2 GigabitEthernet A Group 3 GigabitEthernet A

Rou te 1

468

Multicast Features

Implementation Information
Because protocol control traffic in FTOS is redirected using the MAC address, and multicast control traffic and multicast data traffic might map to the same MAC address, FTOS might forward data traffic with certain MAC addresses to the CPU in addition to control traffic. As the upper five bits of an IP Multicast address are dropped in the translation, 32 different multicast group IDs all map to the same Ethernet address. For example, 224.0.0.5 is a well known IP address for OSPF that maps to the multicast MAC address 01:00:5e:00:00:05. However, 225.0.0.5, 226.0.0.5, etc., map to the same multicast MAC address. The Layer 2 FIB alone cannot differentiate multicast control traffic multicast data traffic with the same address, so if you use IP address 225.0.0.5 for data traffic, both the multicast data and OSPF control traffic match the same entry and are forwarded to the CPU. Therefore, do not use well-known protocol multicast addresses for data transmission, such as the ones below.
Protocol OSPF RIP NTP VRRP PIM-SM Ethernet Address

01:00:5e:00:00:05 01:00:5e:00:00:06 01:00:5e:00:00:09 01:00:5e:00:01:01 01:00:5e:00:00:12 01:00:5e:00:00:0d

The FTOS implementation of MTRACE is in accordance with IETF draft draft-fenner-traceroute-ipm. Multicast is not supported on secondary IP addresses. Egress L3 ACL is not applied to multicast data traffic if multicast routing is enabled.

First Packet Forwarding for Lossless Multicast

Beginning with FTOS version 7.8.1.0 for the E-Series TeraScale, version 8.2.1.0 for E-Series ExaScale, and version 8.3.1.0 on all other FTOS platforms, all initial multicast packets are forwarded to receivers to achieve lossless multicast. In previous versions, when the Dell Networking system is an RP, all initial packets are dropped until PIM creates an (S,G) entry. When the system is an RP and a Source DR, these initial packet drops represent a loss of native data, and when the system is an RP only, the initial packets drops represent a loss of register packets.

Multicast Features | 469

www.dell.com | support.dell.com

Both scenarios might be unacceptable depending on the multicast application. Beginning with the FTOS versions above, when the Dell Networking system is the RP, and has receivers for a group G, it forwards all initial multicast packets for the group based on the (*,G) entry rather than discarding them until the (S,G) entry is created, making Dell Networking systems suitable for applications sensitive to multicast packet loss.
Note: When a source begins sending traffic, the Source DR forwards the initial packets to the RP as encapsulated registered packets. These packets are forwarded via the soft path at a maximum rate of 70 packets/second. Incoming packets beyond this rate are dropped.

Multicast Policies
FTOS offers parallel Multicast features for IPv4 and IPv6. IPv4 Multicast Policies IPv6 Multicast Policies

IPv4 Multicast Policies

Limit the Number of Multicast Routes Prevent a Host from Joining a Group Rate Limit IGMP Join Requests Prevent a PIM Router from Forming an Adjacency Prevent a Source from Registering with the RP Prevent a PIM Router from Processing a Join

Limit the Number of Multicast Routes

Task Limit the total number of multicast routes on the system. Command Syntax
ip multicast-limit

Command Mode CONFIGURATION

Range: 1-50000 Default: 15000

When the limit is reached, FTOS does not process any IGMP or MLD joins to PIMthough it still processes leave messagesuntil the number of entries decreases below 95% of the limit. When the limit falls below 95% after hitting the maximum, the system begins relearning route entries through IGMP, MLD, and MSDP. If the limit is increased after it is reached, join subsequent join requests are accepted. In this case, you must increase the limit by at least 10% for IGMP and MLD to resume.

470

Multicast Features

If the limit is decreased after it is reached, FTOS does not clear the existing sessions. Entries are cleared upon a timeout (you may also clear entries using clear ip mroute).
Note: FTOS waits at least 30 seconds between stopping and starting IGMP join processing. You may experience this delay when manipulating the limit after it is reached.

When the multicast route limit is reached, FTOS displays Message 1.

Message 1 Multicast Route Limit Error
3w1d13h: %RPM0-P:RP2 %PIM-3-PIM_TIB_LIMIT: PIM TIB limit reached. No new routes will be learnt until TIB level falls below low watermark. 3w1d13h: %RPM0-P:RP2 %PIM-3-PIM_TIB_LIMIT: PIM TIB below low watermark. Route learning will begin.

Note: The IN-L3-McastFib CAM partition is used to store multicast routes and is a separate hardware limit that is exists per port-pipe. Any software-configured limit might be superseded by this hardware space limitation. The opposite is also true, the CAM partition might not be exhausted at the time the system-wide route limit set by the ip multicast-limit is reached.

Prevent a Host from Joining a Group

You can prevent a host from joining a particular group by blocking specific IGMP reports. Create an extended access list containing the permissible source-group pairs. Use the command ip igmp access-group access-list-name from INTERFACE mode to apply the access list.
Note: For rules in IGMP access lists, source is the multicast source, not the source of the IGMP packet. For IGMPv2, use the keyword any for source (as shown in Figure 23-2), since IGMPv2 hosts do not know in advance who the source is for the group in which they are interested.

FTOS Behavior: Do not enter the command ip igmp access-group before creating the access-list. If you do, upon entering your first deny rule, FTOS clears multicast routing table and re-learns all groups, even those not covered by the rules in the access-list, because there is an implicit deny all rule at the end of all access-lists. Therefore, configuring an IGMP join request filter in this order might result in data loss. If you must enter the command ip igmp access-group before creating the access-list, prevent FTOS from clearing the routing table by entering a permit any rule with high sequence number before you enter any other rules.

In Figure 23-2, VLAN 400 is configured with an access list to permit only IGMP reports for group 239.0.0.1. Though Receiver 2 sends a membership report for groups 239.0.0.1 and 239.0.0.2, a multicast routing table entry is created only for group 239.0.0.1. VLAN 300 has no access list limiting Receiver 1, so both IGMP reports are accepted, and two corresponding entries are created in the routing table.

Multicast Features | 471

472 www.dell.com | support.dell.com

interface GigabitEthernet 3/21 ip pim sparse-mode ip address 10.11.23.2/24 no shutdown

|
Source 2 10.11.1.2
interface GigabitEthernet 2/1 ip pim sparse-mode ip address 10.11.1.1/24 no shutdown interface GigabitEthernet 3/1 ip pim sparse-mode ip address 10.11.5.1/24 no shutdown interface GigabitEthernet 2/31 ip pim sparse-mode ip address 10.11.23.1/24 no shutdown

Figure 23-2. Preventing a Host from Joining a Group

Multicast Features
Source 1 10.11.5.2 R2 3/1 2/31 R3
interface GigabitEthernet 2/11 ip pim sparse-mode ip address 10.11.12.2/24 no shutdown 239.0.0.2 239.0.0.1

2/1 3/21 RP 2/11 3/11 ip multicast-routing ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 router rip network 10.0.0.0
interface GigabitEthernet 1/21 ip pim sparse-mode ip address 10.11.12.1/24 no shutdown

interface GigabitEthernet 3/11 ip pim sparse-mode ip address 10.11.13.2/24 no shutdown

R1(conf-if-vl-400)# do show ip pim tib

1/21 1/31
interface GigabitEthernet 1/31 ip pim sparse-mode ip address 10.11.13.1/24 no shutdown

R1

R1(conf )#do show run acl ! ip access-list extended igmpjoinfilR2G2 seq 5 permit ip any host 239.0.0.1

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

R1(conf-if-vl-300)# do show ip pim tib PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode interface Vlan 300 ip pim sparse-mode ip address 10.11.3.1/24 untagged GigabitEthernet 1/1 no shutdown

(*, 239.0.0.1), uptime 00:00:06, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:00:06/Never interface Vlan 400 ip pim sparse-mode ip address 10.11.4.1/24 untagged GigabitEthernet 1/2 ip igmp access-group igmpjoinfilR2G2 no shutdown ip igmp snooping enable

Receiver 1 10.11.3.2 Group: 239.0.0.1, 239.0.0.2

(*, 239.0.0.1), uptime 00:00:07, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:00:07/Never (*, 239.0.0.2), uptime 00:01:10, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:01:10/Never

Receiver 2 10.11.4.2 Group: 239.0.0.1, 239.0.0.2

Rate Limit IGMP Join Requests

If you expect a burst of IGMP Joins, protect the IGMP process from overload by limiting that rate at which new groups can be joined using the command ip igmp group-join-limit from INTERFACE mode. Hosts whose IGMP requests are denied will use the retry mechanism built-in to IGMP so that theyre membership is delayed rather than permanently denied. View the enable status of this feature using the command show ip igmp interface from EXEC Privilege mode.

Prevent a PIM Router from Forming an Adjacency

To prevent a router from participating in Protocol Independent Multicast (PIM) (for example, to configure stub multicast routing), use the ip pim neighbor-filter command from INTERFACE mode.

Prevent a Source from Registering with the RP

Use the command ip pim register-filter from CONFIGURATION mode to prevent a source from transmitting to a particular group. This command prevents the PIM source DR from sending register packets to RP for the specified multicast source and group; if the source DR never sends register packets to the RP, no hosts can ever discover the source and create an SPT to it. In Figure 23-3, Source 1 and Source 2 are both transmitting packets for groups 239.0.0.1 and 239.0.0.2. R3 has a PIM register filter that only permits packets destined for group 239.0.0.2. An entry is created for group 239.0.0.1 in the routing table, but no outgoing interfaces are listed. R2 has no filter, so it is allowed to forward both groups. As a result, Receiver 1 receives only one transmission, while Receiver 2 receives duplicate transmissions.

Multicast Features | 473

474 www.dell.com | support.dell.com

interface GigabitEthernet 3/21 ip pim sparse-mode ip address 10.11.23.2/24 no shutdown R3(conf )#do show ip pim tib

|
Source 2 10.11.1.2
interface GigabitEthernet 2/1 ip pim sparse-mode ip address 10.11.1.1/24 no shutdown interface GigabitEthernet 3/1 ip pim sparse-mode ip address 10.11.5.1/24 no shutdown interface GigabitEthernet 2/31 ip pim sparse-mode ip address 10.11.23.1/24 no shutdown

Multicast Features
Source 1 10.11.5.2
PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode (10.11.5.2, 239.0.0.1), uptime 00:05:21, expires 00:02:46, flags: FT Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 3/11 Forward/Sparse 00:00:18/00:03:12

R2(conf )#do show ip pim tib

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

(*, 239.0.0.1), uptime 00:03:00, expires 00:02:32, RP 10.11.12.2, flags: SF Incoming interface: Null, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 2/11 Forward/Sparse 00:03:00/00:02:32

R2 3/1 2/31 R3
interface GigabitEthernet 2/11 ip pim sparse-mode ip address 10.11.12.2/24 no shutdown

239.0.0.2 239.0.0.1

(10.11.1.2, 239.0.0.1), uptime 00:00:44, expires 00:02:56, flags: FT Incoming interface: GigabitEthernet 2/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 2/11 Forward/Sparse 00:00:13/00:03:17

2/1 3/21 2/11

interface GigabitEthernet 1/21 ip pim sparse-mode ip address 10.11.12.1/24 no shutdown

(10.11.5.2, 239.0.0.2), uptime 00:05:08, expires 00:03:26, flags: FT Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Outgoing interface list:

(10.11.5.2, 239.0.0.1), uptime 00:00:26, expires 00:03:06, flags: P Incoming interface: GigabitEthernet 2/31, RPF neighbor 10.11.23.2 Outgoing interface list:

3/11 ip multicast-routing ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 router rip network 10.0.0.0

interface GigabitEthernet 3/11 ip pim sparse-mode ip address 10.11.13.2/24 no shutdown

(*, 239.0.0.2), uptime 00:00:21, expires 00:03:09, RP 10.11.12.2, flags: SF Incoming interface: Null, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 2/11 Forward/Sparse 00:00:21/00:03:09

1/21 RP 1/31
interface GigabitEthernet 1/31 ip pim sparse-mode ip address 10.11.13.1/24 no shutdown

(10.11.1.2, 239.0.0.2), uptime 00:00:02, expires 00:03:28, flags: FT Incoming interface: GigabitEthernet 2/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 2/11 Forward/Sparse 00:00:21/00:03:09

R1

R3(conf )#do show run acl ! ip access-list extended regfilS1G2 seq 5 permit ip host 10.11.5.2 host 239.0.0.1 R3(conf )#do show run pim ! ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 ip pim register-filter regfilS1G2

R1(config-line-console)#do show ip pim tib

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode interface Vlan 400 ip pim sparse-mode ip address 10.11.4.1/24 untagged GigabitEthernet 1/2 no shutdown ip igmp snooping enable

interface Vlan 300 ip pim sparse-mode ip address 10.11.3.1/24 untagged GigabitEthernet 1/1 no shutdown

Figure 23-3. Preventing a Source from Transmitting to a Group

(*, 239.0.0.1), uptime 00:00:43, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:00:43/Never

Receiver 1 10.11.3.2 Group: 239.0.0.2

(10.11.1.2, 239.0.0.1), uptime 00:00:17, expires 00:03:17, flags: CT Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:00:43/Never

(10.11.5.2, 239.0.0.1), uptime 00:00:23, expires 00:03:17, flags: CT Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:00:43/Never

Receiver 2 10.11.4.2 Group: 239.0.0.1

(*, 239.0.0.2), uptime 00:00:40, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:00:40/Never

(10.11.5.2, 239.0.0.2), uptime 00:00:33, expires 00:03:07, flags: CT Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:00:40/Never

Prevent a PIM Router from Processing a Join

Permit or deny PIM Join/Prune messages on an interface using an extended IP access list. Use the command ip pim join-filter to prevent the PIM SM router from creating state based on multicast source and/ or group.

IPv6 Multicast Policies

IPv6 Multicast Policies is available only on platform:

Limit the Number of IPv6 Multicast Routes Prevent an IPv6 Neighbor from Forming an Adjacency Prevent an IPv6 Source from Registering with the RP Prevent an IPv6 PIM Router from Processing an IPv6 Join

Limit the Number of IPv6 Multicast Routes

You can limit the total number of IPv6 multicast routes on the system. The maximum number of multicast entries allowed on each line card is determined by the CAM profile. Multicast routes are stored in the IN-V6-McastFib CAM region, which has a fixed number of entries. Any limit configured via the CLI is superseded by this hardware limit. The opposite is also true; the CAM might not be exhausted at the time the CLI-configured route limit is reached.
Task Limit the total number of IPv6 multicast routes on the system. Command Syntax
ipv6 multicast-limit

Command Mode CONFIGURATION

Range: 1-50000 Default: 15000

Multicast Features | 475

www.dell.com | support.dell.com

Prevent an IPv6 Neighbor from Forming an Adjacency

Task Prevent a router from participating in PIM. Command Syntax
ipv6 pim neighbor-filter access-list

Command Mode CONFIGURATION

FTOS(conf)#ipv6 pim neighbor-filter NEIGH_ACL FTOS(conf)#ipv6 access-list NEIGH_ACL FTOS(conf-ipv6-acl)#show config ! ipv6 access-list NEIGH_ACL seq 5 deny ipv6 host fe80::201:e8ff:fe0a:5ad any seq 10 permit ipv6 any any FTOS(conf-ipv6-acl)#

Prevent an IPv6 Source from Registering with the RP

Task Configured on the source DR, prevent the source DR from sending register packets to the RP for specific sources and groups. Command Syntax
ipv6 pim register-filter access-list

Command Mode CONFIGURATION

FTOS(conf)#ipv6 pim register-filter REG-FIL_ACL FTOS(conf)#ipv6 access-list REG-FIL_ACL FTOS(conf-ipv6-acl)#deny ipv6 165:87:34::10/128 ff0e::225:1:2:0/112 FTOS(conf-ipv6-acl)#permit ipv6 any any FTOS(conf-ipv6-acl)#exit

Prevent an IPv6 PIM Router from Processing an IPv6 Join

Task Permit or deny PIM Join/Prune messages on an interface using an access list. This command prevents the PIM-SM router from creating state based on multicast source and/or group. Command Syntax
ipv6 pim join-filteraccess-list [in | out]

Command Mode INTERFACE

FTOS(conf)#ipv6 access-list JOIN-FIL_ACL FTOS(conf-ipv6-acl)#permit ipv6 165:87:34::0/112 ff0e::225:1:2:0/112 FTOS(conf-ipv6-acl)#permit ipv6 any ff0e::230:1:2:0/112 FTOS(conf-ipv6-acl)#permit ipv6 165:87:32::0/112 any FTOS(conf-ipv6-acl)#exit FTOS(conf)#interface gigabitethernet 0/84 FTOS(conf-if-gi-0/84)#ipv6 pim join-filter JOIN-FIL_ACL in FTOS(conf-if-gi-0/84)#ipv6 pim join-filter JOIN-FIL_ACL out

476

Multicast Features

Multicast Traceroute
Multicast Traceroute is supported only on platform:

MTRACE is an IGMP-based tool that prints that network path that a multicast packet takes from a source to a destination, for a particular group. FTOS has mtrace client and mtrace transmit functionality. MTRACE Clientan mtrace client transmits mtrace queries and prints out the details received responses. MTRACE Transitwhen a Dell Networking system is an intermediate router between the source and destination in an MTRACE query, FTOS computes the RPF neighbor for the source, fills in the request, and forwards the request to the RPF neighbor. While computing the RPF neighbor, static mroutes and mBGP routes are preferred over unicast routes. When a Dell Networking system is the last hop to the destination, FTOS sends a response to the query.
Command Syntax
mtrace multicast-source-address multicast-receiver-address multicast-group-address

Task Print the network path that a multicast packet takes from a multicast source to receiver, for a particular group.

Command Mode EXEC Privilege

Figure 23-4. Tracing a Multicast Route

FTOS#mtrace 10.11.5.2 10.11.3.2 239.0.0.1 Type Ctrl-C to abort. Mtrace from 10.11.5.2 to 10.11.3.2 via group 239.0.0.1 From source (?) to destination (?) Querying full reverse path... 0 10.11.3.2 -1 10.11.3.1 PIM Reached RP/Core [default] -2 10.11.5.2

Multicast Quality of Service

Multicast Quality of Service is supported only on platform:

The Quality of Service (QoS) features available for unicast traffic can be applied to multicast flows. The following QoS features are available: Policy-based QoSClassifying, rate policing, and marking ingress traffic WRED See also Allocate More Buffer Memory for Multicast WRED.

Multicast Features | 477

www.dell.com | support.dell.com

Optimize the E-Series for Multicast Traffic

Optimize the E-Series for Multicast Traffic is supported only on platform:

The default hardware settings for the E-series are for unicast applications like data centers and ISP networks. This means that the E-Series gives priority to unicast data forwarding rather than multicast data forwarding. For multicast intensive applications like trading, Dell Networking recommends reconfiguring some default settings. You may do one or more for the following to optimize the E-Series for your multicast application: Tune the Central Scheduler for Multicast Allocate More Buffer Memory for Multicast WRED Allocate More Bandwidth to Multicast using Egress WFQ

Allocate More Buffer Memory for Multicast WRED

Allocate more buffer memory to multicast WRED (Weighted Random Early Detection) for bursty multicast traffic that might temporarily become oversubscribed. For example, the example WRED profile in Figure 30-14 on page 592 allocates multicast traffic a minimum of 40 megabytes (out of 80 megabytes) of buffer memory and up to 60 megabytes.
Figure 23-5. Allocating More Bandwidth for Multicast WRED
FTOS(Conf)#queue egress multicast linecard all wred-profile Egress FTOS(conf)#wred-profile Egress FTOS(conf-wred)# threshold min 40960 max 61440

Allocate More Bandwidth to Multicast using Egress WFQ

Egress Weighted Fair Queuing (WFQ) determines per port the ratio of egress bandwidth allocated to multicast, replication, and unicast traffic. By default, FTOS provides 1/64 to multicast, 1/64 to replication, and 62/64 for unicast, which is shared between 8 unicast queues. Allocate more bandwidth for multicast using the command queue egress multicast linecard (from CONFIGURATION mode) with the keyword bandwidth-percent. For example, allocate 80% of egress bandwidth to multicast on all line cards using the command queue egress multicast linecard all bandwidth-percent 80.

Tune the Central Scheduler for Multicast

The Central Scheduler is responsible for scheduling unicast and multicast packets via the Terabit backplane. The default configuration of the Central Scheduler is optimized for network environments that forward primarily unicast traffic80% of the scheduler weight is for unicast traffic and 20% is for multicast traffic.

478

Multicast Features

FTOS provides the ability to adjust the scheduling weight for multicast traffic. For example, if the majority of your traffic is multicast, the default configuration might yield greater latency. In this case, allocate more backplane bandwidth for multicast using the command queue multicast bandwidth-percent from CONFIGURATION mode. View your configuration using the command show queue backplane multicast bandwidth-percentage.
Figure 23-6. Tuning the Central Scheduler for Multicast
FTOS#show queue backplane multicast bandwidth-percent Configured multicast bandwidth percentage is 80

Multicast Features | 479

480 www.dell.com | support.dell.com

Multicast Features

24
Open Shortest Path First (OSPFv2 and OSPFv3)
ces Open Shortest Path First version 3 (OSPF for IPv6) is supported on platforms c e
Open Shortest Path First version 2 (OSPF for IPv4) is supported on platforms This chapter is intended to provide a general description of OSPFv2 (OSPF for IPv4) and OSPFv3 (OSPF for IPv6) as supported in the Dell Networking Operating System (FTOS). It is not intended to provide a complete
Note: The fundamental mechanisms of OSPF (flooding, DR election, area support, SPF calculations, etc.) are the same between OSPFv2 and OSPFv3. Where there are differences between the two versions, they are identified and clarified. Except where identified, the information in this chapter applies to both protocol versions.

This chapter includes the following topics: Protocol Overview Implementing OSPF with FTOS Graceful Restart Fast Convergence (OSPFv2, IPv4 only) Multi-Process OSPF (OSPFv2, IPv4 only) RFC-2328 Compliant OSPF Flooding OSPF ACK Packing OSPF Adjacency with Cisco Routers Configuration Information Configuration Task List for OSPFv2 (OSPF for IPv4) Configuration Task List for OSPFv3 (OSPF for IPv6) Sample Configurations for OSPFv2

OSPF protocol standards are listed in the Appendix , Standards Compliance chapter.

Open Shortest Path First (OSPFv2 and OSPFv3) | 481

www.dell.com | support.dell.com

Protocol Overview
Open Shortest Path First (OSPF) routing is a link-state routing protocol that calls for the sending of Link-State Advertisements (LSAs) to all other routers within the same Autonomous System (AS) Areas. Information on attached interfaces, metrics used, and other variables is included in OSPF LSAs. As OSPF routers accumulate link-state information, they use the SPF algorithm (Shortest Path First algorithm) to calculate the shortest path to each node. OSPF routers initially exchange HELLO messages to set up adjacencies with neighbor routers. The HELLO process is used to establish adjacencies between routers of the AS. It is not required that every router within the Autonomous System areas establish adjacencies. If two routers on the same subnet agree to become neighbors through the HELLO process, they begin to exchange network topology information in the form of Link State Advertisements (LSAs). OSPFv3 runs on a per-link basis instead of on a per-IP-subnet basis. All neighbors on all link types are

identified by Router ID (RID). In OSPFv2 neighbors on broadcast and NBMA links are identified by their interface addresses, while neighbors on other types of links are identified by RID.
OSPFv3 removes this inconsistency, and all neighbors on all link types are identified by RID.

Note: OSPFv3 is not backward-compatible with OSPFv2; they can co-exist. To use OSPF with both IPv4 and IPv6, you must run both OSPFv2 and OSPFv3.

Autonomous System (AS) Areas

OSPF operate in a type of hierarchy. The largest entity within the hierarchy is the autonomous system (AS), which is a collection of networks under a common administration that share a common routing strategy. OSPF is an intra-AS (interior gateway) routing protocol, although it is capable of receiving routes from and sending routes to other ASs. An AS can be divided into a number of areas, which are groups of contiguous networks and attached hosts. Routers with multiple interfaces can participate in multiple areas. These routers, Area Border Routers (ABRs), maintain separate databases for each area. Areas are a logical grouping of OSPF routers identified by an integer or dotted-decimal number.
Areas allow you to further organize your routers within in the AS. One or more areas are required within the AS. Areas are valuable in that they allow sub-networks to "hide" within the AS, thus minimizing the size of the routing tables on all routers. An area within the AS may not see the details of another Area's topology. AS areas are known by their area number or the routers IP address.

482

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-1. Autonomous System Areas

Router M Router K Router F Router E Router L

Router D Router C

Area 200 Router G

Area 100 Area 0 Router B Router A Router I Router H

Router J

Area 300

Area Types
The Backbone of the network is Area 0. It is also called Area 0.0.0.0 and is the core of any Autonomous System (AS). All other areas must connect to Area 0. Areas can be defined in such a way that the

backbone is not contiguous. In this case, backbone connectivity must be restored through virtual links. Virtual links are configured between any backbone routers that share a link to a non-backbone area and function as if they were direct links.
An OSPF backbone is responsible for distributing routing information between areas. It consists of all Area Border Routers, networks not wholly contained in any area, and their attached routers. The Backbone is the only area with an default area number. All other areas can have their Area ID assigned in the configuration. Figure 24-1 shows Routers A, B, C, G, H, and I are the Backbone.

Open Shortest Path First (OSPFv2 and OSPFv3) | 483

www.dell.com | support.dell.com

A Stub Area (SA) does not receive external route information, except for the default route. These areas do receive information from inter-area (IA) routes. Note that all routers within an assigned Stub area must be configured as stubby, and no generate LSAs that do not apply. For example, a Type 5 LSA is intended for external areas and the Stubby area routers may not generate external LSAs. Stubby areas cannot be traversed by a virtual link. A Not-So-Stubby Area (NSSA) can import AS external route information and send it to the Backbone. It cannot received external AS information from the Backbone or other areas. It can be traversed by a virtual link. Totally Stubby Areas are referred to as No Summary areas in FTOS.

Networks and Neighbors

As a link-state protocol, OSPF sends routing information to other OSPF routers concerning the state of the links between them. The state (up or down) of those links is important. Routers that share a link become neighbors on that segment. OSPF uses the hello protocol as a neighbor discovery and keep alive mechanism. After two routers are neighbors, they may proceed to exchange and synchronize their databases, which creates an adjacency.

Router Types
Router types are attributes of the OSPF process. A given physical router may be a part of one or more OSPF processes. For example, a router connected to more than one area, receiving routing from a BGP process connected to another AS acts as both an Area Border Router and an Autonomous System Router. Each router has a unique ID, written in decimal format (A.B.C.D). The router ID does not have to be associated with a valid IP address. However, Dell Networking recommends that the router ID and the routers IP address reflect each other, to make troubleshooting easier. Figure 24-2gives some examples of the different router designations.

484

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-2. OSPF Routing Examples

Router M

Interior Router
Router K Router E Router F Router L

Interior Router

Router D Router C

Stub Area
Area 200 Router G

Not So Stubby Area

Area 100

Backbone Area
Area 0

Backbone Router
Router A Router I

Router B

Router H

Area Border Router Interior Router

Interior Router
Router J

Area 300

OSPF AS 9999

Router K

Autonomous System Boundary Router

Autonomous System Boundary Router

Router 8000

Router 82

Router 81

Interior Router OSPF AS 8888

Backbone Router (BR)

A Backbone Router (BR) is part of the OSPF Backbone, Area 0. This includes all Area Border Routers (ABRs). It can also include any routers that connect only to the Backbone and another ABR, but are only part of Area 0, such as Router I in Figure 24-2 above.

Open Shortest Path First (OSPFv2 and OSPFv3) | 485

www.dell.com | support.dell.com

Area Border Router (ABR)

Within an AS, an Area Border (ABR) connects one or more areas to the Backbone. The ABR keeps a copy of the link-state database for every area it connects to, so it may keep multiple copies of the link state database. An Area Border Router (ABR) takes information it has learned on one of its attached areas and can summarize it before sending it out on other areas it is connected to. An ABR can connect to many areas in an AS, and is considered a member of each area it connects to.

Autonomous System Border Router (ASBR)

The Autonomous System Border Area Router (ASBR) connects to more than one AS, and exchanges information with the routers in other ASs. Generally the ASBR connects to a non-Interior Gate Protocol (IGP) such as BGP or uses static routes.

Internal Router (IR)

The Internal Router (IR) has adjacencies with ONLY routers in the same area, as Router E, M and I are shown in Figure 24-2.

Designated and Backup Designated Routers

OSPF elects a Designated Router and a Backup Designated router. Among other things, the designated router is responsible for generating LSAs for the entire multiaccess network. Designated routers allow a reduction in network traffic and in the size of the topological database.
The Designated Router (DR) maintains a complete topology table of the network and sends the updates to the other routers via multicast. All routers in an area form a slave/master relationship with the DR. Every time a router sends an update, it sends it to the Designated Router (DR) and Backup Designated Router (BDR). The DR sends the update out to all other routers in the area. The Backup Designated Router (BDR) is the router that takes over if the DR fails.

Each router exchanges information with the DR and BDR. The DR and BDR relay the information to the other routers. On broadcast network segments the number of OSPF packets is further reduced by the DR and BDR sending such OSPF updates to a multicast IP address that all OSPF routers on the network segment are listening on. These router designations are not the same ad the router IDs discussed earlier. The Designated and Backup Designated Routers are configurable in FTOS. If no DR or BDR is defined in FTOS, the system assigns them. OSPF looks at the priority of the routers on the segment to determine which routers are the DR and BDR. The router with the highest priority is elected the DR. If there is a tie, then the router with the higher Router ID takes precedence. After the DR is elected, the BDR is elected the same way. A router with a router priority set to zero is cannot become the DR or BDR.

486

Open Shortest Path First (OSPFv2 and OSPFv3)

Link-State Advertisements (LSAs)

A Link-State Advertisement (LSA) communicates the router's local routing topology to all other local routers in the same area. OSPFv3 can treat LSAs as having link-local flooding scope, or store and flood them as if they are understood, while ignoring them in their own SPF algorithms. OSPFv2 always discards unknown LSA types.

The LSA types supported by Dell Networking are defined as follows: Type 1 - Router LSA The router lists links to other routers or networks in the same area. Type 1 LSAs are flooded across their own area only. The Link-State ID of the Type 1 LSA is the originating router ID. Type 2 - Network LSA The Designated Router (DR) in an area lists which routers are joined together within the area. Type 2 LSAs are flooded across their own area only. The Link-State ID of the Type 2 LSA is the IP interface address of the DR. Type 3 - Summary LSA (OSPFv2), Inter-Area-Prefix LSA (OSPFv3) An Area Border Router (ABR) takes information it has learned on one of its attached areas and can summarize it before sending it out on other areas it is connected to. The Link-State ID of the Type 3 LSA is the destination network number. Type 4 - AS Border Router Summary LSA (OSPFv2), Inter-Area-Router LSA (OSPFv3) In some cases, Type 5 External LSAs are flooded to areas where the detailed next-hop information may not be available. An Area Border Router will (ABR) flood the information for the router (i.e. the Autonomous System Border Router (ASBR) where the Type 5 advertisement originated. The Link-State ID for Type 4 LSAs is the router ID of the described ASBR. Type 5 - External LSA These LSAs contain information imported into OSPF from other routing processes. They are flooded to all areas, except stub areas. The Link-State ID of the Type 5 LSA is the external network number. Type 7 Routers in a Not-So-Stubby-Area (NSSA) do not receive external LSAs from Area Border Routers (ABRs), but are allowed to send external routing information for redistribution. They use Type 7 LSAs to tell the ABRs about these external routes, which the Area Border Router then translates to Type 5 external LSAs and floods as normal to the rest of the OSPF network. Type 8 - Link LSA (OSPFv3) This LSA carries the IPv6 address information of the local links. Type 9 - Link Local LSA (OSPFv2), Intra-Area-Prefix LSA (OSPFv3) For OSPFv2, this is a link-local "opaque" LSA as defined by RFC2370. For OSPFv3, this LSA carries the IPv6 prefixes of the router and network links.

For all LSA types, there are 20-byte LSA headers. One of the fields of the LSA header is the Link-State ID.

Open Shortest Path First (OSPFv2 and OSPFv3) | 487

www.dell.com | support.dell.com

Each router link is defined as one of four types: type 1, 2, 3, or 4. The LSA includes a link ID field that identifies, by the network number and mask, the object this link connects to. Depending on the type, the link ID has different meanings. 1: point-to-point connection to another router neighboring router 2: connection to a transit network IP address of Designated Router 3: connection to a stub network IP network/subnet number 4: virtual link neighboring router ID

Virtual Links
In the case in which an area cannot be directly connected to Area 0, you must configure a virtual link between that area and Area 0. The two endpoints of a virtual link are ABRs, and the virtual link must be configured in both routers. The common non-backbone area to which the two routers belong is called a transit area. A virtual link specifies the transit area and the router ID of the other virtual endpoint (the other ABR). A Virtual Link cannot be configured through a Stub Area or NSSA.

Router Priority and Cost

Router priority and cost is the method the system uses to rate the routers. For example, if not assigned, the system will select the router with the highest priority as the DR. The second highest priority is the BDR. Priority is a numbered rating 0-255. The higher the number, the higher the priority. Cost is a numbered rating 1-65535. The higher the number, the greater the cost. The cost assigned reflects the cost should the router fail. When a router fails and the cost is assessed, a new priority number results.

488

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-3. Priority and Costs Example

Router 2 Priority 180 Cost 50 Router 3 Priority 100 Cost 25

Router 1 Priority 200 Cost 21

Router 4 Priority 150 Cost 20

Router 1 selected by the system as DR. Router 2 selected by the system as BDR. If R1 fails, the system "subtracts" 21 from R1's priority number. R1's new priority is 179. R2 as both the selected BDR and the now-highest priority, becomes the DR. If R3 fails, the system "subtracts" 50 from its priority. R2's new priority is 130. R4 is now the highest priority and becomes the DR.

Implementing OSPF with FTOS

FTOS supports up to 10,000 OSPF routes. Within that 10,000 up to 8,000 routes can be designated as external and up to 2,000 designated as inter/intra area routes. FTOS version 7.8.1.0 and later support multiple OSPF processes (OSPF MP). Prior to 7.8.1.0, FTOS supports 1 OSPFv2 and 1 OSPFv3 process ID per system. Recall that OSPFv2 and OSPFv3 can coexist but must be configured individually. FTOS supports Stub areas, Totally Stub (No Summary) and Not So Stubby Areas (NSSAs) and supports the following LSAs, as discussed earlier in this document. Router (type 1) Network (type 2) Network Summary (type 3) AS Boundary (type 4) AS External (type 5)

Open Shortest Path First (OSPFv2 and OSPFv3) | 489

www.dell.com | support.dell.com

NSSA External (type 7) Opaque Link-local (type 9)

Graceful Restart
Graceful Restart supported on

c e and s platforms for both Helper and Restart modes.

When a router goes down without a Graceful Restart, there is a potential to lose access to parts of the network due to the necessity of network topology changes. Additionally, LSA flooding and reconvergence can cause substantial delays. It is, therefor, desirable that the network maintain a stable topology if it is possible for data flow to continue uninterrupted. OSPF Graceful Restart recognizes the fact that in a modern router, the control plane and data plane functionality are separate, restarting the control plane functionality (such as the failover of the active RPM to the backup in a redundant configuration), does not necessarily have to interrupt the forwarding of data packets. This behavior is supported because the forwarding tables previously computed by an active RPM have been downloaded into the Forwarding Information Base on the line cards (the data plane), and are still resident. For packets that have existing FIB/CAM entries, forwarding between ingress and egress ports/VLANs etc., can continue uninterrupted while the control plane OSPF process comes back to full functionality and rebuilds its routing tables.

Grace Period
The grace period is the time period that a router defines for other adjacent routers to wait for it to return to full control plane functionality, and which it advertises to its neighbors. When an outage occurs, the neighbors will wait for this time period before flooding LSAs and beginning reconvergence. When a router is attempting to restart gracefully, it will originate Type 9 link local opaque grace-LSAs that notify its helper neighbors that the restart process is beginning. During this period, the helper neighbor routers continue to announce the restarting router as fully adjacent, as long as the network topology remains unchanged. When the restarting router completes its restart, it flushes the grace-LSAs, thereby notifying its neighbors that the restart is complete. This should happen before the grace period expires.

Fast Convergence (OSPFv2, IPv4 only)

Fast Convergence allows you to define the speeds at which LSAs are originated and accepted, and reduce OSPFv2 end-to-end convergence time. FTOS enables you to accept and originate LSAa as soon as they are available to speed up route information propagation. Note that the faster the convergence, the more frequent the route calculations and updates. This will impact CPU utilization and may impact adjacency stability in larger topologies.

Multi-Process OSPF (OSPFv2, IPv4 only)

Multi-Process OSPF is supported on platforms and is supported on OSPFv2 with IPv4 only.

c e and s with FTOS version 7.8.1.0 and later,

490

Open Shortest Path First (OSPFv2 and OSPFv3)

Multi-Process OSPF allows multiple OSPFv2 processes on a single router. Multiple OSPFv2 processes allow for isolating routing domains, supporting multiple route policies and priorities in different domains, and creating smaller domains for easier management. The E-Series supports up to 28 OSPFv2 processes. The C-Series supports up to 6 OSPFv2 processes. The S50 and S25 support up to 4 OSPFv2 processes. The S55, S60, and S4810 support up to 16 OSPFv2 processes. The Z9000 supports up to 3 OSPFv2 processes.

Each OSPFv2 process has a unique process ID and must have an associated Router ID. There must be an equal number of interfaces must be in Layer-3 mode for the number of processes created. For example, if 5 OSPFv2 processes are created on a system, there must be at least 5 interfaces assigned in Layer-3 mode. Each OSPFv2 process is independent. If one process loses adjancency, the other processes continue to function/

Processing SNMP and Sending SNMP Traps

Though there are may be several OSPFv2 processes, only one process can process SNMP requests and send SNMP traps. The mib-binding command identifies one of the OSPVFv2 processes as the process responsible for SNMP management. If the mib-binding command is not specified, the first OSPFv2 process created manages the SNMP processes and traps.

RFC-2328 Compliant OSPF Flooding

In OSPF, flooding is the most resource-consuming task. The flooding algorithm described in RFC 2328 requires that OSPF flood LSAs on all interfaces, as governed by LSA's flooding scope. (Refer to Section 13 of the RFC.) When multiple direct links connect two routers, the RFC 2328 flooding algorithm generates significant redundant information across all links. By default, FTOS implements an enhanced flooding procedure which dynamically and intelligently detects when to optimize flooding. Wherever possible, the OSPF task attempts to reduce flooding overhead by selectively flooding on a subset of the interfaces between two routers. If RFC 2328 flooding behavior is required, enable it by using the command flood-2328 in ROUTER OSPF mode. When enabled, this command configures FTOS to flood LSAs on all interfaces. Confirm RFC 2328 flooding behavior by using the command debug ip ospf packet and look for output similar to the following:

Open Shortest Path First (OSPFv2 and OSPFv3) | 491

www.dell.com | support.dell.com

Figure 24-4. Enabling RFC-2328 Compliant OSPF Flooding

00:10:41 : OSPF(1000:00): Printed only for ACK packets Rcv. v:2 t:5(LSAck) l:64 Acks 2 rid:2.2.2.2 aid:1500 chk:0xdbee aut:0 auk: keyid:0 from:Vl 1000 LSType:Type-5 AS External id:160.1.1.0 adv:6.1.0.0 seq:0x8000000c LSType:Type-5 AS External id:160.1.2.0 adv:6.1.0.0 seq:0x8000000c 00:10:41 : OSPF(1000:00): Rcv. v:2 t:5(LSAck) l:64 Acks 2 rid:2.2.2.2 aid:1500 chk:0xdbee aut:0 auk: keyid:0 from:Vl 100 LSType:Type-5 AS External id:160.1.1.0 adv:6.1.0.0 seq:0x8000000c LSType:Type-5 AS External id:160.1.2.0 adv:6.1.0.0 seq:0x8000000c 00:10:41 : OSPF(1000:00): No change in update packets Rcv. v:2 t:4(LSUpd) l:100 rid:6.1.0.0 aid:0 chk:0xccbd aut:0 auk: keyid:0 from:Gi 10/21 Number of LSA:2 LSType:Type-5 AS External(5) Age:1 Seq:0x8000000c id:170.1.1.0 Adv:6.1.0.0 Netmask:255.255.255.0 fwd:0.0.0.0 E2, tos:0 metric:0 LSType:Type-5 AS External(5) Age:1 Seq:0x8000000c id:170.1.2.0 Adv:6.1.0.0 Netmask:255.255.255.0 fwd:0.0.0.0 E2, tos:0 metric:0

In FTOS Version, 7.5.1.0 use show ip ospf to confirm that RFC-2328 compliant OSPF flooding is enabled, as shown below.
Figure 24-5. Enabling RFC-2328 Compliant OSPF Flooding
FTOS#show ip ospf Routing Process ospf 1 with ID 2.2.2.2 Supports only single TOS (TOS0) routes It is an Autonomous System Boundary Router It is Flooding according to RFC 2328 SPF schedule delay 5 secs, Hold time between two SPFs 10 secs Number of area in this router is 1, normal 0 stub 0 nssa 1 --More--

OSPF ACK Packing

The OSPF ACK Packing feature bundles multiple LS acknowledgements in a single packet, significantly reducing the number of ACK packets transmitted when the number of LSAs increases. This feature also enhances network utilization and reduces the number of small ACK packets sent to a neighboring router. OSPF ACK packing is enabled by default, and non-configurable.

OSPF Adjacency with Cisco Routers

To establish an OSPF adjacency between Dell Networking and Cisco routers, the hello interval and dead interval must be the same on both routers. In FTOS the OSPF dead interval value is, by default, set to 40 seconds, and is independent of the OSPF hello interval. Configuring a hello interval does not change the dead interval in FTOS. In contrast, the OSPF dead interval on a Cisco router is, by default, four

times as long as the hello interval. Changing the hello interval on the Cisco router automatically changes the dead interval as well.

492

Open Shortest Path First (OSPFv2 and OSPFv3)

To ensure equal intervals between the routers, manually set the dead interval of the Dell Networking router to match the Cisco configuration. Use the command ip ospf dead-interval <x> in interface mode:
Figure 24-6. Command Example: ip ospf intervals
FTOS(conf)#int gi 2/2 FTOS(conf-if-gi-2/2)#ip ospf hello-interval 20 FTOS(conf-if-gi-2/2)#ip ospf dead-interval 80 FTOS(conf-if-gi-2/2)#

Dead Interval Set at 4x Hello Interval

Figure 24-7. OSPF Configuration with intervals set

FTOS (conf-if-gi-2/2)#ip ospf dead-interval 20 FTOS (conf-if-gi-2/2)#do show ip os int gi1/3 GigabitEthernet 2/2 is up, line protocol is up Internet Address 20.0.0.1/24, Area 0 Process ID 10, Router ID 1.1.1.2, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 1.1.1.2, Interface address 30.0.0.1 Backup Designated Router (ID) 1.1.1.1, Interface address 30.0.0.2 Timer intervals configured, Hello 20, Dead 80, Wait 20, Retransmit 5 Hello due in 00:00:04 Neighbor Count is 1, Adjacent neighbor count is 1 Adjacent with neighbor 1.1.1.1 (Backup Designated Router) FTOS (conf-if-gi-2/2)#

Dead Interval Set at 4x Hello Interval

For more information regarding this functionality or for assistance, go to http://support.dell.com/force10.

Configuration Information
The interfaces must be in Layer-3 mode (assigned an IP address) and enabled so that they can send and receive traffic. The OSPF process must know about these interfaces. To make the OSPF process aware of these interfaces, they must be assigned to OSPF areas. OSPF must be configured GLOBALLY on the system in CONFIGURATION mode. OSPF features and functions are assigned to each router using the CONFIG-INTERFACE commands for each interface. Note: By default, OSPF is disabled

Configuration Task List for OSPFv2 (OSPF for IPv4)

Open Shortest Path First version 2 (OSPF for IPv4) is supported on platforms

ces

1. Configure a physical interface. Assign an IP address, physical or loopback, to the interface to enable Layer 3 routing. 2. Enable OSPF globally. Assign network area and neighbors. 3. Add interfaces or configure other attributes.

Open Shortest Path First (OSPFv2 and OSPFv3) | 493

www.dell.com | support.dell.com

The following configuration steps include two mandatory steps and several optional ones: Enable OSPFv2 (mandatory) Enable Multi-Process OSPF Assign an OSPFv2 area (mandatory) Enable OSPFv2 on interfaces Configure stub areas Enable passive interfaces Enable fast-convergence Change OSPFv2 parameters on interfaces Enable OSPFv2 authentication Enable graceful restart Configure virtual links Redistribute routes Troubleshooting OSPFv2

For a complete listing of all commands related to OSPFv2, refer to the OSPF section in the FTOS Command Line Interface document.

Enable OSPFv2
Assign an IP address to an interface (physical or Loopback) to enable Layer 3 routing. By default OSPF, like all routing protocols, is disabled. You must configure at least one interface for Layer 3 before enabling OSPFv2 globally. If implementing, Multi-Process OSPF, you must create an equal number of Layer 3 enabled interfaces and OSPF Process IDs. For example, if you create 4 OSPFv2 process IDs, you must have 4 interfaces with Layer 3 enabled. Use these commands on one of the interfaces to enable OSPFv2 routing.
Step 1 Command Syntax
ip address ip-address mask

Command Mode CONFIG-INTERFACE

Usage Assign an IP address to an interface. Format: A.B.C.D/M

If using a Loopback interface, refer to Loopback Interfaces.

no shutdown

CONFIG-INTERFACE

Enable the interface.

494

Open Shortest Path First (OSPFv2 and OSPFv3)

Return to CONFIGURATION mode to enable the OSPF process. The OSPF Process ID is the identifying number assigned to the OSPF process, and the Router ID is the IP address associated with the OSPF process. .
Command Syntax
router ospf process-id [vrf {vrf name}]

Command Mode CONFIGURATION

Usage Enable the OSPFv2 process globally. Range: 0-65535 vrf name: Enter the VRF key word and instance name to tie the OSPF instance to the VRF. All network commands under this OSPF instance are subsequently tied to the VRF instance.

Once the OSPF process and the VRF are tied together, the OSPF Process ID cannot be used again in the system. If you try to enter an OSPF Process ID, or if you try to enable more OSPF processes than available Layer 3 interfaces, prior to assigning an IP address to an interface and setting the no shutdown command, you will see the following message.
Message 1
C300(conf)#router ospf 1 % Error: No router ID available.

In CONFIGURATION ROUTER OSPF mode, assign the Router ID. The Router ID is not required to be the routers IP address. Dell Networking recommends using the IP address as the Router ID for easier management and troubleshooting.
Command Syntax
router-id ip address

Command Mode CONFIG-ROUTER-O SPF-id

Usage Assign the Router ID for the OSPFv2 process. IP Address: A.B.C.D

Use the no router ospf process-id command syntax in the CONFIGURATION mode to disable OSPF. Use the clear ip ospf process-id command syntax in EXEC Privilege mode to reset the OSPFv2 process. Use the show ip ospf process-id command in EXEC mode (Figure 408) to view the current OSPFv2 status.
Figure 24-8. Command Example: show ip ospf process-id
FTOS#show ip ospf 55555 Routing Process ospf 55555 with ID 10.10.10.10 Supports only single TOS (TOS0) routes SPF schedule delay 5 secs, Hold time between two SPFs 10 secs Number of area in this router is 0, normal 0 stub 0 nssa 0 FTOS#

Open Shortest Path First (OSPFv2 and OSPFv3) | 495

www.dell.com | support.dell.com

Enable Multi-Process OSPF

Multi-Process OSPF allows multiple OSPFv2 processes on a single router. The following list shows the number of processes supported on each platform type. The E-Series supports up to 30 OSPFv2 processes. The C-Series supports up to 6 OSPFv2 processes. The S50 and S25 support up to 4 OSPFv2 processes. The S55, S60, and S4810 support up to 16 OSPFv2 processes. The Z9000 supports up to 3 OSPFv2 processes.

Follow the same steps as above, when configuring a single OSPF process. Repeat them as often as necessary for the desired number of processes. Once the process is created, all other configurations apply as usual,
Step 1 Command Syntax
ip address ip-address mask

Command Mode CONFIG-INTERFACE

Usage Assign an IP address to an interface. Format: A.B.C.D/M

If using a Loopback interface, refer to Loopback Interfaces.

no shutdown

CONFIG-INTERFACE

Enable the interface.

Return to CONFIGURATION mode to enable the OSPF process. The OSPF Process ID is the identifying number assigned to the OSPF process, and the Router ID is the IP address associated with the OSPF process. .
Command Syntax
router ospf process-id [vrf {vrf name}]

Command Mode CONFIGURATION

Usage Enable the OSPFv2 process globally. Range: 0-65535 vrf name: Enter the VRF key word and instance name to tie the OSPF instance to the VRF. All network commands under this OSPF instance are subsequently tied to the VRF instance.

Once the OSPF process and the VRF are tied together, the OSPF Process ID cannot be used again in the system. If you try to enable more OSPF processes than available Layer 3 interfaces you will see the following message.
Message 2
C300(conf)#router ospf 1 % Error: No router ID available.

496

Open Shortest Path First (OSPFv2 and OSPFv3)

In CONFIGURATION ROUTER OSPF mode, assign the Router ID. The Router ID is not required to be the routers IP address. Dell Networking recommends using the IP address as the Router ID for easier management and troubleshooting.
Command Syntax
router-id ip address

Command Mode CONFIG-ROUTER-O SPF-id

Usage Assign the Router ID for the OSPFv2 process. IP Address: A.B.C.D

Use the no router ospf process-id command syntax in the CONFIGURATION mode to disable OSPF. Use the clear ip ospf process-id command syntax in EXEC Privilege mode to reset the OSPFv2 process.

Assign an OSPFv2 area

After OSPFv2 is enabled, assign the interface to an OSPF area. Set up OSPF Areas and enable OSPFv2 on an interface with the network command. You must have at least one AS area: Area 0. This is the Backbone Area. If your OSPF network contains more than one area, you must also configure a backbone area (Area ID 0.0.0.0). Any area besides Area 0 can have any number ID assigned to it. The OSPFv2 process evaluates the network commands in the order they are configured. Assign the network address that is most explicit first to include all subnets of that address. For example, if you assign the network address 10.0.0.0 /8, you cannot assign the network address 10.1.0.0 /16 since it is already included in the first network address. When configuring the network command, you must configure a network address and mask that is a superset of the IP subnet configured on the Layer-3 interface to be used for OSPFv2. Use this command in CONFIGURATION ROUTER OSPF mode to set up each neighbor and OSPF area. The Area can be assigned by a number or with an IP interface address.
Command Syntax
network ip-address mask area area-id

Command Mode CONFIG-ROUTER-OSPF-id

Usage Enable OSPFv2 on an interface and assign an network address range to a specific OSPF area. IP Address Format: A.B.C.D/M Area ID Range: 0-65535 or A.B.C.D/M

Enable OSPFv2 on interfaces

Each interface must have OSPFv2 enabled on it. It must be configured for Layer 3 protocol, and not be shutdown. OSPFv2 can also be assigned to a loopback interface as a virtual interface.

Open Shortest Path First (OSPFv2 and OSPFv3) | 497

www.dell.com | support.dell.com

OSPF functions and features, such as MD5 Authentication, Grace Period, Authentication Wait Time, etc, are assigned on a per interface basis. Note: If using features like MD5 Authentication, ensure all the neighboring routers are also configured for MD5. Figure 24-9 presents an example of assigning an IP address to an interface and then assigning an OSPFv2 area that includes that Layer-3 interfaces IP address.
Figure 24-9. Configuring an OSPF Area Example
FTOS#(conf)#int gi 4/44 FTOS(conf-if-gi-4/44)#ip address 10.10.10.10/24 FTOS(conf-if-gi-4/44)#no shutdown FTOS(conf-if-gi-4/44)#ex FTOS(conf)#router ospf 1 FTOS(conf-router_ospf-1)#network 1.2.3.4/24 area 0 FTOS(conf-router_ospf-1)#network 10.10.10.10/24 area 1 FTOS(conf-router_ospf-1)#network 20.20.20.20/24 area 2 FTOS(conf-router_ospf-1)# FTOS# Assign Layer-3 interface with IP Address and no shutdown

Assign interfaces IP Address to an Area

Dell Networking recommends that the OSPFv2 Router ID be the interface IP addresses for easier management and troubleshooting. Use the show config command in CONFIGURATION ROUTER OSPF mode to view the configuration. OSPF, by default, sends hello packets out to all physical interfaces assigned an IP address that are a subset of a network on which OSPF is enabled. Use the show ip ospf interface command (Figure 410) to view the interfaces currently active and the areas assigned to the interfaces.
Figure 24-10. Command Example: show ip ospf process-id interface
FTOS>show ip ospf 1 interface GigabitEthernet 12/17 is up, line protocol is up Internet Address 10.2.2.1/24, Area 0.0.0.0 Process ID 1, Router ID 11.1.2.1, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 11.1.2.1, Interface address 10.2.2.1 Backup Designated Router (ID) 0.0.0.0, Interface address 0.0.0.0 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:04 Neighbor Count is 0, Adjacent neighbor count is 0 GigabitEthernet 12/21 is up, line protocol is up Internet Address 10.2.3.1/24, Area 0.0.0.0 Process ID 1, Router ID 11.1.2.1, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State BDR, Priority 1 Designated Router (ID) 13.1.1.1, Interface address 10.2.3.2 Backup Designated Router (ID) 11.1.2.1, Interface address 10.2.3.1 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:05 Neighbor Count is 1, Adjacent neighbor count is 1 Adjacent with neighbor 13.1.1.1 (Designated Router) FTOS>

498

Open Shortest Path First (OSPFv2 and OSPFv3)

Loopback interfaces also assist in the OSPF process. OSPF will pick the highest interface address as the router-id and a loopback interface address has a higher precedence than other interface addresses. Figure 24-11 gives an example of the show ip ospf process-id interface command with a Loopback interface.
Figure 24-11. Command Example: show ip ospf process-id interface
FTOS#show ip ospf 1 int GigabitEthernet 13/23 is up, line protocol is up Internet Address 10.168.0.1/24, Area 0.0.0.1 Process ID 1, Router ID 10.168.253.2, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State DROTHER, Priority 1 Designated Router (ID) 10.168.253.5, Interface address 10.168.0.4 Backup Designated Router (ID) 192.168.253.3, Interface address 10.168.0.2 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:08 Neighbor Count is 3, Adjacent neighbor count is 2 Adjacent with neighbor 10.168.253.5 (Designated Router) Adjacent with neighbor 10.168.253.3 (Backup Designated Router) Loopback 0 is up, line protocol is up Internet Address 10.168.253.2/32, Area 0.0.0.1 Process ID 1, Router ID 10.168.253.2, Network Type LOOPBACK, Cost: 1 Loopback interface is treated as a stub Host. FTOS#

Configure stub areas

OSPF supports different types of LSAs to help reduce the amount of router processing within the areas. Type 5 LSAs are not flooded into stub areas; the Area Border Router (ABR) advertises a default route into the stub area to which it is attached. Stub area routers use the default route to reach external destinations To ensure connectivity in your OSPFv2 network, never configure the backbone area as a stub area. Use these commands in the following sequence, starting in EXEC Privilege mode to configure a stub area.
Step 1 Command Syntax
show ip ospf process-id [vrf vrf name] database database-summary

Command Mode EXEC Privilege

Usage Review all areas after they were configured to determine which areas are NOT receiving type 5 LSAs. vrf name: Show only the OSPF information tied to the VRF process.

configure

EXEC Privilege

Enter the CONFIGURATION mode.

Open Shortest Path First (OSPFv2 and OSPFv3) | 499

www.dell.com | support.dell.com

Step

Command Syntax
router ospf process-id [vrf {vrf
name}]

Command Mode CONFIGURATION

Usage Enter the ROUTER OSPF mode. Process ID is the ID assigned when configuring OSPFv2 globally (page 58). vrf name: Enter the VRF key word and instance name to tie the OSPF instance to the VRF. All network commands under this OSPF instance are subsequently tied to the VRF instance. Configure the area as a stub area. Use the no-summary keywords to prevent transmission in to the area of summary ASBR LSAs. Area ID is the number or IP address assigned when creating the Area (page 60).

area area-id stub [no-summary]

CONFIG-ROUTER-O SPF-id

Use the show ip ospf database process-id database-summary command syntax (Figure 413) in the EXEC Privilege mode To view which LSAs are transmitted.
Figure 24-12. Command Example: show ip ospf process-id database database-summary
FTOS#show ip ospf 34 database database-summary OSPF Router with ID (10.1.2.100) (Process ID 34) Area ID 2.2.2.2 3.3.3.3 FTOS# Router 1 1 Network S-Net 0 0 0 0 S-ASBR 0 0 Type-7 0 0 Subtotal 1 1

To view information on areas, use the show ip ospf process-id command in the EXEC Privilege mode.

Enable passive interfaces

A passive interface is one that does not send or receive routing information. Enabling passive interface suppresses routing updates on an interface. Although the passive interface will neither send nor receive routing updates, the network on that interface will still be included in OSPF updates sent via other interfaces.

500

Open Shortest Path First (OSPFv2 and OSPFv3)

Use the following command in the ROUTER OSPF mode to suppress the interfaces participation on an OSPF interface. This command stops the router from sending updates on that interface.
Command Syntax
passive-interface {default | interface}

Command Mode CONFIG-ROUTEROSPF-id

Usage Specify whether all or some of the interfaces will be passive. Default enabled passive interfaces on ALL interfaces in the OSPF process. Entering the physical interface type, slot, and number enable passive interface on only the identified interface. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information (e.g. passive-interface gi 2/1). For a port channel, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale (e.g. passive-interface po 100) For a SONET interface, enter the keyword sonet followed by the slot/port information (e.g. passive-interface so 2/2). For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information ( e.g. passive-interface ten 2/3). For a VLAN, enter the keyword vlan followed by a number from 1 to 4094 (e.g. passive-interface vlan 2222). E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

The default keyword sets all interfaces on this OSPF process as passive. The passive interface can be removed from select interfaces using the no passive-interface interface command while passive interface default is configured.

To enable both receiving and sending routing updates, enter the no passive-interface interface command. When you configure a passive interface, the show ip ospf process-id interface command (Figure 413) adds the words passive interface to indicate that hello packets are not transmitted on that interface.

Open Shortest Path First (OSPFv2 and OSPFv3) | 501

www.dell.com | support.dell.com

Figure 24-13.

Command Example: show ip ospf process-id interface

FTOS#show ip ospf 34 int GigabitEthernet 0/0 is up, line protocol is down Internet Address 10.1.2.100/24, Area 1.1.1.1 Process ID 34, Router ID 10.1.2.100, Network Type BROADCAST, Cost: 10 Transmit Delay is 1 sec, State DOWN, Priority 1 Designated Router (ID) 10.1.2.100, Interface address 0.0.0.0 Backup Designated Router (ID) 0.0.0.0, Interface address 0.0.0.0 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 13:39:46 Neighbor Count is 0, Adjacent neighbor count is 0 GigabitEthernet 0/1 is up, line protocol is down Internet Address 10.1.3.100/24, Area 2.2.2.2 Process ID 34, Router ID 10.1.2.100, Network Type BROADCAST, Cost: 10 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 10.1.2.100, Interface address 10.1.3.100 Backup Designated Router (ID) 0.0.0.0, Interface address 0.0.0.0 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Interface is not running No Hellos (Passive interface) Neighbor Count is 0, Adjacent neighbor count is 0 OSPF protocol. Loopback 45 is up, line protocol is up Internet Address 10.1.1.23/24, Area 2.2.2.2 Process ID 34, Router ID 10.1.2.100, Network Type LOOPBACK, Cost: 1 Loopback interface is treated as a stub Host. FTOS#

the

Enable fast-convergence
The fast-convergence CLI sets the minimum origination and arrival LSA parameters to zero (0), allowing rapid route calculation. When fast-convergence is disabled, origination and arrival LSA parameters are set to 5 seconds and 1 second, respectively. Setting the convergence parameter (1-4) indicates the actual convergence level. Each convergence setting adjusts the LSA parameters to zero, but the fast-convergence parameter setting allows for even finer tuning of the convergence speed. The higher the number, the faster the convergence. Use the following command in the ROUTER OSPF mode to enable or disable fast-convergence.
Command Syntax
fast-convergence {number}

Command Mode CONFIG-ROUTEROSPF-id

Usage Enable OSPF fast-convergence and specify the convergence level. Parameter: 1-4 The higher the number, the faster the convergence. When disabled, the parameter is set at 0 (Figure 24-15).

Note: A higher convergence level can result in occasional loss of OSPF adjacency. Generally, convergence level 1 meets most convergence requirements. Higher convergence levels should only be selected following consultation with Dell Networking technical support.

502

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-14 shows the convergence settings when fast-convergence is enabled and Figure 24-15 shows settings when fast-convergence is disabled. These displays appear with the show ip ospf command.
Figure 24-14. Command Example: show ip ospf process-id (fast-convergence enabled)

FTOS(conf-router_ospf-1)#fast-converge 2 FTOS(conf-router_ospf-1)#ex FTOS(conf)#ex FTOS#show ip ospf 1 Routing Process ospf 1 with ID 192.168.67.2 Supports only single TOS (TOS0) routes SPF schedule delay 5 secs, Hold time between two SPFs 10 secs Convergence Level 2 Min LSA origination 0 secs, Min LSA arrival 0 secs Number of area in this router is 0, normal 0 stub 0 nssa 0 FTOS#

Fast-converge parameter setting LSA Parameters

Figure 24-15.

Command example: show ip ospf process-id (fast-convergence disabled)

FTOS#(conf-router_ospf-1)#no fast-converge FTOS#(conf-router_ospf-1)#ex FTOS#(conf)#ex FTOS##show ip ospf 1 Routing Process ospf 1 with ID 192.168.67.2 Supports only single TOS (TOS0) routes SPF schedule delay 5 secs, Hold time between two SPFs 10 secs Convergence Level 0 Min LSA origination 5 secs, Min LSA arrival 1 secs Number of area in this router is 0, normal 0 stub 0 nssa 0 FTOS#

Fast-converge parameter setting LSA Parameters

Change OSPFv2 parameters on interfaces

In FTOS, you can modify the OSPF settings on the interfaces. Some interface parameter values must be consistent across all interfaces to avoid routing errors. For example, you must set the same time interval for the hello packets on all routers in the OSPF network to prevent misconfiguration of OSPF neighbors.

Open Shortest Path First (OSPFv2 and OSPFv3) | 503

www.dell.com | support.dell.com

Use any or all of the following commands in CONFIGURATION INTERFACE mode to change OSPFv2 parameters on the interfaces:
Command Syntax
ip ospf cost

Command Mode CONFIG-INTERFACE

Usage Change the cost associated with OSPF traffic on the interface. Cost: 1 to 65535 (default depends on the interface speed). Change the time interval the router waits before declaring a neighbor dead. Configure Seconds range: 1 to 65535 (default is 40 seconds).

ip ospf dead-interval seconds

CONFIG-INTERFACE

The dead interval must be four times the hello interval. The dead interval must be the same on all routers in the OSPF network.
ip ospf hello-interval seconds

CONFIG-INTERFACE

Change the time interval between hello-packet transmission. Seconds range: from 1 to 65535 (default is 10 seconds).

The hello interval must be the same on all routers in the OSPF network.
ip ospf message-digest-key keyid md5 key

CONFIG-INTERFACE

Use the MD5 algorithm to produce a message digest or key, which is sent instead of the key. Keyid range: 1 to 255 Key: a character string

Be sure to write down or otherwise record the Key. You cannot learn the key once it is configured. You must be careful when changing this key.
ip ospf priority number

CONFIG-INTERFACE

Change the priority of the interface, which is used to determine the Designated Router for the OSPF broadcast network. Number range: 0 to 255 (the default is 1). Change the retransmission interval between LSAs. Seconds range: from 1 to 65535 (default is 5 seconds).

ip ospf retransmit-interval seconds

CONFIG-INTERFACE

The retransmit interval must be the same on all routers in the OSPF network.
ip ospf transmit-delay seconds

CONFIG-INTERFACE

Change the wait period between link state update packets sent out the interface. Seconds range: from 1 to 65535 (default is 1 second).

The transmit delay must be the same on all routers in the OSPF network.

Use the show config command in CONFIGURATION INTERFACE mode (Figure 24-16) to view interface configurations. Use the show ip ospf interface command in EXEC mode to view interface status in the OSPF process.

504

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-16.

Changing the OSPF Cost Value on an Interface

FTOS(conf-if)#ip ospf cost 45 FTOS(conf-if)#show config ! interface GigabitEthernet 0/0 ip address 10.1.2.100 255.255.255.0 no shutdown ip ospf cost 45 FTOS(conf-if)#end FTOS#show ip ospf 34 interface

The change is made on the interface and it is reflected in the OSPF configuration

GigabitEthernet 0/0 is up, line protocol is up Internet Address 10.1.2.100/24, Area 2.2.2.2 Process ID 34, Router ID 10.1.2.100, Network Type BROADCAST, Cost: 45 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 10.1.2.100, Interface address 10.1.2.100 Backup Designated Router (ID) 10.1.2.100, Interface address 0.0.0.0 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:06 Neighbor Count is 0, Adjacent neighbor count is 0 FTOS#

Enable OSPFv2 authentication

Use the following commands in CONFIGURATION INTERFACE mode to enable or change various OSPF authentication parameters:
Command Syntax
ip ospf authentication-key key

Command Mode CONFIG-INTERFACE

Usage Set clear text authentication scheme on the interface. Configure a key that is a text string no longer than eight characters. All neighboring routers must share the same password to exchange OSPF information. Set the authentication change wait time in seconds between 0 and 300 for the interface. This is the amount of time OSPF has available to change its interface authentication type. During the auth-change-wait-time, OSPF sends out packets with both the new and old authentication schemes. This transmission stops when the period ends. The default is 0 seconds.

ip ospf auth-change-wait-time seconds

CONFIG-INTERFACE

Enable graceful restart

Graceful Restart is enabled for the global OSPF process. Use these commands to configure OSPF graceful restart. Refer to Graceful Restart for feature details. The Dell Networking implementation of OSPF graceful restart enables you to specify:
grace periodthe length of time the graceful restart process can last before OSPF terminates it.

Open Shortest Path First (OSPFv2 and OSPFv3) | 505

www.dell.com | support.dell.com

helper-reject neighborsthe router ID of each restart router that does not receive assistance from the

configured router. modethe situation or situations that trigger a graceful restart. rolethe role or roles the configured router can perform.

Note: By default, OSPF graceful restart is disabled.

You enable OSPF graceful restart in CONFIGURATION ROUTER OSPF mode.

Command Syntax
graceful-restart grace-period seconds

Command Mode CONFIG-ROUTEROSPF-id

Usage Enable OSPF graceful-restart globally and set the grace period. Seconds range: between 40 and 3000

This is the period of time that an OSPF routers neighbors will advertise it as fully adjacent, regardless of the synchronization state, during a graceful restart. OSPF terminates this process when the grace period ends.
graceful-restart helper-reject router-id

CONFIG-ROUTEROSPF-id

Enter the Router ID of the OSPF helper router from which the router does not accept graceful restart assistance. This applies to the specified router only. IP Address: A.B.C.D Specify the operating mode in which graceful-restart functions. FTOS supports the following options: Planned-only. The OSPF router supports graceful-restart for planned restarts only. A planned restart is when the user manually enters a fail-over command to force the primary RPM over to the secondary RPM. During a planned restart, OSPF sends out a Grace LSA before the system switches over to the secondary RPM. OSPF also is notified that a planned restart is happening. Unplanned-only. The OSPF router supports graceful-restart for only unplanned restarts. During an unplanned restart, OSPF sends out a Grace LSA once the secondary RPM comes online.

graceful-restart mode [planned-only | unplanned-only]

CONFIG-ROUTEROSPF-id

By default, OSPF supports both planned and unplanned restarts. Selecting one or the other mode restricts OSPF to the single selected mode.
graceful-restart role [helper-only | restart-only]

CONFIG-ROUTEROSPF-id

Configure the graceful restart role or roles that this OSPF router performs. FTOS supports the following options: Helper-only. The OSPF router supports graceful-restart only as a helper router. Restart-only. The OSPF router supports graceful-restart only during unplanned restarts.

By default, OSPF supports both restarting and helper roles. Selecting one or the other role restricts OSPF to the single selected role.

When you configure a graceful restart, the show run ospf command (Figure 24-17) displays information similar to the following.

506

Open Shortest Path First (OSPFv2 and OSPFv3)

Figure 24-17.

Command Example: show run ospf (partial)

FTOS#show run ospf ! router ospf 1 graceful-restart grace-period 300 graceful-restart role helper-only graceful-restart mode unplanned-only graceful-restart helper-reject 10.1.1.1 graceful-restart helper-reject 20.1.1.1 network 10.0.2.0/24 area 0 FTOS#

Use the following command to disable OSPF graceful-restart after you have enabled it.
Command Syntax
no graceful-restart grace-period

Command Mode CONFIG-ROUTEROSPF-id

Usage Disable OSPF graceful-restart. Returns OSPF graceful-restart to its default state.

For more information on OSPF graceful restart, refer to the FTOS Command Line Interface Reference.

Configure virtual links

Areas within OSPF must be connected to the backbone area (Area ID 0.0.0.0). If an OSPF area does not have a direct connection to the backbone, at least one virtual link is required. Virtual links must be configured on an ABR connected to the backbone. hello-interval: help packet retransmit-interval: LSA retransmit interval transmit-delay: LSA transmission delay dead-interval: dead router detection time authentication-key: authentication key message-digest-key: MD5 authentication key

Open Shortest Path First (OSPFv2 and OSPFv3) | 507

www.dell.com | support.dell.com

Use the following command in CONFIGURATION ROUTER OSPF mode to configure virtual links.
Command Syntax
area area-id virtual-link router-id [hello-interval seconds | retransmit-interval seconds | transmit-delay seconds | dead-interval seconds | authentication-key key | message-digest-key keyid md5 key]

Command Mode CONFIG-ROUTEROSPF-id

Usage Configure the optional parameters of a virtual link: Area ID: assigned earlier (0-65535 or A.B.C.D) Router ID: IP address associated with the virtual link neighbor Hello Interval Seconds: 1-8192 (default 10) Retransmit Interval Seconds: 1-3600 (default 5) Transmit Delay Seconds: 1-3600 (default 1) Dead Interval Seconds: 1-8192 (default 40) Authentication Key: 8 characters Message Digest Key: 1-255 MD5 Key: 16 characters

Only the Area ID and Router ID require configuration to create a virtual link. If no other parameter is entered, the defaults are used. Use EITHER the Authentication Key or the Message Digest (MD5) key.

Use the show ip ospf process-id virtual-links command (Figure 24-18) in the EXEC mode to view the virtual link.
Figure 24-18. Command Example: show ip ospf process-id virtual-links
FTOS#show ip ospf 1 virtual-links Virtual Link to router 192.168.253.5 is up Run as demand circuit Transit area 0.0.0.1, via interface GigabitEthernet 13/16, Cost of using 2 Transmit Delay is 1 sec, State POINT_TO_POINT, Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:02 FTOS#

508

Open Shortest Path First (OSPFv2 and OSPFv3)

Filter routes
To filter routes, use prefix lists. OSPF applies prefix lists to incoming or outgoing routes. Incoming routes must meet the conditions of the prefix lists, and if they do not, OSPF does not add the route to the routing table. Configure the prefix list in CONFIGURATION PREFIX LIST mode prior to assigning it to the OSPF process.
Command Syntax
ip prefix-list prefix-name

Command Mode CONFIGURATION

Usage

Create a prefix list and assign it a unique name. You are in PREFIX LIST mode. Create a prefix list with a sequence. number and a deny or permit action. The optional parameters are: ge min-prefix-length: is the minimum prefix length to be matched (0 to 32).
le max-prefix-length: is the maximum prefix length to be matched (0 to 32).

seq sequence-number {deny |permit} ip-prefix [ge min-prefix-length] [le max-prefix-length]

CONFIG- PREFIX LIST

For configuration information on prefix lists, refer to IP Access Control Lists, Prefix Lists, and Route-maps chapter in the FTOS Configuration Guide. Use the following commands in CONFIGURATION-ROUTER OSPF mode to apply prefix lists to incoming or outgoing OSPF routes
Command Syntax
distribute-list prefix-list-name in [interface] distribute-list prefix-list-name out [connected | isis | rip | static]

Command Mode CONFIG-ROUTEROSPF-id CONFIG-ROUTEROSPF-id

Usage Apply a configured prefix list to incoming OSPF routes. Assign a configured prefix list to outgoing OSPF routes.

Open Shortest Path First (OSPFv2 and OSPFv3) | 509

www.dell.com | support.dell.com

Redistribute routes
You can add routes from other routing instances or protocols to the OSPF process. With the redistribute command syntax, you can include RIP, static, or directly connected routes in the OSPF process.
Note: Do not route iBGP routes to OSPF unless there are route-maps associated with the OSPF redistribution.

Use the following command in CONFIGURATION- ROUTER-OSPF mode to redistribute routes:

Command Syntax
redistribute {bgp | connected | isis | rip | static} [metric metric-value | metric-type type-value] [route-map map-name] [tag tag-value]

Command Mode CONFIG-ROUTEROSPF-id

Usage Specify which routes will be redistributed into OSPF process. Configure the following required and optional parameters: bgp, connected, isis, rip, or static: enter one of the keyword to redistribute those routes. rip is supported only on E-Series. metric metric-value range: 0 to 4294967295. metric-type metric-type: 1 for OSPF external route type 1 or 2 for OSPF external route type 2. route-map map-name: enter a name of a configured route map. tag tag-value range: 0 to 4294967295.

To view the current OSPF configuration, use the show running-config ospf command in the EXEC mode or the show config command in the ROUTER OSPF mode
Figure 24-19. Command Example: show config

FTOS(conf-router_ospf)#show config ! router ospf 34 network 10.1.2.32 0.0.0.255 area 2.2.2.2 network 10.1.3.24 0.0.0.255 area 3.3.3.3 distribute-list dilling in FTOS(conf-router_ospf)#

510

Open Shortest Path First (OSPFv2 and OSPFv3)

Troubleshooting OSPFv2
FTOS has several tools to make troubleshooting easier. Be sure to check the following, as these are typical issues that interrupt an OSPFv2 process. Note that this is not a comprehensive list, just some examples of typical troubleshooting checks. Has OSPF been enabled globally? Is the OSPF process active on the interface? Are adjacencies established correctly? Are the interfaces configured for Layer 3 correctly? Is the router in the correct area type? Have the routes been included in the OSPF database? Have the OSPF routes been included in the routing table (not just the OSPF database)?

Some useful troubleshooting commands are: show interfaces show protocols debug IP OSPF events and/or packets show neighbors show virtual links show routes

Use the show running-config ospf command to see the state of all the enabled OSPFv2 processes.
Command Syntax
show running-config ospf

Command Mode EXEC Privilege

Usage View the summary of all OSPF process IDs enables on the router.

Open Shortest Path First (OSPFv2 and OSPFv3) | 511

www.dell.com | support.dell.com

Figure 24-20.

Command Example: show running-config ospf

FTOS#show run ospf ! router ospf 3 ! router ospf 4 router-id 4.4.4.4 network 4.4.4.0/28 area 1 ! router ospf 5 ! router ospf 6 ! router ospf 7 mib-binding ! router ospf 8 ! router ospf 90 area 2 virtual-link 4.4.4.4 area 2 virtual-link 90.90.90.90 retransmit-interval 300 ! ipv6 router ospf 999 default-information originate always router-id 10.10.10.10 FTOS#

Use the following commands in EXEC Privilege mode to get general route and links status information.
Command Syntax
show ip route summary

Command Mode EXEC Privilege EXEC Privilege

Usage View the summary information of the IP routes View the summary information for the OSPF database

show ip ospf database

Use the following command in EXEC Privilege mode to view the OSPFv2 configuration for a neighboring router:
Command Syntax
show ip ospf neighbor

Command Mode EXEC Privilege

Usage View the configuration of OSPF neighbors.

512

Open Shortest Path First (OSPFv2 and OSPFv3)

Use the following command in EXEC Privilege mode to configure the debugging options of an OSPFv2 process:
Command Syntax
debug ip ospf process-id [event | packet | spf]

Command Mode EXEC Privilege

Usage View debug messages. To view debug messages for a specific OSPF process ID, enter debug ip ospf process-id. If you do not enter a process ID, the command applies to the first OSPF process. To view debug messages for a specific operation, enter one of the optional keywords: event: view OSPF event messages packet: view OSPF packet information. spf: view shortest path first (spf) information.

Configuration Task List for OSPFv3 (OSPF for IPv6)

Open Shortest Path First version 3 (OSPF for IPv6) is supported on platforms

ce

The configuration options of OSPFv3 are the same as those for OSPFv2, but may be configured with differently labeled commands. Process IDs and areas need to be specified. Interfaces and addresses need to be included in the process. Areas can be defined as stub or totally stubby. The interfaces must be in IPv6 Layer-3 mode (assigned an IPv6 IP address) and enabled so that they can send and receive traffic. The OSPF process must know about these interfaces. To make the OSPF process aware of these interfaces, they must be assigned to OSPF areas. TheOSPFv3 ipv6 ospf area command enables OSPFv3 on the interface and places the interface in an area. With OSPFv2, two commands are required to accomplish the same tasks: the router ospf command to create the OSPF process, then the network area command to enable OSPF on an interface. Note that the OSPFv2 network area command can enable OSPF on multiple interfaces with the single command, while the OSPFv3 ipv6 ospf area command must be configured on each interface that will be running OSPFv3. All IPv6 addresses on an interface are included in the OSPFv3 process that is created on the interface. OSPFv3 for IPv6 is enabled by specifying an OSPF Process ID and an Area in the INTERFACE mode. If an OSPFv3 process has not yet been created, it is created automatically. All IPv6 addresses configured on the interface are included in the specified OSPF process.
Note: IPv6 and OSPFv3 do not support Multi-Process OSPF. Only a single OSPFv3 process is can be enabled.

Enable IPv6 Unicast Routing Assign IPv6 addresses on an interface Assign Area ID on interface Assign OSPFv3 Process ID and Router ID Globally

Open Shortest Path First (OSPFv2 and OSPFv3) | 513

www.dell.com | support.dell.com

Configure stub areas Configure Passive-Interface Redistribute routes Configure a default route

Enable IPv6 Unicast Routing

Command Syntax
ipv6 unicast routing

Command Mode CONFIGURATION

Usage Enables IPv6 unicast routing globally.

Assign IPv6 addresses on an interface

Command Syntax
ipv6 address ipv6 address

Command Mode CONF-INT-type slot/port

Usage Assign IPv6 address to the interface. IPv6 addresses are normally written as eight groups of four hexadecimal digits, where each group is separated by a colon (:). FORMAT: A:B:C::F/128 Bring the interface up.

no shutdown

CONF-INT-type slot/port

Assign Area ID on interface

Command Syntax
ipv6 ospf process-id area area-id

Command Mode CONF-INT-type slot/port

Usage Assign the OSPFv3 process and an OSPFv3 area to this interface. process-id: The Process ID number assigned above. area-id: the area ID for this interface.

The ipv6 ospf area command enables OSPFv3 on an interface and places the interface in the specified area. Additionally, it creates the OSPFv3 process with ID on the router. OSPFv2 required two commands are required to accomplish the same tasks: the router ospf command to create the OSPF process, then the network area command to enable OSPFv2 on an interface. Note that the OSPFv2 network area command can enable OSPFv2 on multiple interfaces with the single command, whereas the OSPFv3 ipv6 ospf area command must be configured on each interface that will be running OSPFv3.

514

Open Shortest Path First (OSPFv2 and OSPFv3)

Assign OSPFv3 Process ID and Router ID Globally

Command Syntax
ipv6 router ospf {process ID}

Command Mode CONFIGURATION

Usage Enable the OSPFv3 process globally and enter OSPFv3 mode. Range: 0-65535 Assign the Router ID for this OSPFv3 process number: IPv4 address Format: A.B.C.D

router-id {number}

CONF-IPV6-ROUTER-OSPF

Note: The router-id for an OSPFv3 router is entered as an IPv4 IP address.

Configure stub areas

Command Syntax
area area-id stub [no-summary]

Command Mode CONF-IPV6-ROUTER-OSPF

Usage Configure the area as a stub area. Use the no-summary keywords to prevent transmission in to the area of summary ASBR LSAs. Area ID is a number or IP address assigned when creating the Area. The Area ID can be represented as a number between 0 65536 if a dotted decimal format is assigned, rather than an IP address.

Open Shortest Path First (OSPFv2 and OSPFv3) | 515

www.dell.com | support.dell.com

Configure Passive-Interface
Use the following command to suppress the interfaces participation on an OSPFv3 interface. This command stops the router from sending updates on that interface.
Command Syntax
passive-interface {type slot/port}

Command Mode CONF-IPV6-ROUTER-OSPF

Usage Specify whether some or all some of the interfaces will be passive. Interface identifies the specific interface that will be passive. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information (e.g. passive-interface gi 2/1). For a port channel, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale (e.g. passive-interface po 100) For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information ( e.g. passive-interface ten 2/3). For a VLAN, enter the keyword vlan followed by a number from 1 to 4094 (e.g. passive-interface vlan 2222). E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

To enable both receiving and sending routing updates, enter the no passive-interface interface command. When you configure a passive interface, the show ipv6 ospf interface command adds the words passive interface to indicate that hello packets are not transmitted on that interface.

516

Open Shortest Path First (OSPFv2 and OSPFv3)

Redistribute routes
You can add routes from other routing instances or protocols to the OSPFv3 process. With the redistribute command syntax, you can include RIP, static, or directly connected routes in the OSPF process.
Command Syntax
redistribute {bgp | connected | static} [metric metric-value | metric-type type-value] [route-map map-name] [tag tag-value]

Command Mode CONF-IPV6-ROUTER-OSPF

Usage Specify which routes will be redistributed into OSPF process. Configure the following required and optional parameters: bgp, connected, or static: enter one of the keyword to redistribute those routes. metric metric-value range: 0 to 4294967295. metric-type metric-type: 1 for OSPFv3 external route type 1 OR 2 for OSPFv3 external route type 2. route-map map-name: enter a name of a configured route map. tag tag-value range: 0 to 4294967295.

Configure a default route

Configure FTOS to generate a default external route into the OSPFv3 routing domain.
Command Syntax
default-information originate [always [metric metric-value] [metric-type type-value]] [route-map map-name]

Command Mode CONF-IPV6-ROUTER-OSPF

Usage Specify the information for the default route. Configure the following required and optional parameters: always: indicate that default route information must always be advertised metric metric-value range: 0 to 4294967295. metric-type metric-type: 1 for OSPFv3 external route type 1 OR 2 for OSPFv3 external route type 2. route-map map-name: enter a name of a configured route map.

Open Shortest Path First (OSPFv2 and OSPFv3) | 517

www.dell.com | support.dell.com

Troubleshooting OSPFv3
FTOS has several tools to make troubleshooting easier. Be sure to check the following, as these are typical issues that interrupt the OSPFv3 process. Note that this is not a comprehensive list, just some examples of typical troubleshooting checks. Has OSPF been enabled globally? Is the OSPF process active on the interface? Are adjacencies established correctly? Are the interfaces configured for Layer 3 correctly? Is the router in the correct area type? Have the routes been included in the OSPF database? Have the OSPF routes been included in the routing table (not just the OSPF database)?

Some useful troubleshooting commands are: show ipv6 interfaces show ipv6 protocols debug IPv6 OSPF events and/or packets show ipv6 neighbors show virtual links show ipv6 routes

Use the following commands in EXEC Privilege mode to get general route and links status information.
Command Syntax
show ipv6 route summary

Command Mode EXEC Privilege EXEC Privilege

Usage View the summary information of the IPv6 routes View the summary information for the OSPFv3 database

show ipv6 ospf database

Use the following command in EXEC Privilege mode to view the OSPF configuration for a neighboring router:
Command Syntax
show ipv6 ospf neighbor

Command Mode EXEC Privilege

Usage View the configuration of OSPFv3 neighbors.

518

Open Shortest Path First (OSPFv2 and OSPFv3)

Use the following command in EXEC Privilege mode to configure the debugging options of an OSPFv3 process:
Command Syntax
debug ipv6 ospf packet {type slot/port}

Command Mode EXEC Privilege

Usage View debug messages for all OSPFv3 interfaces. packet: view OSPF packet information. For a Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information (e.g. passive-interface gi 2/1). For a port channel, enter the keyword port-channel followed by a number from 1 to 255 for TeraScale and ExaScale (e.g. passive-interface po 100) For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information ( e.g. passive-interface ten 2/3). For a VLAN, enter the keyword vlan followed by a number from 1 to 4094 (e.g. passive-interface vlan 2222). E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

Sample Configurations for OSPFv2

The following configurations are examples for enabling OSPFv2. These are not comprehensive directions. They are intended to give you a some guidance with typical configurations. You can copy and paste from these examples to your CLI. Be sure you make the necessary changes to support your own IP Addresses, Interfaces, Names, etc.

Basic OSPFv2 Router Topology

The following illustration is a sample basic OSPFv2 topology.

Open Shortest Path First (OSPFv2 and OSPFv3) | 519

www.dell.com | support.dell.com

Figure 24-21.

Basic topology and CLI commands for OSPFv2

OSPF AREA 0

GI 1/1

GI 2/1

GI 1/2

GI 2/2

GI 3/1

GI 3/2

router ospf 11111 network 10.0.11.0/24 area 0 network 10.0.12.0/24 area 0 network 192.168.100.0/24 area 0 ! interface GigabitEthernet 1/1 ip address 10.1.11.1/24 no shutdown ! interface GigabitEthernet 1/2 ip address 10.2.12.2/24 no shutdown ! interface Loopback 10 ip address 192.168.100.100/24 no shutdown

router ospf 33333 network 192.168.100.0/24 area 0 network 10.0.13.0/24 area 0 network 10.0.23.0/24 area 0 ! interface Loopback 30 ip address 192.168.100.100/24 no shutdown ! interface GigabitEthernet 3/1 ip address 10.1.13.3/24 no shutdown ! interface GigabitEthernet 3/2 ip address 10.2.13.3/24 no shutdown

router ospf 22222 network 192.168.100.0/24 area 0 network 10.2.21.0/24 area 0 network 10.2.22.0/24 area 0 ! interface Loopback 20 ip address 192.168.100.20/24 no shutdown ! interface GigabitEthernet 2/1 ip address 10.2.21.2/24 no shutdown ! interface GigabitEthernet 2/2 ip address 10.2.22.2/24 no shutdown

520

Open Shortest Path First (OSPFv2 and OSPFv3)

25
PIM Sparse-Mode
PIM Sparse-Mode is supported on platforms:

ces

PIM-Sparse Mode (PIM-SM) is a multicast protocol that forwards multicast traffic to a subnet only upon request using a PIM Join message; this behavior is the opposite of PIM-Dense Mode, which forwards multicast traffic to all subnets until a request to stop.

Implementation Information
The Dell Networking implementation of PIM-SM is based on the IETF Internet Draft draft-ietf-pim-sm-v2-new-05. C-Series supports a maximum of 31 PIM interfaces and 4K multicast entries including (*,G), and (S,G) entries. There is no limit on the number of PIM neighbors C-Series can have. S-Series supports a maximum of 31 PIM interfaces and 2K multicast entries including (*,G), and (S,G) entries. There is no limit on the number of PIM neighbors S-Series can have. E-Series supports a maximum of 511 PIM interfaces and 50K multicast entries including (*,G), (S,G), and (S,G,rpt) entries. There is no limit on the number of PIM neighbors E-Series can have. The SPT-Threshold is zero, which means that the last-hop designated router (DR) joins the shortest path tree (SPT) to the source upon receiving the first multicast packet. FTOS reduces the number of control messages sent between multicast routers by bundling Join and Prune requests in the same message. FTOS supports PIM-SM on physical, VLAN, and port-channel interfaces. FTOS supports 2000 IPv6 multicast forwarding entries, with up to 128 PIM-SSM neighbors/interfaces. PIM-SM on VLAN interfaces is supported on the E-Series on TeraScale platforms only. IPv6 Multicast is not supported on SONET interfaces.

Protocol Overview
PIM-SM initially uses unidirectional shared trees to forward multicast traffic; that is, all multicast traffic must flow only from the Rendezvous Point (RP) to the receivers. Once a receiver receives traffic from the RP, PM-SM switches to shortest path trees (SPT) to forward multicast traffic. Every multicast group has an RP and a unidirectional shared tree (group-specific shared tree).

PIM Sparse-Mode | 521

www.dell.com | support.dell.com

Requesting Multicast Traffic

A host requesting multicast traffic for a particular group sends an IGMP Join message to its gateway router. The gateway router is then responsible for joining the shared tree to the RP (RPT) so that the host can receive the requested traffic. 1. Upon receiving an IGMP Join message, the receiver gateway router (last-hop DR) creates a (*,G) entry in its multicast routing table for the requested group. The interface on which the join message was received becomes the outgoing interface associated with the (*,G) entry. 2. The last-hop DR sends a PIM Join message to the RP. All routers along the way, including the RP, create an (*,G) entry in their multicast routing table, and the interface on which the message was received becomes the outgoing interface associated with the (*,G) entry. This process constructs an RPT branch to the RP. 3. If a host on the same subnet as another multicast receiver sends an IGMP report for the same multicast group, the gateway takes no action. If a router between the host and the RP receives a PIM Join message for which it already has a (*,G) entry, the interface on which the message was received is added to the outgoing interface list associated with the (*,G) entry, and the message is not (and does not need to be) forwarded towards the RP.

Refusing Multicast Traffic

A host requesting to leave a multicast group sends an IGMP Leave message to the last-hop DR. If the host is the only remaining receiver for that group on the subnet, the last-hop DR is responsible for sending a PIM Prune message up the RPT to prune its branch to the RP. 1. Upon receiving an IGMP Leave message, the gateway removes the interface on which it is received from the outgoing interface list of the (*,G) entry. If the (*,G) entry has no remaining outgoing interfaces, multicast traffic for that group is no longer forwarded to that subnet. 2. If the (*,G) entry has no remaining outgoing interfaces, the last-hop DR sends a PIM Prune message to towards the RP. All routers along the way remove the interface on which the message was received from the outgoing interface list of the (*,G) entry. If on any router there is at least one outgoing interface listed for that (*,G) entry, the Prune message is not forwarded.

Sending Multicast Traffic

With PIM-SM, all multicast traffic must initially originate from the RP. A source must unicast traffic to the RP so that the RP can learn about the source and create an SPT to it. Then the last-hop DR may create an SPT directly to the source. 1. The source gateway router (first-hop DR) receives the multicast packets and creates an (S,G) entry in its multicast routing table. The first-hop DR encapsulates the initial multicast packets in PIM Register packets and unicasts them to the RP. 2. The RP decapsulates the PIM Register packets and forwards them if there are any receivers for that group. The RP sends a PIM Join message towards the source. All routers between the RP and the

522

PIM Sparse-Mode

source, including the RP, create an (S,G) entry and list the interface on which the message was received as an outgoing interface, thus recreating a SPT to the source. 3. Once the RP starts receiving multicast traffic via the (S,G) it unicasts a Register-Stop message to the first-hop DR so that multicast packets are no longer encapsulated in PIM Register packets and unicast. Upon receiving the first multicast packet from a particular source, the last-hop DR sends a PIM Join message to the source to create an SPT to it. 4. There are two paths, then, between the receiver and the source, a direct SPT and an RPT. One router will receive a multicast packet on two interfaces from the same source in this case; this router prunes the shared tree by sending a PIM Prune message to the RP that tells all routers between the source and the RP to remove the outgoing interface from the (*,G) entry, and tells the RP to prune its SPT to the source with a Prune message.
FTOS Behavior: When the router creates an SPT to the source, there are then two paths between the receiver and the source, the SPT and the RPT. Until the router can prune itself from the RPT, the receiver receives duplicate multicast packets which may cause disruption. Therefore, the router must prune itself from the RPT as soon as possible. FTOS optimizes the shared to shortest-path tree switchover latency by copying and forwarding the first (S,G) packet received on the SPT to the PIM task immediately upon arrival. The arrival of the (S,G) packet confirms for PIM that the SPT is created, and that it can prune itself from the shared tree.

Important Points to Remember

If a loopback interface with a /32 mask is used as the RP, you must enable PIM Sparse-mode on the interface.

Configure PIM-SM
Configuring PIM-SM is a two-step process: 1. Enable multicast routing using the command ip multicast-routing from CONFIGURATION mode. 2. Select a Rendezvous Point. 3. Enable PIM-SM on an interface. See page 524.

Related Configuration Tasks

Configurable S,G Expiry Timers Configure a Static Rendezvous Point Configure a Designated Router Create Multicast Boundaries and Domains PIM-SM Graceful Restart Monitoring PIM

PIM Sparse-Mode | 523

www.dell.com | support.dell.com

Enable PIM-SM
You must enable PIM-SM on each participating interface:
Step 1 2 Task Enable multicast routing on the system. Enable PIM-Sparse Mode Command
ip multicast-routing ip pim sparse-mode

Command Mode CONFIGURATION INTERFACE

Display which interfaces are enabled with PIM-SM using the command show ip pim interface from EXEC Privilege mode, as shown in Figure 25-1.
Figure 25-1. Viewing PIM-SM Enabled Interfaces
FTOS#show ip pim interface Address Interface VIFindex Ver/ Mode 189.87.5.6 Gi 4/11 0x2 v2/S 189.87.3.2 Gi 4/12 0x3 v2/S 189.87.31.6 Gi 7/11 0x0 v2/S 189.87.50.6 Gi 7/13 0x4 v2/S FTOS# Nbr Count 1 1 0 1 Query Intvl 30 30 30 30 DR Prio 1 1 1 1 DR 127.87.5.6 127.87.3.5 127.87.31.6 127.87.50.6

Note: You can influence the selection of the Rendezvous Point by enabling PIM-Sparse Mode on a loopback interface and assigning a low IP address.

Display PIM neighbors for each interface using the command show ip pim neighbor from EXEC Privilege mode, as shown in Figure 25-2.
Figure 25-2. Viewing PIM Neighbors Command Example
FTOS#show ip pim neighbor Neighbor Interface Address 127.87.5.5 Gi 4/11 127.87.3.5 Gi 4/12 127.87.50.5 Gi 7/13 FTOS# Uptime/Expires 01:44:59/00:01:16 01:45:00/00:01:16 00:03:08/00:01:37 Ver v2 v2 v2 DR Prio/Mode 1 / S 1 / DR 1 / S

Display the PIM routing table using the command show ip pim tib from EXEC privilege mode, as shown in Figure 25-3.

524

PIM Sparse-Mode

Figure 25-3. Viewing the PIM Multicast Routing Table

FTOS#show ip pim tib PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode (*, 192.1.2.1), uptime 00:29:36, expires 00:03:26, RP 10.87.2.6, flags: SCJ Incoming interface: GigabitEthernet 4/12, RPF neighbor 10.87.3.5 Outgoing interface list: GigabitEthernet 4/11 GigabitEthernet 7/13 (10.87.31.5, 192.1.2.1), uptime 00:01:24, expires 00:02:26, flags: FT Incoming interface: GigabitEthernet 7/11, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 4/11 GigabitEthernet 4/12 GigabitEthernet 7/13 --More--

Configurable S,G Expiry Timers

By default S, G entries expire in 210 seconds. You can configure a global expiry time (for all (S,G) entries) or configure a expiry time for a particular entry. If both are configured, the ACL supercedes the global configuration for the specified entries. When an expiry time created, deleted, or updated, the changes are applied when the keep alive timer refreshes. To configure a global expiry time:
Task Enable global expiry timer for S, G entries Range 211-86400 seconds Default: 210 Command
ip pim sparse-mode sg-expiry-timer seconds

Command Mode CONFIGURATION

Configure the expiry time for a particular (S,G) entry:

Step 1 2 Task Create an Extended ACL Specify the source and group to which the timer will be applied using extended ACLs with permit rules only. Command Syntax
ip access-list extended access-list-name

Command Mode CONFIGURATION CONFIG-EXT-NACL

[seq sequence-number] permit ip source-address/ mask | any | host source-address} {destination-address/mask | any | host destination-address}

PIM Sparse-Mode | 525

www.dell.com | support.dell.com

Step 3

Task Set the expiry time for a specific (S,G) entry (Figure 25-4). Range 211-86400 seconds Default: 210

Command Syntax
ip pim sparse-mode sg-expiry-timer seconds sg-list access-list-name

Command Mode CONFIGURATION

Note: The expiry time configuration is nullified, and the default global expiry time is used if: an ACL is specified in the ip pim sparse-mode sg-expiry-timer command, but the ACL has not been created or is a standard ACL. if the expiry time is specified for an (S,G) entry in a deny rule. Figure 25-4. Configuring an (S,G) Expiry Time
FTOS(conf)#ip access-list extended SGtimer FTOS(config-ext-nacl)#permit ip 10.1.2.3/24 225.1.1.0/24 FTOS(config-ext-nacl)#permit ip any 232.1.1.0/24 FTOS(config-ext-nacl)#permit ip 100.1.1.0/16 any FTOS(config-ext-nacl)#show conf ! ip access-list extended SGtimer seq 5 permit ip 10.1.2.0/24 225.1.1.0/24 seq 10 permit ip any 232.1.1.0/24 seq 15 permit ip 100.1.0.0/16 any FTOS(config-ext-nacl)#exit FTOS(conf)#ip pim sparse-mode sg-expiry-timer 1800 sg-list SGtimer

Display the expiry time configuration using the show running-configuration [acl | pim] command from EXEC Privilege mode.

Configure a Static Rendezvous Point

The rendezvous point is a PIM-enabled interface on a router that acts as the root a group-specific tree; every group must have an RP. Identify an RP by the IP address of a PIM-enabled or loopback interface using the command ip pim rp-address, as shown in Figure 25-5.
Figure 25-5. Electing a Rendezvous Point
FTOS#sh run int loop0 ! interface Loopback 0 ip address 1.1.1.1/32 ip pim sparse-mode no shutdown FTOS#sh run pim ! ip pim rp-address 1.1.1.1 group-address 224.0.0.0/4

526

PIM Sparse-Mode

Override Bootstrap Router Updates

PIM-SM routers need to know the address of the RP for each group for which they have (*,G) entry. This address is obtained automatically through the bootstrap router (BSR) mechanism or a static RP configuration. If you have configured a static RP for a group, use the option override with the command ip pim rp-address to override bootstrap router updates with your static RP configuration. If you do not use this option, the RPs advertised in the BSR updates take precedence over any statically configured RPs. Display the assigned RP for a group using the command show ip pim rp from EXEC privilege mode, as shown in Figure 25-6.
Figure 25-6. Displaying the Rendezvous Point for a Multicast Group
FTOS#show ip pim rp Group RP 225.0.1.40 165.87.50.5 226.1.1.1 165.87.50.5

mapping

Display the assigned RP for a group range (group-to-RP mapping) using the command show ip pim rp command in EXEC privilege mode.
Figure 25-7. Display the Rendezvous Point for a Multicast Group Range
FTOS#show ip pim rp mapping PIM Group-to-RP Mappings Group(s): 224.0.0.0/4, Static RP: 165.87.50.5, v2

Configure a Designated Router

Multiple PIM-SM routers might be connected to a single LAN segment. One of these routers is elected to act on behalf of directly connected hosts. This router is the Designated Router (DR). The DR is elected using hello messages. Each PIM router learns about its neighbors by periodically sending a hello message out of each PIM-enabled interface. Hello messages contain the IP address of the interface out of which it is sent and a DR priority value. The router with the greatest priority value is the DR. If the priority value is the same for two routers, then the router with the greatest IP address is the DR. By default the DR priority value is 192, so the IP address determines the DR. Assign a DR priority value using the command ip pim dr-priority priority-value from INTERFACE mode. Change the interval at which a router sends hello messages using the command ip pim query-interval seconds from INTERFACE mode. Display the current value of these parameter using the command show ip pim interface EXEC Privilege mode.

PIM Sparse-Mode | 527

www.dell.com | support.dell.com

Create Multicast Boundaries and Domains

A PIM domain is a contiguous set of routers that all implement PIM and are configured to operate within a common boundary defined by PIM Multicast Border Routers (PMBRs). PMBRs connect each PIM domain to the rest of the internet. Create multicast boundaries and domains by filtering inbound and outbound Bootstrap Router (BSR) messages per interface, use the ip pim bsr-border command. This command is applied to the subsequent inbound and outbound updates. Already existing BSR advertisements are removed by timeout. Remove candidate RP advertisements using the clear ip pim rp-mapping command.

PIM-SM Graceful Restart

e PIM-SM Graceful Restart is supported only on platform ex with FTOS 8.2.1.0 and later.
PIM-SM Graceful Restart is supported only on platform When a PIM neighbor restarts and the liveliness timer for that neighbor expires, the join/prune states received from the neighbor expire, and the corresponding interfaces are removed from the outgoing list of multicast entries. The effect of this is that active multicast sessions are brought down. FTOS supports PIM-SM graceful restart based on the GenID. Per RFC 4601, hello messages should contain a Generation_Identifier option, which contains a randomly generated value (GenID) that is regenerated each time PIM forwarding is started or restarted on the interface, including when the router restarts. When a router receives from a neighbor a hello message with a new GenID, any old hello information about that neighbor should be discarded and superseded by the information from the new hello message. FTOS supports graceful restart based on the GenID. A Dell Networking PIM router announces its graceful restart capability to its neighbors up front as an option in its hello messages. If a graceful-restart capable router recognizes that a graceful-restart capable neighbor has restarted, it preserves the state from the neighbor and continues forwarding multicast traffic while the neighbor restarts. The router holds on to the entries learned from the neighbor for the graceful restart interval. If it does not receive a hello from the neighbor within this time, it purges all state associated with the neighbor. If the neighbor restarts and sends a hello with a new GenID before this interval expires, the router sends a join message towards the neighbor for the relevant entries.

If a graceful-restart capable router restarts, the router preserves all multicast entries in hardware until it receives and consolidates joins from its graceful-restart capable neighbors. The router is not taken off the forwarding path during restart.

528

PIM Sparse-Mode

Enable PIM-SM graceful restart (non-stop forwarding capability) using the command ip pim graceful-restart nsf from CONFIGURATION mode. There are two options with this command:
restart-time

is the time required by the Dell Networking system to restart. The default value is 180

seconds.
stale-entry-time

is the maximum amount of time that the Dell Networking system preserves entries from a restarting neighbor. The default value is 60 seconds.

In helper-only mode, the system preserves the PIM states of a neighboring router while the neighbor gracefully restarts, but the Dell Networking system allows itself to be taken off the forwarding path if it restarts. Enable this mode using the command ip pim graceful-restart helper-only. This mode takes precedence over any graceful restart configuration.

Monitoring PIM
The PIM MIB is supported only on platform

FTOS fully supports the PIM MIB as specified in RFC 5060 with some exceptions. The following tables are not supported: pimBidirDFElectionTable pimAnycastRPSetTable The OIDs related to InvalidRegisterMsgs reflect the last received invalid register message. Similarly, the OIDs related to InvalidJoinPruneMsgs reflect the last received invalid Join or Prune message. OIDs which refer to any timer show the time that the timer started; it is 0 otherwise.

PIM Sparse-Mode | 529

530 www.dell.com | support.dell.com

PIM Sparse-Mode

26
PIM Source-Specific Mode
PIM Source-Specific Mode is supported on platforms:

ces

PIM-Source-Specific Mode (PIM-SSM) is a multicast protocol that forwards multicast traffic from a single source to a subnet. In the other versions of Protocol Independent Multicast (PIM), a receiver subscribes to a group only. The receiver receives traffic not just from the source in which it is interested but from all sources sending to that group. PIM-SSM requires that receivers specify the sources in which they are interested using IGMPv3 include messages to avoid receiving unwanted traffic. PIM-SSM is more efficient than PIM-SM because it immediately creates shortest path trees (SPT) to the source rather than first using shared trees. PIM-SM requires a shared tree rooted at the RP because IGMPv2 receivers do not know about the source sending multicast data. Multicast traffic passes from the source to the receiver through the RP, until the receiver learns the source address, at which point it switches to the SPT. PIM-SSM uses IGMPv3. Since receivers subscribe to a source and group, the RP and shared tree is unnecessary, so only SPTs are used. On Dell Networking systems, it is possible to use PIM-SM with IGMPv3 to achieve the same result, but PIM-SSM eliminates the unnecessary protocol overhead. PIM-SSM also solves the multicast address allocation problem. Applications should use unique multicast addresses because if multiple applications use the same address, receivers receive unwanted traffic. However, global multicast address space is limited. Currently GLOP/EGLOP is used to statically assign Internet-routable multicast addresses, but each autonomous system number yields only 255 multicast addresses. For short-term applications, an address could be leased, but no global dynamic multicast address allocation scheme has been accepted yet. PIM-SSM eliminates the need for unique multicast addresses because routing decisions for (S1, G1) are independent from (S2, G1). As a result, subnets do not receive unwanted traffic when multiple applications use the same address. In Figure 26-1, Receiver 1 is an IGMPv2 host. The packets for group 239.0.0.2 travel to it first via the RP, then by the SPT. Receiver 2 is an IGMPv3 host. The packets for group 239.0.0.1 travel only via the STP.

PIM Source-Specific Mode | 531

532 www.dell.com | support.dell.com

R3(conf )#do show ip pim tib interface GigabitEthernet 3/21 ip pim sparse-mode ip address 10.11.23.2/24 no shutdown

|
Source 2 10.11.1.2
interface GigabitEthernet 2/1 ip pim sparse-mode ip address 10.11.1.1/24 no shutdown interface GigabitEthernet 3/1 ip pim sparse-mode ip address 10.11.5.1/24 no shutdown interface GigabitEthernet 2/31 ip pim sparse-mode ip address 10.11.23.1/24 no shutdown

PIM Source-Specific Mode

Source 1 10.11.5.2
PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode (10.11.5.2, 239.0.0.1), uptime 00:00:21, expires 00:03:14, flags: FT Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 3/11 Forward/Sparse 00:00:15/00:03:15

R2(conf )#do show ip pim tib

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

(*, 239.0.0.2), uptime 00:02:19, expires 00:03:13, RP 10.11.12.2, flags: S Incoming interface: Null, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 2/11 Forward/Sparse 00:02:19/00:03:13

(10.11.5.2, 239.0.0.2), uptime 00:00:44, expires 00:02:51, flags: P Incoming interface: GigabitEthernet 2/31, RPF neighbor 10.11.23.2 Outgoing interface list:

R2 3/1 2/31 R3
interface GigabitEthernet 2/11 ip pim sparse-mode ip address 10.11.12.2/24 no shutdown

2/1 3/21 RP 2/11 3/11 ip multicast-routing ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 router rip network 10.0.0.0

(10.11.5.2, 239.0.0.2), uptime 00:00:49, expires 00:03:04, flags: FT Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 3/11 Forward/Sparse 00:00:49/00:02:41

interface GigabitEthernet 3/11 ip pim sparse-mode ip address 10.11.13.2/24 no shutdown

R1(conf )#do show ip pim tib

interface GigabitEthernet 1/21 ip pim sparse-mode ip address 10.11.12.1/24 no shutdown

1/21 1/31
interface GigabitEthernet 1/31 ip pim sparse-mode ip address 10.11.13.1/24 no shutdown

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

R1

Figure 26-1. PIM-SM with IGMPv2 versus PIM-SM with IGMPv3

(10.11.5.2, 239.0.0.1), uptime 00:00:02, expires 00:00:00, flags: CJ Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:00:02/Never interface Vlan 300 ip pim sparse-mode ip address 10.11.3.1/24 untagged GigabitEthernet 1/1 no shutdown

(*, 239.0.0.2), uptime 00:02:12, expires 00:00:00, RP 10.11.12.2, flags: SCJ Incoming interface: GigabitEthernet 1/21, RPF neighbor 10.11.12.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:02:12/Never interface Vlan 400 ip pim sparse-mode ip address 10.11.4.1/24 untagged GigabitEthernet 1/2 ip igmp version 3 no shutdown

(10.11.5.2, 239.0.0.2), uptime 00:00:36, expires 00:03:14, flags: CT Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:02:12/Never ip igmp snooping enable

Receiver 1 10.11.3.2 Group: 239.0.0.2

Receiver 2 10.11.4.2 Group: 239.0.0.1 Source: 10.11.5.2

Implementation Information
The Dell Networking implementation of PIM-SSM is based on RFC 3569. C-Series supports a maximum of 31 PIM interfaces and 4K multicast entries including (*,G), and (S,G) entries. There is no limit on the number of PIM neighbors C-Series can have. S-Series supports a maximum of 31 PIM interfaces and 2K multicast entries including (*,G), and (S,G) entries. There is no limit on the number of PIM neighbors S-Series can have. E-Series supports a maximum of 511 PIM interfaces and 50K multicast entries including (*,G), (S,G), and (S,G,rpt) entries. There is no limit on the number of PIM neighbors E-Series can have. FTOS reduces the number of control messages sent between multicast routers by bundling Join and Prune requests in the same message.

Important Points to Remember

The default SSM range is 232/8 always. Applying an SSM range does not overwrite the default range. Both the default range and SSM range are effective even when the default range is not added to the SSM ACL. Extended ACLs cannot be used for configuring SSM range. Be sure to create the ACL first and then apply it to the SSM range. The default range is always supported, so range can never be smaller than the default.

Configure PIM-SM
Configuring PIM-SSM is a one-step process: 1. Configure PIM-SM. See page 521. 2. Enable PIM-SSM for a range of addresses. See page 534.

Related Configuration Tasks

Use PIM-SSM with IGMP version 2 Hosts

PIM Source-Specific Mode | 533

www.dell.com | support.dell.com

Enable PIM-SSM
To enable PIM-SSM:
Step 1 Task Create an ACL that uses permit rules to specify what range of addresses should use SSM. You must at least include one rule, permit 232.0.0.0/8, which is the default range for PIM-SSM. Enter the command ip pim ssm-range and specify the ACL you created. Command Syntax
ip access-list standard name

Command Mode CONFIGURATION

ip pim ssm-range acl-name

CONFIGURATION

Display address ranges in the PIM-SSM range using the command show ip pim ssm-range from EXEC Privilege mode.
Figure 26-2. Enabling PIM-SSM
R1(conf)#do show run pim ! ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 ip pim ssm-range ssm R1(conf)#do show run acl ! ip access-list standard ssm seq 5 permit host 239.0.0.2 R1(conf)#do show ip pim ssm-range Group Address / MaskLen 239.0.0.2 / 32

Use PIM-SSM with IGMP version 2 Hosts

PIM-SSM requires receivers that support IGMP version 3. You can employ PIM-SSM even when receivers support only IGMP version 1 or version 2 by translating (*,G) entries to (S,G) entries. Translate (*,G) entries to (S,G) entries using the command ip igmp ssm-map acl source from CONFIGURATION mode. In a standard access list, specify the groups or the group ranges that you want to map to a source. Then, specify the multicast source. When a SSM map is in place and FTOS cannot find any matching access lists for a group, it continues to create (*,G) entries because there is an implicit deny for unspecified groups in the ACL. When you remove the mapping configuration, FTOS removes the corresponding (S,G) states that it created and reestablishes the original (*,G) states. You may enter multiple ssm-map commands for different access lists. You may also enter multiple ssm-map commands for the same access list, as long as they use different source addresses.

534

PIM Source-Specific Mode

When an extended ACL is associated with this command, FTOS displays an error message. If you apply an extended ACL before you create it, FTOS accepts the configuration, but when the ACL is later defined, FTOS ignores the ACL and the stated mapping has no effect.

Display the source to which a group is mapped using the command show ip igmp ssm-map [group], as shown in Figure 26-4 on page 537. If use the group option, the command displays the group-to-source mapping even if the group is not currently in the IGMP group table. If you do not specify the group option, then the display is a list of groups currently in the IGMP group table that have a group-to-source mapping. Display the list of sources mapped to a group currently in the IGMP group table using the command show ip igmp groups group detail, as shown in Figure 26-4 on page 537.

PIM Source-Specific Mode | 535

536 www.dell.com | support.dell.com

R3(conf )#do show ip pim tib interface GigabitEthernet 3/21 ip pim sparse-mode ip address 10.11.23.2/24 no shutdown

|
Source 2 10.11.1.2
interface GigabitEthernet 2/1 ip pim sparse-mode ip address 10.11.1.1/24 no shutdown interface GigabitEthernet 3/1 ip pim sparse-mode ip address 10.11.5.1/24 no shutdown interface GigabitEthernet 2/31 ip pim sparse-mode ip address 10.11.23.1/24 no shutdown

PIM Source-Specific Mode

Source 1 10.11.5.2
PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode (10.11.5.2, 239.0.0.1), uptime 00:01:34, expires 00:02:58, flags: FT Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Registering Outgoing interface list: GigabitEthernet 3/11 Forward/Sparse 00:01:34/00:03:01 (10.11.5.2, 239.0.0.2), uptime 00:00:17, expires 00:00:00, flags: FJ Incoming interface: GigabitEthernet 3/1, RPF neighbor 0.0.0.0 Outgoing interface list: GigabitEthernet 3/11 Forward/Sparse 00:00:17/00:03:

R2(conf )#do show ip pim tib

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

R2 3/1 2/31 R3
interface GigabitEthernet 2/11 ip pim sparse-mode ip address 10.11.12.2/24 no shutdown

2/1 3/21 RP 2/11 3/11 ip multicast-routing ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 router rip network 10.0.0.0
interface GigabitEthernet 1/21 ip pim sparse-mode ip address 10.11.12.1/24 no shutdown

interface GigabitEthernet 3/11 ip pim sparse-mode ip address 10.11.13.2/24 no shutdown

R3(conf )#do show run pim ! ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 ip pim ssm-range ssm R1(conf )#do show run acl ! ip access-list standard ssm seq 5 permit host 239.0.0.2

R1(conf )#do show ip pim tib

1/21 1/31
interface GigabitEthernet 1/31 ip pim sparse-mode ip address 10.11.13.1/24 no shutdown

PIM Multicast Routing Table Flags: D - Dense, S - Sparse, C - Connected, L - Local, P - Pruned, R - RP-bit set, F - Register flag, T - SPT-bit set, J - Join SPT, M - MSDP created entry, A - Candidate for MSDP Advertisement K - Ack-Pending State Timers: Uptime/Expires Interface state: Interface, next-Hop, State/Mode

R1

(10.11.5.2, 239.0.0.1), uptime 00:01:50, expires 00:03:28, flags: CT Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 400 Forward/Sparse 00:01:50/Never interface Vlan 400 ip pim sparse-mode ip address 10.11.4.1/24 untagged GigabitEthernet 1/2 ip igmp version 3 no shutdown

Figure 26-3. Using PIM-SM with IGMPv2 versus PIM-SSM with IGMPv2

(10.11.5.2, 239.0.0.2), uptime 00:00:33, expires 00:00:00, flags: CJ Incoming interface: GigabitEthernet 1/31, RPF neighbor 10.11.13.2 Outgoing interface list: Vlan 300 Forward/Sparse 00:00:33/Never

interface Vlan 300 ip pim sparse-mode ip address 10.11.3.1/24 untagged GigabitEthernet 1/1 no shutdown

R1(conf )#do show run pim ! ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 ip pim ssm-range ssm R1(conf )#do show run acl ! ip access-list standard map seq 5 permit host 239.0.0.2 ! ip access-list standard ssm seq 5 permit host 239.0.0.2

ip igmp snooping enable

Receiver 1 10.11.3.2 Group: 239.0.0.2 Source: 10.11.5.2 Receiver 2 10.11.4.2 Group: 239.0.0.1

Figure 26-4. Configuring PIM-SSM with IGMPv2

R1(conf)#do show run pim ! ip pim rp-address 10.11.12.2 group-address 224.0.0.0/4 ip pim ssm-range ssm R1(conf)#do show run acl ! ip access-list standard map seq 5 permit host 239.0.0.2 ! ip access-list standard ssm seq 5 permit host 239.0.0.2 R1(conf)#ip igmp ssm-map map 10.11.5.2 R1(conf)#do show ip igmp groups Total Number of Groups: 2 IGMP Connected Group Membership Group Address Interface Mode 239.0.0.2 Vlan 300 IGMPv2-Compat Member Ports: Gi 1/1 239.0.0.1 Vlan 400 INCLUDE R1(conf)#do show ip igmp ssm-map IGMP Connected Group Membership Group Address Interface Mode 239.0.0.2 Vlan 300 IGMPv2-Compat Member Ports: Gi 1/1 R1(conf)#do show ip igmp ssm-map 239.0.0.2 SSM Map Information Group : 239.0.0.2 Source(s) : 10.11.5.2 R1(conf)#do show ip igmp groups detail Interface Group Uptime Expires Router mode Last reporter Last reporter mode Last report received Group source list Source address 10.11.5.2 Vlan 300 239.0.0.2 00:00:01 Never IGMPv2-Compat 10.11.3.2 IGMPv2 Join Uptime 00:00:01 Expires Never

Uptime 00:00:07 00:00:10

Expires Never Never

Last Reporter 10.11.3.2 10.11.4.2

Uptime 00:00:36

Expires Never

Last Reporter 10.11.3.2

Interface Vlan 400 Group 239.0.0.1 Uptime 00:00:05 Expires Never Router mode INCLUDE Last reporter 10.11.4.2 Last reporter mode INCLUDE Last report received ALLOW Group source list Source address 10.11.5.2 Member Ports: Gi 1/2

Uptime 00:00:05

Expires 00:02:04

PIM Source-Specific Mode | 537

538 www.dell.com | support.dell.com

PIM Source-Specific Mode

27
Port Monitoring
Port Monitoring is supported on platforms:

ces

Port Monitoring is a feature that copies all incoming or outgoing packets on one port and forwards (mirrors) them to another port. The source port is the monitored port (MD) and the destination port is the monitoring port (MG). Port Monitoring functionality is different between platforms, but the behavior is the same, with highlighted exceptions. This chapter is divided into the following sections: Important Points to Remember Port Monitoring on E-Series Port Monitoring on C-Series and S-Series Configuring Port Monitoring Flow-based Monitoring

Important Points to Remember

Port Monitoring is supported on physical ports only; VLAN and port-channel interfaces do not support port monitoring. A SONET port may only be a monitored port. The Monitored (source, MD) and Monitoring ports (destination, MG) must be on the same switch. In general, a monitoring port should have no ip address and no shutdown as the only configuration; FTOS permits a limited set of commands for monitoring ports; display them using the command ?. A monitoring port also may not be a member of a VLAN. There may only be one destination port in a monitoring session. A source port (MD) can only be monitored by one destination port (MG). The following error is displayed if you try to assign a monitored port to more than one monitoring port.
FTOS(conf)#mon ses 1 FTOS(conf-mon-sess-1)#$gig 0/0 destination gig 0/60 direction both FTOS(conf-mon-sess-1)#do show mon ses SessionID Source Destination Direction Mode ----------------------------------1 Gi 0/0 Gi 0/60 both interface FTOS(conf-mon-sess-1)#mon ses 2 FTOS(conf-mon-sess-2)#source gig 0/0 destination gig 0/61 direction both % Error: MD port is already being monitored.

Type ---Port-based

Port Monitoring | 539

www.dell.com | support.dell.com

The C-Series and S-Series may only have four destination ports per port-pipe. There is no limitation on the total number of monitoring sessions.

Table 27-1 lists the maximum number of monitoring sessions per system. For the C-Series and S-Series, the total number of sessions is derived by consuming a unique destination port in each session, in each port-pipe.
Table 27-1. System C150 C300 S50V, S50N S25P Maximum Number of Monitoring Sessions per System Maximum Sessions System E1200/E1200i (TeraScale) E1200i (ExaScale) E600/E600i (TeraScale) E600i (ExaScale) E300 Maximum Sessions 28

(Note) (Note) (Note) (Note)

14

Note: On the C-Series and S-Series, there is no limit to the number of monitoring sessions per system, provided that there are only 4 destination ports per port-pipe. If each monitoring session has a unique destination port, then the maximum number of session is 4 per port-pipe.

Port Monitoring on E-Series

Both the E-Series TeraScale and E-Series ExaScale support the following. FTOS supports one destination (MG) port per monitoring session. The same destination port (MG) can be used in another monitoring session. One destination (MG) port can monitor up to 28 source (MD) ports. A port cannot be defined as both a source (MD) and a destination (MG) port (Message 1).

Message 1 Cannot define source (MD) and destination (MG) on same port
% Error: MD port is already being monitored.

E-Series TeraScale
The E-Series TeraScale system supports 1 monitoring session per port-pipe. E-Series TeraScale supports a maximum of 28 port pipes.

540

Port Monitoring

On the E-Series TeraScale, FTOS supports a single source-destination statement in a monitor session (Message 2). E-Series TeraScale supports only one source and one destination port per port-pipe (Message 3). Therefore, the E-Series TeraScale supports as many monitoring sessions as there are port-pipes in the system.
Message 2 Multiple Source-Destination Statements Error Message on E-Series TeraScale
% Error: Remove existing monitor configuration.

Message 3 One Source/Destination Port per Port-pipe Error Message on E-Series TeraScale
% Error: Some port from this port pipe is already configured as MD. % Error: Some port from this port pipe is already configured as MG.

Figure 27-1 illustrates a possible port monitoring configuration on the E-Series.

Figure 27-1. Port Monitoring Configurations on the E-Series
Line Card 0 Port-Pipe 0 Monitor Session 0 Monitor Session 1 Monitor Session 2 Monitor Session 3 MD MG MD MD MD
Port Monitoring 002

Line Card 1 Port-Pipe 0 Port-Pipe 1

Port-Pipe 1

MG

E-Series ExaScale
FTOS on E-Series ExaScale supports a single destination (MG) port monitoring multiple multiple source (MD) ports in one monitor session. One monitor session can have only one destination (MG) port. The same destination (MG) port can be uses with multiple monitoring sessions. There is no restriction on the number of source (MD) or destination (MG) ports on the chassis because there is no port-pipe restriction on the E-Series ExaScale system. There is no restriction to the number of monitoring sessions supported on the E-Series ExaScale system.

Port Monitoring on C-Series and S-Series

The C-Series and S-Series support multiple source-destination statements in a monitor session, but there may only be one destination port in a monitoring session (Message 4).
Message 4 One Destination Port in a Monitoring Session Error Message on C-Series and S-Series
% Error: Only one MG port is allowed in a session.

Port Monitoring | 541

www.dell.com | support.dell.com

The number of source ports FTOS allows within a port-pipe is equal to the number of physical ports in the port-pipe (n). However, n number of ports may only have four different destination ports (Message 5).
Figure 27-2. Number of Monitoring Ports on the C-Series and S-Series
FTOS#show mon session SessionID Source Destination Direction Mode ----------------------------------0 Gi 0/13 Gi 0/1 rx interface 10 Gi 0/14 Gi 0/2 rx interface 20 Gi 0/15 Gi 0/3 rx interface 30 Gi 0/16 Gi 0/37 rx interface FTOS(conf)#mon ses 300 FTOS(conf-mon-sess-300)#source gig 0/17 destination gig 0/4 direction tx % Error: Exceeding max MG ports for this MD port pipe. FTOS(conf-mon-sess-300)# FTOS(conf-mon-sess-300)#source gig 0/17 destination gig 0/1 direction tx FTOS(conf-mon-sess-300)#do show mon session SessionID Source Destination Direction Mode ----------------------------------0 Gi 0/13 Gi 0/1 rx interface 10 Gi 0/14 Gi 0/2 rx interface 20 Gi 0/15 Gi 0/3 rx interface 30 Gi 0/16 Gi 0/37 rx interface 300 Gi 0/17 Gi 0/1 tx interface FTOS(conf-mon-sess-300)# Type ---Port-based Port-based Port-based Port-based

Type ---Port-based Port-based Port-based Port-based Port-based

In Figure 27-2, ports 0/13, 0/14, 0/15, and 0/16 all belong to the same port-pipe. They are pointing to four different destinations (0/1, 0/2, 0/3, and 0/37). Now it is not possible for another source port from the same port-pipe (for example, 0/17) to point to another new destination (for example, 0/4). If you attempt to configure another destination, Message 5 appears. However, you can configure another monitoring session that uses one of previously used destination ports, as shown in Figure 27-3.
Figure 27-3. Number of Monitoring Ports on the C-Series and S-Series
FTOS(conf)#mon ses 300 FTOS(conf-mon-sess-300)#source gig 0/17 destination gig 0/4 direction tx % Error: Exceeding max MG ports for this MD port pipe. FTOS(conf-mon-sess-300)# FTOS(conf-mon-sess-300)#source gig 0/17 destination gig 0/1 direction tx FTOS(conf-mon-sess-300)#do show mon session SessionID Source Destination Direction Mode ----------------------------------0 Gi 0/13 Gi 0/1 rx interface 10 Gi 0/14 Gi 0/2 rx interface 20 Gi 0/15 Gi 0/3 rx interface 30 Gi 0/16 Gi 0/37 rx interface 300 Gi 0/17 Gi 0/1 tx interface

Type ---Port-based Port-based Port-based Port-based Port-based

In Figure 27-4, 0/25 and 0/26 belong to Port-pipe 1. This port-pipe again has the same restriction of only four destination ports, new or used.

542

Port Monitoring

Figure 27-4. Number of Monitoring Ports on the C-Series and S-Series

FTOS(conf-mon-sess-300)#do show mon session SessionID Source Destination -----------------------0 Gi 0/13 Gi 0/1 10 Gi 0/14 Gi 0/2 20 Gi 0/15 Gi 0/3 30 Gi 0/16 Gi 0/37 100 Gi 0/25 Gi 0/38 110 Gi 0/26 Gi 0/39 300 Gi 0/17 Gi 0/1 FTOS(conf-mon-sess-300)# Direction --------rx rx rx rx tx tx tx Mode ---interface interface interface interface interface interface interface Type ---Port-based Port-based Port-based Port-based Port-based Port-based Port-based

A source port may only be monitored by one destination port (Message 6), but a destination port may monitor more than one source port. Given these parameters, Figure 27-1 illustrates conceptually the possible port monitoring configurations on the C-Series and S-Series.
Message 5 One Destination Port in a Monitoring Session Error Message on C-Series and S-Series
% Error: Exceeding max MG ports for this MD port pipe.

Message 6 One Destination Port per Source Port Error Message

% Error: MD port is already being monitored.

Figure 27-5. Port Monitoring Configurations on the C-Series and S-Series

Line Card 0 Port-Pipe 0 Monitor Session 0 Monitor Session 1 MD MD Monitor Session 2 MD MG MG MG
Port Monitoring 003

Line Card 1 Port-Pipe 0 Port-Pipe 1

Port-Pipe 1

MD MD

MG MD

FTOS Behavior: On the C-Series and S-Series, all monitored frames are tagged if the configured monitoring direction is transmit (TX), regardless of whether the monitored port (MD) is a Layer 2 or Layer 3 port. If the MD port is a Layer 2 port, the frames are tagged with the VLAN ID of the VLAN to which the MD belongs. If the MD port is a Layer 3 port, the frames are tagged with VLAN ID 4095. If the MD port is in a Layer 3 VLAN, the frames are tagged with the respective Layer 3 VLAN ID. For example, in the configuration source gig 6/0 destination gig 6/1 direction tx, if the MD port gigabitethernet 6/0 is an untagged member of any VLAN, all monitored frames that the MG port gigabitethernet 6/1 receives are tagged with the VLAN ID of the MD port. Similarly, if BPDUs are transmitted, the MG port receives them tagged with the VLAN ID 4095. This behavior might result in a difference between the number of egress packets on the MD port and monitored packets on the MG port.

Port Monitoring | 543

www.dell.com | support.dell.com

FTOS Behavior: The C-Series and S-Series continue to mirror outgoing traffic even after an MD participating in Spanning Tree Protocol transitions from the forwarding to blocking.

Configuring Port Monitoring

To configure port monitoring:
Step 1 Task Verify that the intended monitoring port has no configuration other than no shutdown, as shown in Figure 27-6. Create a monitoring session using the command monitor session from CONFIGURATION mode, as shown in Figure 27-6. Specify the source and destination port and direction of traffic, as shown in Figure 27-6. Command Syntax
show interface

Command Mode EXEC Privilege

monitor session

CONFIGURATION

source

MONITOR SESSION

Display monitor sessions using the command show monitor session from EXEC Privilege mode, as shown in Figure 27-6.
Figure 27-6. Configuring Port-based Monitoring
FTOS(conf-if-gi-1/2)#show config ! interface GigabitEthernet 1/2 no ip address no shutdown FTOS(conf-if-gi-1/2)#exit FTOS(conf)#monitor session 0 FTOS(conf-mon-sess-0)#source gig 1/1 dest gig 1/2 direction rx FTOS(conf-mon-sess-0)#exit FTOS(conf)#do show monitor session 0 SessionID Source Destination Direction -------------------------------0 Gi 1/1 Gi 1/2 rx FTOS(conf)#

Mode ---interface

Type ---Port-based

In Figure 27-7, the host and server are exchanging traffic which passes through interface gigabitethernet 1/ 1. Interface gigabitethernet 1/1 is the monitored port and gigabitethernet 1/2 is the monitoring port, which is configured to only monitor traffic received on gigabitethernet 1/1 (host-originated traffic).

544

Port Monitoring

Figure 27-7. Port Monitoring Example

Host Traffic 1/1 Server Traffic 1/2 Host Server 1/3

FTOS(conf-if-gi-1/2)#show config ! interface GigabitEthernet 1/2 no ip address no shutdown Sniffer FTOS(conf )#monitor session 0 FTOS(conf-mon-sess-0)#source gig 1/1 destination gig 1/2 direction rx

Port Monitoring | 545

www.dell.com | support.dell.com

Flow-based Monitoring
Flow-based Monitoring is supported only on platform

Flow-based monitoring conserves bandwidth by monitoring only specified traffic instead all traffic on the interface. This feature is particularly useful when looking for malicious traffic. It is available for Layer 2 and Layer 3 ingress and egress traffic. You may specify traffic using standard or extended access-lists. To configure flow-based monitoring:
Step 4 5 Task Enable flow-based monitoring for a monitoring session. Define in an access-list rules that include the keyword
monitor. FTOS only considers for port monitoring traffic matching rules with the keyword monitor.

Command Syntax
flow-based enable ip access-list

Command Mode MONITOR SESSION CONFIGURATION

See Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps. 6 Apply the ACL to the monitored port. See Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps.
ip access-group access-list

INTERFACE

View an access-list that you applied to an interface using the command show ip accounting access-list from EXEC Privilege mode, as shown in Figure 27-8.

546

Port Monitoring

Figure 27-8. Configuring Flow-based Monitoring

FTOS(conf)#monitor session 0 FTOS(conf-mon-sess-0)#flow-based enable FTOS(conf)#ip access-list ext testflow FTOS(config-ext-nacl)#seq 5 permit icmp any any count bytes monitor FTOS(config-ext-nacl)#seq 10 permit ip 102.1.1.0/24 any count bytes monitor FTOS(config-ext-nacl)#seq 15 deny udp any any count bytes FTOS(config-ext-nacl)#seq 20 deny tcp any any count bytes FTOS(config-ext-nacl)#exit FTOS(conf)#interface gig 1/1 FTOS(conf-if-gi-1/1)#ip access-group testflow in FTOS(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 ip address 10.11.1.254/24 ip access-group testflow in shutdown FTOS(conf-if-gi-1/1)#exit FTOS(conf)#do show ip accounting access-list testflow ! Extended Ingress IP access list testflow on GigabitEthernet 1/1 Total cam count 4 seq 5 permit icmp any any monitor count bytes (0 packets 0 bytes) seq 10 permit ip 102.1.1.0/24 any monitor count bytes (0 packets 0 bytes) seq 15 deny udp any any count bytes (0 packets 0 bytes) seq 20 deny tcp any any count bytes (0 packets 0 bytes) FTOS(conf)#do show monitor session 0 SessionID Source Destination Direction Mode ----------------------------------0 Gi 1/1 Gi 1/2 rx interface

Type ---Flow-based

Port Monitoring | 547

548 www.dell.com | support.dell.com

Port Monitoring

28
Private VLANs
FTOS 7.8.1.0 adds a Private VLAN (PVLAN) feature for the C-Series and S-Series:

cs

For syntax details on the commands discussed in this chapter, see the Private VLANs Commands chapter in the FTOS Command Reference. This chapter contains the following major sections: Private VLAN Concepts Private VLAN Commands Private VLAN Configuration Task List Private VLAN Configuration Example Inspecting the Private VLAN Configuration

Private VLANs extend the FTOS security suite by providing Layer 2 isolation between ports within the same VLAN. A private VLAN partitions a traditional VLAN into subdomains identified by a primary and secondary VLAN pair. Private VLANs block all traffic to isolated ports except traffic from promiscuous ports. Traffic received from an isolated port is forwarded only to promiscuous ports or trunk ports. Example uses of PVLANs: A hotel can use an isolated VLAN in a private VLAN to provide Internet access for its guests, while stopping direct access between the guest ports. A service provider can provide Layer 2 security for customers and use the IP addresses more efficiently, by using a separate community VLAN per customer, while at the same time using the same IP subnet address space for all community and isolated VLANs mapped to the same primary VLAN. In more detail, community VLANs are especially useful in the service provider environment, because, multiple customers are likely to maintain servers that must be strictly separated in customer-specific groups. A set of servers owned by a customer could comprise a community VLAN, so that those servers could communicate with each other, and would be isolated from other customers. Another customer might have another set of servers in another community VLAN. Another customer might want an isolated VLAN, which is has one or more ports that are also isolated from each other.

Private VLANs | 549

www.dell.com | support.dell.com

Private VLAN Concepts

The VLAN types in a private VLAN (PVLAN) include: Community VLAN A community VLAN is a type of secondary VLAN in a primary VLAN: Ports in a community VLAN can communicate with each other. Ports in a community VLAN can communicate with all promiscuous ports in the primary VLAN. A community VLAN can only contain ports configured as host.

Isolated VLAN An isolated VLAN is a type of secondary VLAN in a primary VLAN: Ports in an isolated VLAN cannot talk directly to each other. Ports in an isolated VLAN can only communicate with promiscuous ports in the primary VLAN. An isolated VLAN can only contain ports configured as host.

Primary VLANA primary VLAN is the base VLAN of a private VLAN: A switch can have one or more primary VLANs, and it can have none. A primary VLAN has one or more secondary VLANs. A primary VLAN and each of its secondary VLANs decrement the available number of VLAN IDs in the switch. A primary VLAN has one or more promiscuous ports. A primary VLAN might have one or more trunk ports, or none.

Secondary VLAN A secondary VLAN is a subdomain of the primary VLAN. There are two types of secondary VLAN community VLAN and isolated VLAN. PVLAN port types: Community port: A community port is, by definition, a port that belongs to a community VLAN and is allowed to communicate with other ports in the same community VLAN and with promiscuous ports. Host port: A host port, in the context of a private VLAN, is a port in a secondary VLAN: The port must first be assigned that role in INTERFACE mode. A port assigned the host role cannot be added to a regular VLAN. Isolated port: An isolated port is, by definition, a port that, in Layer 2, can only communicate with promiscuous ports that are in the same PVLAN. Promiscuous port: A promiscuous port is, by definition, a port that is allowed to communicate with any other port type in the PVLAN: A promiscuous port can be part of more than one primary VLAN. A promiscuous port cannot be added to a regular VLAN. Trunk port: A trunk port, by definition, carries traffic between switches: A trunk port in a PVLAN is always tagged. Primary or secondary VLAN traffic is carried by the trunk port in tagged mode. The tag on the packet helps identify the VLAN to which the packet belongs. A trunk port can also belong to a regular VLAN (non-private VLAN).

550

Private VLANs

Each of the port types can be any type of physical Ethernet port, including port channels (LAGs). For details on port channels, see Port Channel Interfaces in Chapter 15, Interfaces. For an introduction to VLANs, see Chapter 20, Layer 2.

Private VLAN Commands

The commands dedicated to supporting the Private VLANs feature are:
Table 28-1. Task Enable/disable Layer 3 communication between secondary VLANs. Private VLAN Commands Command Syntax [no] ip local-proxy-arp Note: Even after ip-local-proxy-arp is disabled (no ip-local-proxy-arp) in a secondary VLAN, Layer 3 communication may happen between some secondary VLAN hosts, until the ARP timeout happens on those secondary VLAN hosts. [no] private-vlan mode {community | isolated | primary} [no] private-vlan mapping secondary-vlan vlan-list
show interfaces private-vlan [interface interface] show vlan private-vlan [community | interface | isolated | primary | primary_vlan | interface interface] show vlan private-vlan mapping

Command Mode INTERFACE VLAN

Set the mode of the selected VLAN to community, isolated, or primary. Map secondary VLANs to the selected primary VLAN. Display type and status of PVLAN interfaces. Display PVLANs and/or interfaces that are part of a PVLAN. Display primary-secondary VLAN mapping. Set the PVLAN mode of the selected port.

INTERFACE VLAN INTERFACE VLAN EXEC EXEC Privilege EXEC EXEC Privilege EXEC EXEC Privilege INTERFACE

switchport mode private-vlan {host | promiscuous | trunk}

Note: Secondary VLANs are Layer 2 VLANs, so even if they are operationally down while primary VLANs are operationally up, Layer 3 traffic will still be transmitted across secondary VLANs.

The outputs of the following commands are augmented in FTOS 7.8.1.0 to provide PVLAN data:
show arp:

See the IP Routing Commands chapter in the FTOS Command Reference. show vlan: See the Layer 2 Commands chapter in the FTOS Command Reference.

Private VLANs | 551

www.dell.com | support.dell.com

Private VLAN Configuration Task List

The following sections contain the procedures that configure a private VLAN: Creating PVLAN ports Creating a Primary VLAN Creating a Community VLAN Creating an Isolated VLAN

Creating PVLAN ports

Private VLAN ports are those that will be assigned to the private VLAN (PVLAN).
Step 1 Command Syntax
interface interface no shutdown
switchport switchport mode private-vlan {host | promiscuous | trunk}

Command Mode CONFIGURATION INTERFACE INTERFACE INTERFACE

Purpose Access the INTERFACE mode for the port that you want to assign to a PVLAN. Enable the port. Set the port in Layer 2 mode. Select the PVLAN mode: host (port in isolated or community VLAN) promiscuous (intra-VLAN communication port) trunk (inter-switch PVLAN hub port)

2 3 4

For interface details, see Enable a Physical Interface in Chapter 15, Interfaces.
Note: Interfaces that are configured as PVLAN ports cannot be added to regular VLANs. Conversely, regular ports (ports not configured as PVLAN ports) cannot be added to PVLANs.

Figure 28-1 shows the use of the switchport mode private-vlan command on a port and on a port channel:
Figure 28-1. Examples of switchport mode private-vlan Command
FTOS#conf FTOS(conf)#interface GigabitEthernet 2/1 FTOS(conf-if-gi-2/1)#switchport mode private-vlan promiscuous FTOS(conf)#interface GigabitEthernet 2/2 FTOS(conf-if-gi-2/2)#switchport mode private-vlan host FTOS(conf)#interface GigabitEthernet 2/3 FTOS(conf-if-gi-2/3)#switchport mode private-vlan trunk FTOS(conf)#interface GigabitEthernet 2/2 FTOS(conf-if-gi-2/2)#switchport mode private-vlan host

552

Private VLANs

Creating a Primary VLAN

A primary VLAN is a port-based VLAN that is specifically enabled as a primary VLAN to contain the promiscuous ports and PVLAN trunk ports for the private VLAN. A primary VLAN also contains a mapping to secondary VLANs, which are comprised of community VLANs and isolated VLANs.
Step 1 Command Syntax
interface vlan vlan-id no shutdown
private-vlan mode primary private-vlan mapping secondary-vlan vlan-list

Command Mode CONFIGURATION INTERFACE VLAN INTERFACE VLAN INTERFACE VLAN

Purpose Access the INTERFACE VLAN mode for the VLAN to which you want to assign the PVLAN interfaces. Enable the VLAN. Set the PVLAN mode of the selected VLAN to primary. Map secondary VLANs to the selected primary VLAN. The list of secondary VLANs can be: Specified in comma-delimited (VLAN-ID,VLAN-ID) or hyphenated-range format (VLAN-ID-VLAN-ID). Specified with this command even before they have been created. Amended by specifying the new secondary VLAN to be added to the list. Add promiscuous ports as tagged or untagged interfaces. Add PVLAN trunk ports to the VLAN only as tagged interfaces. Interfaces can be entered singly or in range format, either comma-delimited (slot/port,port,port) or hyphenated (slot/port-port). Only promiscuous ports or PVLAN trunk ports can be added to the PVLAN (no host or regular ports). (OPTIONAL) Assign an IP address to the VLAN. (OPTIONAL) Enable/disable Layer 3 communication between secondary VLANs.

2 3 4

tagged interface or untagged interface

INTERFACE VLAN

6 7

ip address ip address ip local-proxy-arp

INTERFACE VLAN INTERFACE VLAN

Note: If a promiscuous or host port is untagged in a VLAN and it receives a tagged packet in the same VLAN, the packet will NOT be dropped.

Private VLANs | 553

www.dell.com | support.dell.com

Creating a Community VLAN

A community VLAN is a secondary VLAN of the primary VLAN in a private VLAN. The ports in a community VLAN can talk to each other and with the promiscuous ports in the primary VLAN.
Step 1 Command Syntax
interface vlan vlan-id

Command Mode CONFIGURATION INTERFACE VLAN INTERFACE VLAN INTERFACE VLAN

Purpose Access the INTERFACE VLAN mode for the VLAN that you want to make a community VLAN. Enable the VLAN. Set the PVLAN mode of the selected VLAN to community. Add one or more host ports to the VLAN. The interfaces can be entered singly or in range format, either comma-delimited (slot/port,port,port) or hyphenated (slot/
port-port).

2 3 4

no shutdown private-vlan mode community tagged interface

or
untagged interface

Only host (isolated) ports can be added to the VLAN.

Creating an Isolated VLAN

An isolated VLAN is a secondary VLAN of a primary VLAN. Its ports can only talk with the promiscuous ports in that primary VLAN.
Step 1 Command Syntax
interface vlan vlan-id

Command Mode CONFIGURATION INTERFACE VLAN INTERFACE VLAN INTERFACE VLAN

Purpose Access the INTERFACE VLAN mode for the VLAN that you want to make an isolated VLAN. Enable the VLAN. Set the PVLAN mode of the selected VLAN to isolated. Add one or more host ports to the VLAN. The interfaces can be entered singly or in range format, either comma-delimited (slot/port,port,port) or hyphenated (slot/
port-port).

2 3 4

no shutdown private-vlan mode isolated tagged interface

or
untagged interface

Only ports defined as host can be added to the VLAN.

Figure 28-2 shows the use of the PVLAN commands that are used in VLAN INTERFACE mode to configure the PVLAN member VLANs (primary, community, and isolated VLANs):

554

Private VLANs

Figure 28-2. Configuring VLANs for a Private VLAN

FTOS#conf FTOS(conf)# interface vlan 10 FTOS(conf-vlan-10)# private-vlan mode primary FTOS(conf-vlan-10)# private-vlan mapping secondary-vlan 100-101 FTOS(conf-vlan-10)# untagged Gi 2/1 FTOS(conf-vlan-10)# tagged Gi 2/3 FTOS(conf)# interface vlan 101 FTOS(conf-vlan-101)# private-vlan mode community FTOS(conf-vlan-101)# untagged Gi 2/10 FTOS(conf)# interface vlan 100

Private VLAN Configuration Example

Figure 28-3. Sample Private VLAN Topology

The following configuration is based on the example diagram, above: On C300-1: Gi 0/0 and Gi 23 are configured as promiscuous ports, assigned to the primary VLAN, VLAN 4000. Gi 0/25 is configured as a PVLAN trunk port, also assigned to the primary VLAN 4000. Gi 0/24 and Gi 0/47 are configured as host ports and assigned to the isolated VLAN, VLAN 4003. Gi 4/0 and Gi 23 are configured as host ports and assigned to the community VLAN, VLAN 4001. Gi 4/24 and Gi 4/47 are configured as host ports and assigned to community VLAN 4002.

Private VLANs | 555

www.dell.com | support.dell.com

The result is that: The ports in community VLAN 4001 can communicate directly with each other and with promiscuous ports. The ports in community VLAN 4002 can communicate directly with each other and with promiscuous ports The ports in isolated VLAN 4003 can only communicate with the promiscuous ports in the primary VLAN 4000. All the ports in the secondary VLANs (both community and isolated VLANs) can only communicate with ports in the other secondary VLANs of that PVLAN over Layer 3, and only when the command ip local-proxy-arp is invoked in the primary VLAN.
Note: Even after ip-local-proxy-arp is disabled (no ip-local-proxy-arp) in a secondary VLAN, Layer 3 communication may happen between some secondary VLAN hosts, until the ARP timeout happens on those secondary VLAN hosts.

In parallel, on S50-1: Gi 0/3 is a promiscuous port and Gi 0/25 is a PVLAN trunk port, assigned to the primary VLAN 4000. Gi 0/4-6 are host ports. Gi 0/4 and Gi 0/5 are assigned to the community VLAN 4001, while Gi 0/6 is assigned to the isolated VLAN 4003.

The result is that: The S50V ports would have the same intra-switch communication characteristics as described above for the C300. For transmission between switches, tagged packets originating from host PVLAN ports in one secondary VLAN and destined for host PVLAN ports in the other switch travel through the promiscuous ports in the local VLAN 4000 and then through the trunk ports (0/25 in each switch).

Inspecting the Private VLAN Configuration

The standard methods of inspecting configurations also apply in PVLANs: Within the INTERFACE and INTERFACE VLAN modes, use the show config command to display the specific interface configuration. Inspect the running-config, and, with the grep pipe option (show running-config | grep string), you can display a specific part of the running-config. Figure 28-8 shows the PVLAN parts of the running-config from the S50V switch in the topology diagram shown in Figure 28-3, above. You can also use one of three show commands that are specific to the Private VLAN feature: show interfaces private-vlan [interface interface]: Display the type and status of the configured PVLAN interfaces. See the example output in the Security chapter of the FTOS Command Reference. show vlan private-vlan [community | interface | isolated | primary | primary_vlan | interface interface]: Display the configured PVLANs or interfaces that are part of a PVLAN. Figure 28-4 shows the results of using the command without command options on the C300 switch in the topology diagram shown in Figure 28-3, above, while Figure 28-5 shows the results on the S50V.

556

Private VLANs

primary-secondary VLAN mapping. See the example output from the S50V, above, in Figure 28-6. Two show commands revised to display PVLAN data are:
show arp show vlan: See

show vlan private-vlan mapping: Display the

revised output in Figure 28-7.

Figure 28-4. show vlan private-vlan Example Output from C300

c300-1#show vlan private-vlan Primary Secondary Type Active ------- --------- --------- -----4000 Primary Yes 4001 Community Yes 4002 Community Yes 4003 Isolated Yes Ports -----------------------------------------Gi 0/0,23,25 Gi 4/0,23 Gi 4/24,47 Gi 0/24,47

Figure 28-5. show vlan private-vlan Example Output from S50V

S50-1#show vlan private-vlan Primary Secondary Type Active ------- --------- --------- -----4000 Primary Yes 4001 Community Yes 4003 Isolated Yes Ports -----------------------------------------Gi 0/3,25 Gi 0/4-5 Gi 0/6

Figure 28-6. show vlan private-vlan mapping Example Output from S50V
S50-1#show vlan private-vlan mapping Private Vlan: Primary : 4000 Isolated : 4003 Community : 4001

In the following screenshot, note the addition of the PVLAN codes P, I, and C in the left column:
Figure 28-7. show vlan Example Output from S50V
S50V#show vlan Codes: Q: U x G NUM 1 100 200 201 * - Default VLAN, G - GVRP VLANs, P - Primary, C - Community, I - Isolated Untagged, T - Tagged Dot1x untagged, X - Dot1x tagged GVRP tagged, M - Vlan-stack Status Inactive Inactive Inactive Inactive Description Q Ports

* P I

PVLAN codes

primary VLAN in PVLAN isolated VLAN in VLAN 200

T Gi 0/19-20 T Gi 0/21

Private VLANs | 557

www.dell.com | support.dell.com

Figure 28-8. Example running-config Output of PVLAN Configuration from S50V

! interface GigabitEthernet 0/3 no ip address switchport switchport mode private-vlan promiscuous no shutdown ! interface GigabitEthernet 0/4 no ip address switchport switchport mode private-vlan host no shutdown ! interface GigabitEthernet 0/5 no ip address switchport switchport mode private-vlan host no shutdown ! interface GigabitEthernet 0/6 no ip address switchport switchport mode private-vlan host no shutdown ! interface GigabitEthernet 0/25 no ip address switchport switchport mode private-vlan trunk no shutdown ! interface Vlan 4000 private-vlan mode primary private-vlan mapping secondary-vlan 4001-4003 no ip address tagged GigabitEthernet 0/3,25 no shutdown ! interface Vlan 4001 private-vlan mode community

558

Private VLANs

29
Per-VLAN Spanning Tree Plus
Per-VLAN Spanning Tree Plus is supported platforms:

ces

Protocol Overview
Per-VLAN Spanning Tree Plus (PVST+) is a variation of Spanning Treedeveloped by a third party that allows you to configure a separate Spanning Tree instance for each VLAN. For more information on Spanning Tree, see Chapter 38, Spanning Tree Protocol.
Figure 29-1. Per-VLAN Spanning Tree

STI 1/2/3 root R2

STI 1: VLAN 100 STI 2: VLAN 200 STI 3: VLAN 300

R3
3/22

2/32

2/12
Forwarding
kin g

3/12

Bl

oc

1/22

X X X 1/32

R1

Per-VLAN Spanning Tree Plus | 559

www.dell.com | support.dell.com

FTOS supports three other variations of Spanning Tree, as shown in Table 29-1.
Table 29-1. FTOS Supported Spanning Tree Protocols IEEE Specification 802.1d 802.1w 802.1s Third Party

Dell Networking Term Spanning Tree Protocol Rapid Spanning Tree Protocol Multiple Spanning Tree Protocol Per-VLAN Spanning Tree Plus

Implementation Information
The FTOS implementation of PVST+ is based on IEEE Standard 802.1d. The FTOS implementation of PVST+ uses IEEE 802.1s costs as the default costs (Table 29-2). Other implementations use IEEE 802.1d costs as the default costs if you are using Dell Networking systems in a multi-vendor network, verify that the costs are values you intended. You must allocate at least the default minimum amount of Layer 2 ACL CAM space when employing PVST+ on the E-Series. See Configure Ingress Layer 2 ACL Sub-partitions. On the C-Series and S-Series, you can enable PVST+ on 254 VLANs.

Configure Per-VLAN Spanning Tree Plus

Configuring PVST+ is a four-step process: 1. Configure interfaces for Layer 2. 2. Place the interfaces in VLANs. 3. Enable PVST+. See page 561. 4. Optionally, for load balancing, select a non-default bridge-priority for a VLAN. See page 561.

Related Configuration Tasks

Modify Global PVST+ Parameters Modify Interface PVST+ Parameters Configure an EdgePort Flush MAC Addresses after a Topology Change Preventing Network Disruptions with BPDU Guard SNMP Traps for Root Elections and Topology Changes Configuring Spanning Trees as Hitless

560

Per-VLAN Spanning Tree Plus

PVST+ in Multi-vendor Networks PVST+ Extended System ID PVST+ Sample Configurations

Enable PVST+
When you enable PVST+, FTOS instantiates STP on each active VLAN. To enable PVST+ globally:
Step 1 2 Task Enter PVST context. Enable PVST+. Command Syntax
protocol spanning-tree pvst no disable

Command Mode PROTOCOL PVST PROTOCOL PVST

Disable PVST+
Task Disable PVST+ globally. Disable PVST+ on an interface, or remove a PVST+ parameter configuration. Command Syntax
disable no spanning-tree pvst

Command Mode PROTOCOL PVST INTERFACE

Display your PVST+ configuration by entering the command show config from PROTOCOL PVST context, as shown in fig.
Figure 29-2. Display the PVST+ Configuration
FTOS_E600(conf-pvst)#show config verbose ! protocol spanning-tree pvst no disable vlan 100 bridge-priority 4096

Influence PVST+ Root Selection

In Figure 29-1, all VLANs use the same forwarding topology because R2 is elected the root, and all GigabitEthernet ports have the same cost. Figure 29-3 changes the bridge priority of each bridge so that a different forwarding topology is generated for each VLAN. This behavior demonstrates how you can use PVST+ to achieve load balancing.

Per-VLAN Spanning Tree Plus | 561

www.dell.com | support.dell.com

Figure 29-3. Load Balancing with PVST+

STI 2 root
vlan 100 bridge-priority 4096

R2

STI 1: VLAN 100 STI 2: VLAN 200 STI 3: VLAN 300

STI 3 root R3
3/22 X 3/12
vlan 100 bridge-priority 4096

2/32

Blocking

2/12
Forwarding

1/22

X X 1/32

STI 1 root
vlan 100 bridge-priority 4096

R1

The bridge with the bridge value for bridge priority is elected root. Since all bridges use the default priority (until configured otherwise), lowest MAC address is used as a tie-breaker. Assign bridges a low non-default value for bridge priority to increase the likelihood that it will be selected as the STP root.
Task Assign a bridge priority. Range: 0 to 61440 Default: 32768 Command Syntax
vlan bridge-priority

Command Mode PROTOCOL PVST

Display the PVST+ forwarding topology by entering the command show spanning-tree pvst [vlan vlan-id] from EXEC Privilege mode, as shown in Figure 29-4.

562

Per-VLAN Spanning Tree Plus

Figure 29-4. Display the PVST+ Forwarding Topology

FTOS_E600(conf)#do show spanning-tree pvst vlan 100 VLAN 100 Root Identifier has priority 4096, Address 0001.e80d.b6d6 Root Bridge hello time 2, max age 20, forward delay 15 Bridge Identifier has priority 4096, Address 0001.e80d.b6d6 Configured hello time 2, max age 20, forward delay 15 We are the root of VLAN 100 Current root has priority 4096, Address 0001.e80d.b6d6 Number of topology changes 5, last change occurred 00:34:37 ago on Gi 1/32 Port 375 (GigabitEthernet 1/22) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.375 Designated root has priority 4096, address 0001.e80d.b6:d6 Designated bridge has priority 4096, address 0001.e80d.b6:d6 Designated port id is 128.375 , designated path cost 0 Number of transitions to forwarding state 2 BPDU sent 1159, received 632 The port is not in the Edge port mode Port 385 (GigabitEthernet 1/32) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.385 Designated root has priority 4096, address 0001.e80d.b6:d6 Designated bridge has priority 4096, address 0001.e80d.b6:d6 Designated port id is 128.385 , designated path cost 0

Modify Global PVST+ Parameters

The root bridge sets the values for forward-delay, and hello-time and overwrites the values set on other PVST+ bridges. Forward-delay is the amount of time an interface waits in the Listening State and the Learning State before it transitions to the Forwarding State. Hello-time is the time interval in which the bridge sends Bridge Protocol Data Units (BPDUs). Max-age is the length of time the bridge maintains configuration information before it refreshes that information by recomputing the PVST+ topology.

To change PVST+ parameters, use the following commands on the root bridge:
Task Change the forward-delay parameter. Range: 4 to 30 Default: 15 seconds Change the hello-time parameter. Note: With large configurations (especially those with more ports) Dell Networking recommends that you increase the hello-time. Range: 1 to 10 Default: 2 seconds Command Syntax
vlan forward-delay

Command Mode PROTOCOL PVST

vlan hello-time

PROTOCOL PVST

Per-VLAN Spanning Tree Plus | 563

www.dell.com | support.dell.com

Task Change the max-age parameter. Range: 6 to 40 Default: 20 seconds

Command Syntax
vlan max-age

Command Mode PROTOCOL PVST

The values for global PVST+ parameters are given in the output of the command show spanning-tree pvst, as shown in Figure 29-4.

Modify Interface PVST+ Parameters

You can adjust two interface parameters to increase or decrease the probability that a port becomes a forwarding port: Port cost is a value that is based on the interface type. The greater the port cost, the less likely the port will be selected to be a forwarding port. Port priority influences the likelihood that a port will be selected to be a forwarding port in case that several ports have the same port cost.

Table 29-2 lists the default values for port cost by interface.
Table 29-2. Port Cost 100-Mb/s Ethernet interfaces 1-Gigabit Ethernet interfaces 10-Gigabit Ethernet interfaces Port Channel with 100 Mb/s Ethernet interfaces Port Channel with 1-Gigabit Ethernet interfaces Port Channel with 10-Gigabit Ethernet interfaces PVST+ Default Port Cost Values Default Value 200000 20000 2000 180000 18000 1800

Note: The FTOS implementation of PVST+ uses IEEE 802.1s costs as the default costs. Other implementations use IEEE 802.1d costs as the default costs if you are using Dell Networking systems in a multi-vendor network, verify that the costs are values you intended.

To change the port cost or priority of an interface:

Task Change the port cost of an interface. Range: 0 to 200000 Default: see Table 29-2. Command Syntax
spanning-tree pvst vlan cost

Command Mode INTERFACE

564

Per-VLAN Spanning Tree Plus

Task Change the port priority of an interface. Range: 0 to 240, in increments of 16 Default: 128

Command Syntax
spanning-tree pvst vlan priority

Command Mode INTERFACE

pvst,

The values for interface PVST+ parameters are given in the output of the command show spanning-tree as shown in Figure 29-4.

Configure an EdgePort
The EdgePort feature enables interfaces to begin forwarding traffic approximately 30 seconds sooner. In this mode an interface forwards frames by default until it receives a BPDU that indicates that it should behave otherwise; it does not go through the Learning and Listening states. The bpduguard shutdown-on-violation option causes the interface hardware to be shutdown when it receives a BPDU. When only bpduguard is implemented, although the interface is placed in an Error Disabled state when receiving the BPDU, the physical interface remains up and spanning-tree will drop packets in the hardware after a BPDU violation. BPDUs are dropped in the software after receiving the BPDU violation. This feature is the same as PortFast mode in Spanning Tree. Caution: Configure EdgePort only on links connecting to an end station. EdgePort can cause loops if it is enabled on an interface connected to a network. To enable EdgePort on an interface, use the following command:
Task Enable EdgePort on an interface. Command Syntax
spanning-tree pvst edge-port [bpduguard | shutdown-on-violation]

Command Mode INTERFACE

The EdgePort status of each interface is given in the output of the command show spanning-tree pvst, as shown in Figure 29-4.

Per-VLAN Spanning Tree Plus | 565

www.dell.com | support.dell.com

FTOS Behavior: Regarding bpduguard shutdown-on-violation behavior: 1If the interface to be shutdown is a port channel then all the member ports are disabled in the hardware. 2When a physical port is added to a port channel already in error disable state, the new member port will also be disabled in the hardware. 3When a physical port is removed from a port channel in error disable state, the error disabled state is cleared on this physical port (the physical port will be enabled in the hardware). 4The reset linecard command does not clear the error disabled state of the port or the hardware disabled state. The interface continues to be disables in the hardware. The error disabled state can be cleared with any of the following methods: Perform an shutdown command on the interface. Disable the shutdown-on-violation command on the interface ( no spanning-tree stp-id portfast [bpduguard | [shutdown-on-violation]] ). Disable spanning tree on the interface (no spanning-tree in INTERFACE mode). Disabling global spanning tree (no spanning-tree in CONFIGURATION mode).

PVST+ in Multi-vendor Networks

Some non-Dell Networking systems which have hybrid ports participating in PVST+ transmit two kinds of BPDUs: an 802.1D BPDU and an untagged PVST+ BPDU. Dell Networking systems do not expect PVST+ BPDU (tagged or untagged) on an untagged port. If this happens, FTOS places the port in error-disable state. This behavior might result in the network not converging. To prevent FTOS from executing this action, use the command no spanning-tree pvst err-disable cause invalid-pvst-bpdu. After you configure this command, if the port receives a PVST+ BPDU, the BPDU is dropped, and the port remains operational.

PVST+ Extended System ID

In Figure 29-5, ports P1 and P2 are untagged members of different VLANs. These ports are untagged because the hub is VLAN unaware. There is no data loop in the above scenario, however, PVST+ can be employed to avoid potential misconfigurations. If PVST+ is enabled on the Dell Networking switch in this network, P1 and P2 receive BPDUs from each other. Ordinarily, the Bridge ID in the frame matches the Root ID, a loop is detected, and the rules of convergence require that P2 move to blocking state because it has the lowest port ID. To keep both ports in forwarding state, use Extend System ID. Extend System ID augments the Bridge ID with a VLAN ID to differentiate BPDUs on each VLAN so that PVST+ does not detect a loop, and both ports can remain in forwarding state.

566

Per-VLAN Spanning Tree Plus

Figure 29-5. PVST+ with Extend System ID

Dell Force10 System

P1 untagged in VLAN 10

VLAN unaware Hub

P2 untagged in VLAN 20

moves to blocking unless Extended System ID is enabled

Task Augment the Bridge ID with the VLAN ID.

Command Syntax
extend system-id

Command Mode PROTOCOL PVST

FTOS(conf-pvst)#do show spanning-tree pvst vlan 5 brief VLAN 5 Executing IEEE compatible Spanning Tree Protocol Root ID Priority 32773, Address 0001.e832.73f7 Root Bridge hello time 2, max age 20, forward delay 15 Bridge ID Priority 32773 (priority 32768 sys-id-ext 5), Address 0001.e832.73f7 We are the root of Vlan 5 Configured hello time 2, max age 20, forward delay 15 ...

PVST+ Sample Configurations

Figure 29-6, Figure 29-7, and Figure 29-8 provide the running configurations for the topology shown in Figure 29-3.

Per-VLAN Spanning Tree Plus | 567

www.dell.com | support.dell.com

Figure 29-6. PVST+ Sample Configuration: R1 Running-configuration

interface GigabitEthernet 1/22 no ip address switchport no shutdown ! interface GigabitEthernet 1/32 no ip address switchport no shutdown ! protocol spanning-tree pvst no disable vlan 100 bridge-priority 4096 interface Vlan 100 no ip address tagged GigabitEthernet 1/22,32 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 1/22,32 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 1/22,32 no shutdown ! protocol spanning-tree pvst no disable vlan 100 bridge-priority 4096

568

Per-VLAN Spanning Tree Plus

Figure 29-7. PVST+ Sample Configuration: R2 Running-configuration

interface GigabitEthernet 2/12 no ip address switchport no shutdown ! interface GigabitEthernet 2/32 no ip address switchport no shutdown ! interface Vlan 100 no ip address tagged GigabitEthernet 2/12,32 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 2/12,32 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 2/12,32 no shutdown ! protocol spanning-tree pvst no disable vlan 200 bridge-priority 4096

Figure 29-8. PVST+ Sample Configuration: R3 Running-configuration

interface GigabitEthernet 3/12 no ip address switchport no shutdown ! interface GigabitEthernet 3/22 no ip address switchport no shutdown ! interface Vlan 100 no ip address tagged GigabitEthernet 3/12,22 no shutdown ! interface Vlan 200 no ip address tagged GigabitEthernet 3/12,22 no shutdown ! interface Vlan 300 no ip address tagged GigabitEthernet 3/12,22 no shutdown ! protocol spanning-tree pvst no disable vlan 300 bridge-priority 4096

Per-VLAN Spanning Tree Plus | 569

www.dell.com | support.dell.com |

570

Per-VLAN Spanning Tree Plus

30
Quality of Service
Quality of Service (QoS) is supported on platforms:

ces

Differentiated service is accomplished by classifying and queuing traffic, and assigning priorities to those queues. The E-Series has eight unicast queues per port and 128 multicast queues per-port pipe. Traffic is queued on ingress and egress. By default, on ingress, all data traffic is mapped to Queue 0, and all control traffic is mapped to Queue 7. On egress control traffic is mapped across all eight queues. All queues are serviced using the Weighted Fair Queuing scheduling algorithm. You can only manage queuing prioritization on egress. The C-Series traffic has eight queues per port. Four queues are for data traffic and four are for control traffic. All queues are serviced using the Deficit Round Robin scheduling algorithm. You can only manage queuing prioritization on egress.

Table 30-1. Feature

FTOS Support for Port-based, Policy-based, and Multicast QoS Features Platform ces ces ces ces e ces ces ces ces ces ces Ingress + Egress Ingress Egress Direction Ingress + Egress Ingress

Port-based QoS Configurations Set dot1p Priorities for Incoming Traffic Honor dot1p Priorities on Ingress Traffic Configure Port-based Rate Policing Configure Port-based Rate Limiting Configure Port-based Rate Shaping Policy-based QoS Configurations Classify Traffic Create a Layer 3 class map Set DSCP values for egress packets based on flow Create a Layer 2 class map

Quality of Service | 571

www.dell.com | support.dell.com

Table 30-1. Feature

FTOS Support for Port-based, Policy-based, and Multicast QoS Features Platform ces ces ces ces ces ces e ces ces e ces ces ces ecs ces ces ces ces e e e e ces e cs Ingress Egress Egress Egress Ingress + Egress Ingress Egress Direction Ingress + Egress Ingress

Create a QoS Policy Create an input QoS policy Configure policy-based rate policing Set a DSCP value for egress packets Set a dot1p value for egress packets Create an output QoS policy Configure policy-based rate limiting Configure policy-based rate shaping Allocate bandwidth to queue Specify WRED drop precedence Create Policy Maps Create Input Policy Maps Honor DSCP values on ingress packets Honoring dot1p values on ingress packets Create Output Policy Maps Specify an aggregate QoS policy QoS Rate Adjustment Strict-priority Queueing Weighted Random Early Detection Create WRED Profiles Configure WRED for Storm Control Allocating Bandwidth to Multicast Queues Pre-calculating Available QoS CAM Space Viewing QoS CAM Entries Configure Quality of Service for an Office VOIP Deployment

572

Quality of Service

Figure 30-1. Dell Networking QoS Architecture

Marking (DiffServ, 802.1p, Exp)

Ingress Packet Processing

Packet Classification (ACL)

Rate Policing

Buffers & Class-based Queues

Switching

Rate Limiting

Buffers & Class-based Queues

Egress Packet Processing

Traffic Shaping

Egress Congestion Management (WFQ Scheduling)

Congestion Avoidance (WRED)

Implementation Information
Dell Networking QoS implementation complies with IEEE 802.1p User Priority Bits for QoS Indication. It also implements these Internet Engineering Task Force (IETF) documents: RFC 2474, Definition of the Differentiated Services Field (DS Field) in the IPv4 Headers RFC 2475, An Architecture for Differentiated Services RFC 2597, Assured Forwarding PHB Group RFC 2598, An Expedited Forwarding PHB

You cannot configure port-based and policy-based QoS on the same interface, and SONET line cards support only port-based QoS.

Port-based QoS Configurations

You can configure the following QoS features on an interface: Set dot1p Priorities for Incoming Traffic Configure Port-based Rate Policing

Quality of Service | 573

www.dell.com | support.dell.com

Configure Port-based Rate Limiting Configure Port-based Rate Shaping Storm Control

Set dot1p Priorities for Incoming Traffic

Change the priority of incoming traffic on the interface using the command dot1p-priority from INTERFACE mode, as shown in Figure 30-2. FTOS places traffic marked with a priority in a queue based on Table 30-2. If you set a dot1p priority for a port-channel, all port-channel members are configured with the same value. You cannot assign a dot1p value to an individual interfaces in a port-channel.
FTOS Behavior: The C-Series and S-Series distribute eight dot1p priorities across four data queues. This is different from the E-Series, which distributes eight dot1p priorities across eight queues (Table 30-2).

Table 30-2.

dot1p-priority values and queue numbers E-Series Queue Number 2 0 1 3 4 5 6 7 C-Series Queue Number 1 0 0 1 2 2 3 3 S-Series Queue Number 1 0 0 1 2 2 3 3

dot1p 0 1 2 3 4 5 6 7

Figure 30-2. Configuring dot1p Priority on an Interface

FTOS#config FTOS(conf)#interface gigabitethernet 1/0 FTOS(conf-if)#switchport FTOS(conf-if)#dot1p-priority 1 FTOS(conf-if)#end FTOS#

Honor dot1p Priorities on Ingress Traffic

By default FTOS does not honor dot1p priorities on ingress traffic. Use the command service-class dynamic dot1p from INTERFACE mode to honor dot1p priorities on ingress traffic, as shown in Figure 30-3. You can configure this feature on physical interfaces and port-channels, but you cannot configure it on individual interfaces in a port channel.

574

Quality of Service

On the C-Series and S-Series you can configure service-class dynamic dot1p from CONFIGURATION mode, which applies the configuration to all interfaces. A CONFIGURATION mode service-class dynamic dot1p entry supersedes any INTERFACE entries. See Mapping dot1p values to service queues.
Note: You cannot configure service-policy input and service-class dynamic dot1p on the same interface. Figure 30-3. service-class dynamic dot1p Command Example
FTOS#config t FTOS(conf)#interface gigabitethernet 1/0 FTOS(conf-if)#service-class dynamic dot1p FTOS(conf-if)#end FTOS#

Priority-tagged Frames on the Default VLAN

Priority-tagged Frames on the Default VLAN is available only on platforms:

ex c s

Priority-tagged frames are 802.1Q tagged frames with VLAN ID 0. For VLAN classification these packets are treated as untagged. However, the dot1p value is still honored when service-class dynamic dot1p or trust dot1p is configured. When priority-tagged frames ingress an untagged port or hybrid port the frames are classified to the default VLAN of the port, and to a queue according to their dot1p priority dot1p priority if service-class dynamic dotp or trust dot1p are configured. When priority-tagged frames ingress a tagged port, the frames are dropped because for a tagged port the default VLAN is 0.
FTOS Behavior: Hybrid ports can receive untagged, tagged, and priority tagged frames. The rate metering calculation might be inaccurate for untagged ports, since an internal assumption is made that all frames are treated as tagged. Internally the ASIC adds a 4-bytes tag to received untagged frames. Though these 4-bytes are not part of the untagged frame received on the wire, they are included in the rate metering calculation resulting in metering inaccuracy.

Configure Port-based Rate Policing

Rate policing ingress traffic on an interface using the command rate police from INTERACE mode, as shown in Figure 30-4. If the interface is a member of a VLAN, you may specify the VLAN for which ingress packets are policed.
Figure 30-4. Rate Policing Ingress Traffic
FTOS#config t FTOS(conf)#interface gigabitethernet 1/0 FTOS(conf-if)#rate police 100 40 peak 150 50 FTOS(conf-if)#end FTOS#

Quality of Service | 575

www.dell.com | support.dell.com

Figure 30-5. Displaying your Rate Policing Configuration

FTOS#show interfaces gigabitEthernet 1/2 rate police Rate police 300 (50) peak 800 (50) Traffic Monitor 0: normal 300 (50) peak 800 (50) Out of profile yellow 23386960 red 320605113 Traffic Monitor 1: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 2: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 3: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 4: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 5: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 6: normal NA peak NA

Configure Port-based Rate Limiting

Configure Port-based Rate Limiting is supported only on platform

FTOS Behavior: On the C-Series and S-Series, rate shaping is effectively rate limiting because of its smaller buffer size.

Rate limit egress traffic on an interface using the command rate limit from INTERFACE mode, as shown in Figure 30-6. If the interface is a member of a VLAN, you may specify the VLAN for which egress packets are rate limited.
Figure 30-6. Rate Limiting Egress Traffic
FTOS#config t FTOS(conf)#interface gigabitethernet 1/0 FTOS(conf-if)#rate limit 100 40 peak 150 50 FTOS(conf-if)#end FTOS#

Display how your rate limiting configuration affects traffic using the keyword rate limit with the command show interfaces, as shown in Figure 30-7.

576

Quality of Service

Figure 30-7. Displaying How Your Rate Limiting Configuration Affects Traffic
FTOS#show interfaces gigabitEthernet 1/1 rate limit Rate limit 300 (50) peak 800 (50) Traffic Monitor 0: normal 300 (50) peak 800 (50) Out of profile yellow 23386960 red 320605113 Traffic Monitor 1: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 2: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 3: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 4: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 5: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 6: normal NA peak NA Out of profile yellow 0 red 0 Traffic Monitor 7: normal NA peak NA Out of profile yellow 0 red 0 Total: yellow 23386960 red 320605113

Configure Port-based Rate Shaping

Configure Port-based Rate Limiting is supported only on platform

ces

FTOS Behavior: On the C-Series and S-Series, rate shaping is effectively rate limiting because of its smaller buffer size. On the S60, rate shaping on tagged ports is slightly greater than the configured rate and rate shaping on untagged ports is slightly less than configured rate.

Rate shaping buffers, rather than drops, traffic exceeding the specified rate until the buffer is exhausted. If any stream exceeds the configured bandwidth on a continuous basis, it can consume all of the buffer space that is allocated to the port. Apply rate shaping to outgoing traffic on a port using the command rate shape from INTERFACE mode, as shown in Figure 30-8. Apply rate shaping to a queue using the command rate-shape from QoS Policy mode.
FTOS#config FTOS(conf)#interface gigabitethernet 1/0 FTOS(conf-if)#rate shape 500 50 FTOS(conf-if)#end FTOS#

Figure 30-8. Applying Rate Shaping to Outgoing Traffic

Policy-based QoS Configurations

Policy-based QoS configurations consist of the components shown in Figure 30-9.

Quality of Service | 577

www.dell.com | support.dell.com

Figure 30-9. Constructing Policy-based QoS Configurations

Interface

Input Service Policy

Output Service Policy

0 Input Policy Map

7 Input Policy Map

0 Output Policy Map

7 Output Policy Map

Class Map

DSCP

Input QoS Policy

Output QoS Policy

L3 ACL

L3 Fields

Rate Policing

Outgoing Marking

Rate Limiting

WRED

B/W %

Classify Traffic
Class maps differentiate traffic so that you can apply separate quality of service policies to each class. For both class maps, Layer 2 and Layer 3, FTOS matches packets against match criteria in the order that you configure them.

Create a Layer 3 class map

A Layer 3 class map differentiates ingress packets based on DSCP value or IP precedence, and characteristics defined in an IP ACL. You may specify more than one DSCP and IP precedence value, but only one value must match to trigger a positive match for the class map. 1. Create a match-any class map using the command class-map match-any or a match-all class map using the command class-map match-all from CONFIGURATION mode, as shown in Figure 30-10. 2. Once you create a class-map, FTOS places you in CLASS MAP mode. From this mode, specify your match criteria using the command match ip, as shown in Figure 30-10. Match-any class maps allow up to five ACLs, and match-all class-maps allow only one ACL. 3. After you specify your match criteria, link the class-map to a queue using the command service-queue from POLICY MAP mode, as shown in Figure 30-10.

578

Quality of Service

Figure 30-10.

Using the Order Keyword in ACLs

FTOS(conf)#ip access-list standard acl1 FTOS(config-std-nacl)#permit 20.0.0.0/8 FTOS(config-std-nacl)#exit FTOS(conf)#ip access-list standard acl2 FTOS(config-std-nacl)#permit 20.1.1.0/24 order 0 FTOS(config-std-nacl)#exit FTOS(conf)#class-map match-all cmap1 FTOS(conf-class-map)#match ip access-group acl1 FTOS(conf-class-map)#exit FTOS(conf)#class-map match-all cmap2 FTOS(conf-class-map)#match ip access-group acl2 FTOS(conf-class-map)#exit FTOS(conf)#policy-map-input pmap FTOS(conf-policy-map-in)#service-queue 7 class-map cmap1 FTOS(conf-policy-map-in)#service-queue 4 class-map cmap2 FTOS(conf-policy-map-in)#exit FTOS(conf)#interface gig 1/0 FTOS(conf-if-gi-1/0)#service-policy input pmap

Create a Layer 2 class map

All class maps are Layer 3 by default; you can create a Layer 2 class map by specifying the option layer2 with the class-map command. A Layer 2 class map differentiates traffic according to 802.1p value and/or characteristics defined in a MAC ACL. 1. Create a match-any class map using the command class-map match-any or a match-all class map using the command class-map match-all from CONFIGURATION mode, and enter the keyword layer2. 2. Once you create a class-map, FTOS places you in CLASS MAP mode. From this mode, specify your match criteria using the command match mac. Match-any class maps allow up to five access-lists, and match-all class-maps allow only one. You can match against only one VLAN ID. 3. After you specify your match criteria, link the class-map to a queue using the command service-queue from POLICY MAP mode.

Determine the order in which ACLs are used to classify traffic

When you link class-maps to queues using the command service-queue, FTOS matches the class-maps according to queue priority (queue numbers closer to 0 have lower priorities). For example, in Figure 30-10, class-map cmap2 is matched against ingress packets before cmap1. ACLs acl1 and acl2 have overlapping rules because the address range 20.1.1.0/24 is within 20.0.0.0/8. Therefore, (without the keyword order) packets within the range 20.1.1.0/24 match positive against cmap1 and are buffered in queue 7, though you intended for these packets to match positive against cmap2 and be buffered in queue 4. In cases such as these, where class-maps with overlapping ACL rules are applied to different queues, use the order keyword to specify the order in which you want to apply ACL rules, as shown in Figure 30-10. The order can range from 0 to 254. FTOS writes to the CAM ACL rules with lower order numbers (order numbers closer to 0) before rules with higher order numbers so that packets are matched as you intended. By default, all ACL rules have an order of 254.

Quality of Service | 579

www.dell.com | support.dell.com

Set DSCP values for egress packets based on flow

Match-any Layer 3 flows may have several match criteria. All flows that match at least one of the match criteria are mapped to the same queue since they are in the same class map. Setting a DSCP value from QOS-POLICY-IN mode (see Set a DSCP value for egress packets) assigns the same DSCP value to all of the matching flows in the class-map. The Flow-based DSCP Marking feature allows you to assign different DSCP to each match criteria CLASS-MAP mode using the option set-ip-dscp with the match command so that matching flows within a class map can have different DSCP values, as shown in Figure 30-11. The values you set from CLASS-MAP mode override the value you QoS input policy DSCP value, and packets matching the rule are marked with the specified value.
Figure 30-11.
FTOS#show ! class-map match ip match ip match ip

Marking Flows in the Same Queue with Different DSCP Values

run class-map match-any example-flowbased-dscp access-group test set-ip-dscp 2 access-group test1 set-ip-dscp 4 precedence 7 set-ip-dscp 1

FTOS#show run qos-policy-input ! qos-policy-input flowbased set ip-dscp 3 FTOS# show cam layer3 linecard 2 port-set 0 Cam Port Dscp Proto Tcp Src Dst SrcIp DstIp DSCP Queue Index Flag Port Port Marking ----------------------------------------------------------------------------------------------16260 1 0 TCP 0x0 0 0 1.1.1.0/24 0.0.0.0/0 2 0 16261 1 0 UDP 0x0 0 0 2.2.2.2/32 0.0.0.0/0 4 0 16262 1 56 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 1 0 24451 1 0 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0

Display configured class maps and match criteria

Display all class-maps or a specific class map using the command show qos class-map from EXEC Privilege mode.

580

Quality of Service

FTOS Behavior: An explicit deny any rule in a Layer 3 ACL used in a (match any or match all) class-map creates a default to Queue 0" entry in the CAM, which causes unintended traffic classification. Below, traffic is classified in two Queues, 1 and 2. Class-map ClassAF1 is match any, and ClassAF2 is match all.
FTOS#show running-config policy-map-input ! policy-map-input PolicyMapIn service-queue 1 class-map ClassAF1 qos-policy QosPolicyIn-1 service-queue 2 class-map ClassAF2 qos-policy QosPolicyIn-2 FTOS#show running-config class-map ! class-map match-any ClassAF1 match ip access-group AF1-FB1 set-ip-dscp 10 match ip access-group AF1-FB2 set-ip-dscp 12 match ip dscp 10 set-ip-dscp 14 ! class-map match-all ClassAF2 match ip access-group AF2 match ip dscp 18 FTOS#show running-config ACL ! ip access-list extended AF1-FB1 seq 5 permit ip host 23.64.0.2 any seq 10 deny ip any any ! ip access-list extended AF1-FB2 seq 5 permit ip host 23.64.0.3 any seq 10 deny ip any any ! ip access-list extended AF2 seq 5 permit ip host 23.64.0.5 any seq 10 deny ip any any FTOS#show cam layer3-qos interface gigabitethernet 4/49 Cam Port Dscp Proto Tcp Src Dst SrcIp DstIp DSCP Queue Index Flag Port Port Marking --------------------------------------------------------------------------------------------20416 1 18 IP 0x0 0 0 23.64.0.5/32 0.0.0.0/0 20 2 20417 1 18 IP 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0 20418 1 0 IP 0x0 0 0 23.64.0.2/32 0.0.0.0/0 10 1 20419 1 0 IP 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0 20420 1 0 IP 0x0 0 0 23.64.0.3/32 0.0.0.0/0 12 1 20421 1 0 IP 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0 20422 1 10 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 14 1 24511 1 0 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0

Above, the ClassAF1 does not classify traffic as intended. Traffic matching the first match criteria is classified to Queue 1, but all other traffic is classified to Queue 0 as a result of CAM entry 20419. When the explicit deny any rule is removed from all three ACLs, the CAM reflects exactly the desired classification.
FTOS#show cam layer3-qos interface gigabitethernet 4/49 Cam Port Dscp Proto Tcp Src Dst SrcIp DstIp DSCP Queue Index Flag Port Port Marking --------------------------------------------------------------------------------------------20416 1 18 IP 0x0 0 0 23.64.0.5/32 0.0.0.0/0 20 2 20417 1 0 IP 0x0 0 0 23.64.0.2/32 0.0.0.0/0 10 1 20418 1 0 IP 0x0 0 0 23.64.0.3/32 0.0.0.0/0 12 1 20419 1 10 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 14 1 24511 1 0 0 0x0 0 0 0.0.0.0/0 0.0.0.0/0 0

Quality of Service | 581

www.dell.com | support.dell.com

Create a QoS Policy

There are two types of QoS policies: input and output. Input QoS policies regulate Layer 3 and Layer 2 ingress traffic. The regulation mechanisms for input QoS policies are rate policing and setting priority values. There are two types of input QoS policies: Layer 3 and Layer 2. Layer 3 QoS input policies allow you to rate police and set a DSCP or dot1p value. Layer 2 QoS input policies allow you to rate police and set a dot1p value.

Output QoS policies regulate Layer 3 egress traffic. The regulation mechanisms for output QoS policies are rate limiting, rate shaping, and WRED.
Note: When changing a "service-queue" configuration in a QoS policy map, all QoS rules are deleted and re-added automatically to ensure that the order of the rules is maintained. As a result, the Matched Packets value shown in the "show qos statistics" command is reset.

Create an input QoS policy

To create an input QoS policy: 1. Create a Layer 3 input QoS policy using the command qos-policy-input from CONFIGURATION mode. Create a Layer 2 input QoS policy by specifying the keyword layer2 after the command qos-policy-input. 2. Once you create an input QoS policy, do one or more of the following: Configure policy-based rate policing Set a DSCP value for egress packets Set a dot1p value for egress packets

Configure policy-based rate policing

Rate police ingress traffic using the command rate-police from QOS-POLICY-IN mode.

Set a DSCP value for egress packets

Set a DSCP value for egress packets based on ingress QoS classification, as shown in Figure 30-2. The 6 bits that are used for DSCP are also used to identify the queue in which traffic is buffered. When you set a DSCP value, FTOS displays an informational message advising you of the queue to which you should apply the QoS policy (using the command service-queue from POLICY-MAP-IN mode). If you apply the QoS policy to a queue other than the one specified in the informational message, FTOS replaces the first 3 bits in the DSCP field with the queue ID you specified.

582

Quality of Service

Figure 30-12.

Marking DSCP Values for Egress Packets

FTOS#config FTOS(conf)#qos-policy-input my-input-qos-policy FTOS(conf-qos-policy-in)#set ip-dscp 34 % Info: To set the specified DSCP value 34 (100-010 b) the QoS policy must be mapped to queue 4 (100 b). FTOS(conf-qos-policy-in)#show config ! qos-policy-input my-input-qos-policy set ip-dscp 34 FTOS(conf-qos-policy-in)#end FTOS#

Set a dot1p value for egress packets

Set a dot1p value for egress packets using the command set mac-dot1p from QOS-POLICY-IN mode.

Create an output QoS policy

To create an output QoS policy: 1. Create an output QoS policy using the command qos-policy-output from CONFIGURATION mode. 2. Once you configure an output QoS policy, do one or more of the following Configure policy-based rate limiting Configure policy-based rate shaping Allocate bandwidth to queue Specify WRED drop precedence

Configure policy-based rate limiting

Configure policy-based rate limiting is supported only on platform

Policy-based rate limiting is configured the same way as port-based rate limiting except that the command from QOS-POLICY-OUT mode is rate-limit rather than rate limit as it is in INTERFACE mode.

Configure policy-based rate shaping

Rate shape egress traffic using the command rate-shape from QOS-POLICY-OUT mode.

Allocate bandwidth to queue

The E-Series schedules unicast, multicast, and replication traffic for egress based on the Weighted Fair Queuing algorithm. The C-Series and S-Series schedule packets for egress based on Deficit Round Robin (DRR). These strategies both offer a guaranteed data rate. To allocate an amount bandwidth to a queue using the command bandwidth-percentage on the E-Series.

Quality of Service | 583

www.dell.com | support.dell.com

To allocate bandwidth to queues on the C-Series and S-Series, assign each queue a weight ranging from 1 to 1024, in increments of 2n, using the command bandwidth-weight. Table 30-3 shows the default bandwidth weights for each queue, and their equivalent percentage which is derived by dividing the bandwidth weight by the sum of all queue weights.
Table 30-3. Default Bandwidth Weights for C-Series and S-Series Equivalent Percentage 6.67% 13.33% 26.67% 53.33%

Queue 0 1 2 3

Default Weight 1 2 4 8

There are two key differences between allocating bandwidth by weight on the C-Series and S-Series and allocating bandwidth by percentage on the E-Series: 1. Assigning a weight to one queue affects the amount of bandwidth that is allocated to other queues. Therefore, whenever you are allocating bandwidth to one queue, Dell Networking recommends that you evaluate your bandwidth requirements for all other queues as well. 2. Because you are required to choose a bandwidth weight in increments of 2n you may not be able to achieve exactly a target bandwidth allocation. Table 30-4 shows an example of choosing bandwidth weights for all four queues to achieve a target bandwidth allocation.
Table 30-4. Assigning Bandwidth Weights for the C-Series and S-Series Equivalent Percentage 0.44% 28.44% 56.89% 14.22% Target Allocation 1% 25% 60% 14%

Queue 0 1 2 3

Weight 1 64 128 32

Specify WRED drop precedence

Specify WRED drop precedence is supported only on platform

Specify a WRED profile to yellow and/or green traffic using the command wred from QOS-POLICY-OUT mode. See Apply a WRED profile to traffic.

Create Policy Maps

There are two types of policy maps: input and output.

584

Quality of Service

Create Input Policy Maps

There are two types of input policy-maps: Layer 3 and Layer 2. 1. Create a Layer 3 input policy map using the command policy-map-input from CONFIGURATION mode. Create a Layer 2 input policy map by specifying the keyword layer2 with the policy-map-input command. 2. Once you create an input policy map, do one or more of the following: Apply a class-map or input QoS policy to a queue Apply an input QoS policy to an input policy map Honor DSCP values on ingress packets Honoring dot1p values on ingress packets

3. Apply the input policy map to an interface. See page 60.

FTOS Behavior: On ExaScale, FTOS cannot classify protocol traffic on a Layer 2 interface using Layer 3 policy map. The packets always take the default queue, Queue 0, and cannot be rate-policed.

Apply a class-map or input QoS policy to a queue

Assign an input QoS policy to a queue using the command service-queue from POLICY-MAP-IN mode.

Apply an input QoS policy to an input policy map

Apply an input QoS policy to an input policy map using the command policy-aggregate from POLICY-MAP-IN mode.

Honor DSCP values on ingress packets

FTOS provides the ability to honor DSCP values on ingress packets using Trust DSCP feature. Enable this feature using the command trust diffserv from POLICY-MAP-IN mode. Table 30-5 lists the standard DSCP definitions, and indicates to which queues FTOS maps DSCP values. When Trust DSCP is configured the matched packets and matched bytes counters are not incremented in show qos statistics.
Table 30-5. Default DSCP to Queue Mapping Traditional IP Precedence Network Control Internetwork Control EF (Expedited Forwarding) AF4 (Assured Forwarding) CRITIC/ECP Flash Override E-Series Internal Queue ID 7 6 5 4 C-Series Internal Queue ID 3 3 2 2 S-Series Internal DSCP/CP Queue ID decimal 3 3 2 2 3247 4863

DSCP/CP hex range (XXX)xxx DSCP Definition 111XXX 110XXX 101XXX 100XXX

Quality of Service | 585

www.dell.com | support.dell.com

DSCP/CP hex range (XXX)xxx DSCP Definition 011XXX 010XXX 001XXX 000XXX AF3 AF2 AF1 BE (Best Effort)

Traditional IP Precedence Flash Immediate Priority Best Effort

E-Series Internal Queue ID 3 2 1 0

C-Series Internal Queue ID 1 1 0 0

S-Series Internal DSCP/CP Queue ID decimal 1 1 0 0 015 1631

Honoring dot1p values on ingress packets

FTOS provides the ability to honor dot1p values on ingress packets with the Trust dot1p feature. Enable Trust dot1p using the command trust dot1p from POLICY-MAP-IN mode. Table 30-6 specifies the queue to which the classified traffic is sent based on the dot1p value.
Table 30-6. Default dot1p to Queue Mapping E-Series Queue ID 2 0 1 3 4 5 6 7 C-Series Queue ID 1 0 0 1 2 2 3 3 S-Series Queue ID 1 0 0 1 2 2 3 3

dot1p 0 1 2 3 4 5 6 7

The dot1p value is also honored for frames on the default VLAN; see Priority-tagged Frames on the Default VLAN.

Fall Back to trust diffserve or dot1p

Fall Back to trust diffserve or dot1p is available only on platforms:

When using QoS service policies with multiple class maps, you can configure FTOS to use the incoming DSCP or dot1p marking as a secondary option for packet queuing in the event that no match occurs in the class maps. When class-maps are used, traffic is matched against each class-map sequentially from first to last. The sequence is based on the priority of the rules, as follows: 1. rules with lowest priority, or in the absence of a priority configuration, 2. rules of the next numerically higher queue

586

Quality of Service

By default, if no match occurs, the packet is queued to the default queue, Queue 0. In the following configuration, packets are classified to queues using the three class maps:
! policy-map-input input-policy service-queue 1 class-map qos-BE1 service-queue 3 class-map qos-AF3 service-queue 4 class-map qos-AF4 ! class-map match-any qos-AF3 match ip dscp 24 match ip access-group qos-AF3-ACL ! class-map match-any qos-AF4 match ip dscp 32 match ip access-group qos-AF4-ACL ! class-map match-all qos-BE1 match ip dscp 0 match ip access-group qos-BE1-ACL

The packet classification logic for the above configuration is as follows: 1. Match packets against match-any qos-AF4. If a match exists, queue the packet as AF4 in Queue 4, and if no match exists, go to the next class map. 2. Match packets against match-any qos-AF3. If a match exists, queue the packet as AF3 in Queue 3, and if no match exists, go to the next class map. 3. Match packets against match-all qos-BE1. If a match exists, queue the packet as BE1, and if no match exists, queue the packets to the default queue, Queue 0. You can optionally classify packets using their DSCP marking, instead of placing packets in Queue 0, if no match occurs. In the above example, if no match occurs against match-all qos-BE1, the classification logic continues: 4. Queue the packet according to the DSCP marking. The DSCP to Queue mapping will be as per the Table 30-5. The behavior is similar for trust dot1p fallback in a Layer2 input policy map; the dot1p-to-queue mapping is according to Table 30-6. To enable Fall Back to trust diffserve or dot1p:
Task Classify packets according to their DSCP value as a secondary option in case no match occurs against the configured class maps. Command Syntax
trust {diffserve | dot1p} fallback

Command Mode POLICY-MAP-IN

Quality of Service | 587

www.dell.com | support.dell.com

Mapping dot1p values to service queues

Mapping dot1p values to service queues is available only on platforms:

cs

On the C-Series and S-Series all traffic is by default mapped to the same queue, Queue 0. If you honor dot1p on ingress, then you can create service classes based the queueing strategy in Table 30-6 using the command service-class dynamic dot1p from INTERFACE mode. You may apply this queuing strategy globally by entering this command from CONFIGURATION mode. All dot1p traffic is mapped to Queue 0 unless service-class dynamic dot1p is enabled on an interface or globally. Layer 2 or Layer 3 service policies supersede dot1p service classes.

Guaranteeing bandwidth to dot1p-based service queues

Guarantee a minimum bandwidth to queues globally from CONFIGURATION mode with the command service-class bandwidth-weight. The command is applied in the same way as the bandwidth-weight command in an output QoS policy (see Allocate bandwidth to queue). The bandwidth-weight command in QOS-POLICY-OUT mode supersedes the service-class bandwidth-weight command.

Apply an input policy map to an interface

Apply an input policy map to an interface using the command service-policy input from INTERFACE mode. Specify the keyword layer2 if the policy map you are applying a Layer 2 policy map; in this case, the INTERFACE must be in switchport mode. You can apply the same policy map to multiple interfaces, and you can modify a policy map after you apply it. You cannot apply a class-map and QoS policies to the same interface. You cannot apply an input Layer 2 QoS policy on an interface you also configure with vlan-stack access. If you apply a service policy that contains an ACL to more than one interface, FTOS uses ACL optimization to conserves CAM space. The ACL Optimization behavior detects when an ACL already exists in the CAM and rather than writing it to the CAM multiple times.

Create Output Policy Maps

Create Output Policy Maps is supported only on platform

1. Create an output policy map using the command policy-map-output from CONFIGURATION mode. 2. Once you create an output policy map, do one or more of the following: Apply an output QoS policy to a queue Specify an aggregate QoS policy Apply an output policy map to an interface

3. Apply the policy map to an interface. See page 61.

588

Quality of Service

Apply an output QoS policy to a queue

Apply an output QoS policy to queues using the command service-queue from INTERFACE mode.

Specify an aggregate QoS policy

Specify an aggregate QoS policy using the command policy-aggregate from POLICY-MAP-OUT mode.

Apply an output policy map to an interface

Apply an input policy map to an interface using the command service-policy output from INTERFACE mode. You can apply the same policy map to multiple interfaces, and you can modify a policy map after you apply it.

QoS Rate Adjustment

The Ethernet packet format consists of: Preamble: 7 bytes Preamble Start Frame Delimiter (SFD): 1 byte Destination MAC Address: 6 bytes Source MAC Address: 6 bytes Ethernet Type/Length: 2 bytes Payload: (variable) Cyclic Redundancy Check (CRC): 4 bytes Inter-frame Gap (IFG): (variable)

By default, while rate limiting, policing, and shaping, FTOS does not include the Preamble, SFD, or the IFG fields. These fields are overhead; only the fields from MAC Destination Address to the CRC are used for forwarding and are included in these rate metering calculations. You can optionally include overhead fields in rate metering calculations by enabling QoS Rate Adjustment. QoS Rate Adjustment is disabled by default, and no qos-rate-adjust is listed in the running-configuration.
Task Include a specified number of bytes of packet overhead to include in rate limiting, policing, and shaping calculations. For example, to include the Preamble and SFD, enter qos-rate-adjust 8. For variable length overhead fields you must know the number of bytes you want to include. Command Syntax
qos-rate-adjust overhead-bytes

Command Mode CONFIGURATION

Default: Disabled C-Series and S-Series Range: 1-31 E-Series Range: 1-144

Quality of Service | 589

www.dell.com | support.dell.com

Strict-priority Queueing
You can assign strict-priority to one unicast queue, 1-7, using the command strict-priority from CONFIGURATION mode. Strict-priority means that FTOS dequeues all packets from the assigned queue before servicing any other queues. The strict-priority supersedes bandwidth-percentage an bandwidth-weight percentage configurations. A queue with strict-priority can starve other queues in the same port-pipe. On the E-Series, this configuration is applied to the queue on both ingress and egress.

Weighted Random Early Detection

Weighted Random Early Detection is supported only on platform

Weighted Random Early Detection (WRED) congestion avoidance mechanism that drops packets to prevent buffering resources from being consumed. Traffic is a mixture of various kinds of packets. The rate at which some types of packets arrive might be greater than others. In this case, the space on the BTM (ingress or egress) can be consumed by only one or a few types of traffic, leaving no space for other types. A WRED profile can be applied to a policy-map so that specified traffic can be prevented from consuming too much of the BTM resources. WRED uses a profile to specify minimum and maximum threshold values. The minimum threshold is the allotted buffer space for specified traffic, for example 1000KB on egress. If the 1000KB is consumed, packets will be dropped randomly at an exponential rate until the maximum threshold is reached (Figure 30-13); this is the early detection part of WRED. If the maximum threshold2000KB, for exampleis reached, then all incoming packets are dropped until less than 2000KB of buffer space is consumed by the specified traffic.
Figure 30-13. Packet Drop Rate for WREDl
All Pckts

No Packets Buffered

Packet Drop Rate

Allotted Space

Early Warning

0 Pckts 0KB Min Max Buffer Space fnC0045mp Total Buffer Space

590

Quality of Service

You can create a custom WRED profile or use on of the five pre-defined profiles listed in Table 30-7.
Table 30-7. Pre-defined WRED Profiles Maximum Threshold 0 2048 4096 8192 16384

Default Profile Minimum Name Threshold wred_drop wred_ge_y wred_ge_g wred_teng_y wred_teng_g 0 1024 2048 4096 8192

Create WRED Profiles

To create a WRED profile: 1. Create a WRED profile using the command wred from CONFIGURATION mode. 2. The command wred places you in WRED mode. From this mode, specify minimum and maximum threshold values using the command threshold.

Apply a WRED profile to traffic

Once you create a WRED profile you must specify to which traffic FTOS should apply the profile. FTOS assigns a color (also called drop precedence)red, yellow, or greento each packet based on it DSCP value before queuing it. DSCP is a 6 bit field. Dell Networking uses the first three bits of this field (DP) to determine the drop precedence. DP values of 110 and 100 map to yellow, and all other values map to green. If you do not configure FTOS to honor DSCP values on ingress (Honor DSCP values on ingress packets) see all traffic defaults to green drop precedence. Assign a WRED profile to either yellow or green traffic from QOS-POLICY-OUT mode using the command wred.

Configure WRED for Storm Control

Configure WRED for Storm Control is supported only on platform

Storm control limits the percentage of the total bandwidth that broadcast traffic can consume on an interface (if configured locally) or on all interfaces (if configured globally). For storm-control broadcast 50 out, the total bandwidth that broadcast traffic can consume on egress on a 1Gbs interface is 512Mbs. The method by which packets are selected to be dropped is the "tail-drop" method, where packets exceeding the specified rate are dropped.

Quality of Service | 591

www.dell.com | support.dell.com

WRED can be used in combination with storm control to regulate broadcast and unknown-unicast traffic. This feature is available through an additional option in command storm-control [broadcast | unknown-unicast] at CONFIGURATION. See the FTOS Command Line Reference for information on using this command. Using the command storm-control broadcast 50 out wred-profile, for example, first the total bandwidth that broadcast traffic can consume is reduced to 50% of line rate. Even though broadcast traffic is restricted, the rate of outgoing broadcast traffic might be greater than other traffic, and if so, broadcast packets would consume too much buffer space. So, the wred-profile option is added to limit the amount of buffer space that broadcast traffic can consume.

Display Default and Configured WRED Profiles

wred-profile

Display default and configured WRED profiles and their threshold values using the command show qos from EXEC mode, as shown in Figure 30-14.
Figure 30-14. Displaying WRED Profiles
FTOS#show qos wred-profile Wred-profile-name wred_drop wred_ge_y wred_ge_g wred_teng_y wred_teng_g min-threshold 0 1000 2000 4000 8000 max-threshold 0 2000 4000 8000 16000

Display WRED Drop Statistics

Display the number of packets FTOS dropped by WRED Profile using the command show qos statistics from EXEC Privilege mode, as shown in Figure 30-15.

592

Quality of Service

Figure 30-15.

show qos statistics Command Example

FTOS#show qos statistics wred-profile Interface Gi 5/11 Queue# Drop-statistic WRED-name 0 Green Yellow Out of Green Yellow Out of Green Yellow Out of Green Yellow Out of Green Yellow Out of Green Yellow Out of Green Yellow Out of Green Yellow Out of WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile WRED1 WRED2 Profile

Min 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20

Max 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

Dropped Pkts 51623 51300 0 52082 51004 0 50567 49965 0 50477 49815 0 50695 49476 0 50245 49535 0 50033 49595 0 50474 49522 0

FTOS#

FTOS Behavior: The C-Series fetches the per-queue packet count via class-maps. The count is the number of packets matching the ACL entries in class-map. Every time the class-map or policy-map is modified, the ACL entries are re-written to the Forwarding Processor, and the queue statistics are cleared. This behavior is different from the E-Series. The E-Series fetches the packet count directly from counters at each queue, which allows queue statistics to persist until explicitly cleared via the CLI.

Allocating Bandwidth to Multicast Queues

Allocating Bandwidth to Multicast Queues is supported on platform:

The E-Series has 128 multicast queues per port-pipe, which are transparent, and eight unicast queues per port. You can allocate a specific bandwidth percentage per port-pipe to multicast traffic using the command queue egress multicast bandwidth-percentage from CONFIGURATION mode. If you configure bandwidth-percentage for unicast only, 1/8 of the port bandwidth is reserved for multicast, and the remaining bandwidth is distributed based on your configuration. If you configure multicast bandwidth, after assigning the specified amount of bandwidth to multicast the remaining bandwidth is distributed according to the WFQ algorithm. If you configure bandwidth-percentage for both unicast and multicast, then bandwidth is assigned based on your configuration for multicast then unicast (based on the remaining available bandwidth), and the remaining bandwidth is distributed among the other queues.

Quality of Service | 593

www.dell.com | support.dell.com

For example, if you configure 70% bandwidth to multicast, 80% bandwidth to one queue in unicast and 0 % to all remaining unicast queues, then first, FTOS assigns 70% bandwidth to multicast, then FTOS derives the 80% bandwidth for unicast from the remaining 30% of total bandwidth.

Pre-calculating Available QoS CAM Space

Pre-calculating Available QoS CAM Space is supported on platforms:

ces

Before version 7.3.1 there was no way to measure the number of CAM entries a policy-map would consume (the number of CAM entries that a rule uses is not predictable; 1 to 16 entries might be used per rule depending upon its complexity). Therefore, it was possible to apply to an interface a policy-map that requires more entries than are available. In this case, the system writes as many entries as possible, and then generates an CAM-full error message (Message 1). The partial policy-map configuration might cause unintentional system behavior.
Message 1 QoS CAM Region Exceeded
%EX2YD:12 %DIFFSERV-2-DSA_QOS_CAM_INSTALL_FAILED: Not enough space in L3 Cam(PolicyQos) for class 2 (Gi 12/20) entries on portpipe 1 for linecard 12 %EX2YD:12 %DIFFSERV-2DSA_QOS_CAM_INSTALL_FAILED: Not enough space in L3 Cam(PolicyQos) for class 5 (Gi 12/22) entries on portpipe 1 for linecard 12

The command test cam-usage enables you to verify that there are enough available CAM entries before applying a policy-map to an interface so that you avoid exceeding the QoS CAM space and partial configurations. This command measures the size of the specified policy-map and compares it to the available CAM space in a partition for a specified port-pipe. Test the policy-map size against the CAM space for a specific port-pipe or all port-pipes using these commands:
test cam-usage service-policy input policy-map {linecard | stack-unit } number port-set number test cam-usage service-policy input policy-map {linecard | stack-unit } all

The output of this command, shown in Figure 30-16, displays: the estimated number of CAM entries the policy-map will consume whether or not the policy-map can be applied the number of interfaces in a port-pipe to which the policy-map can be applied

Specifically: Available CAM is the available number of CAM entries in the specified CAM partition for the specified line card or stack-unit port-pipe. Estimated CAM is the estimated number of CAM entries that the policy will consume when it is applied to an interface.

594

Quality of Service

Status indicates whether or not the specified policy-map can be completely applied to an interface in the port-pipe. Allowed indicates that the policy-map can be applied because the estimated number of CAM entries is less or equal to the available number of CAM entries. The number of interfaces in the port-pipe to which the policy-map can be applied is given in parenthesis. Exception indicates that the number of CAM entries required to write the policy-map to the CAM is greater than the number of available CAM entries, and therefore the policy-map cannot be applied to an interface in the specified port-pipe.
Note: The command show cam-usage provides much of the same information as test cam-usage, but whether or not a policy-map can be successfully applied to an interface cannot be determined without first measuring how many CAM entries the policy-map would consume; the command test cam-usage is useful because it provides this measurement.

Figure 30-16.

test cam-usage Command Example

FTOS# test cam-usage service-policy input pmap_l2 linecard 0 port-set 0 Linecard | Port-pipe | CAM Partition | Available CAM | Estimated CAM | Status =============================================================================== 0 0 L2ACL 500 200 Allowed(2)

Viewing QoS CAM Entries

Viewing QoS CAM Entries is supported only on platform

View Layer 2 QoS CAM entries using the command show cam layer3-qos from EXEC Privilege mode. View Layer 3 QoS CAM entries using the command show cam layer2-qos from EXEC Privilege mode.

Configure Quality of Service for an Office VOIP Deployment

Configure Quality of Service for an Office VOIP Deployment is supported on platforms

cs

There are multiple ways you can use QoS to map ingress phone and PC traffic so that you can give them each a different quality of service.

Honor the incoming DSCP value

On both the C-Series or S-Series, if you know traffic originating from the phone is tagged with the DSCP value of 46 (EF), you might make the associated queue a strict priority queue, as shown in Figure 30-17; on the C-Series and S-Series, FTOS maps DSCP 46 to queue 2 (see Table 30-5 on page 585 in the QoS chapter.)

Quality of Service | 595

www.dell.com | support.dell.com

Figure 30-17.

Honoring the DSCP Value on Incoming Voice Data

FTOS#sh run policy-map-input ! policy-map-input HonorDSCP trust diffserv FTOS#sh run int gigabitethernet 6/11 ! interface GigabitEthernet 6/11 description "IP Phone X" no ip address portmode hybrid switchport service-policy input HonorDSCP power inline auto no shutdown FTOS#sh run | grep strict-priority strict-priority unicast 2

Honor the incoming dot1p value

On the C-Series, if you know traffic originating from the phone is tagged with a dot1p value of 5, you might make the associated queue a strict priority queue, as shown in Figure 30-18; on the C-Series, FTOS maps dot1p priority 5 to queue 2.
Figure 30-18. Honoring the Dot1P Value on Incoming Voice Traffic

FTOS#sh run int gi 6/10 ! interface GigabitEthernet 6/10 description "IP Phone X" no ip address portmode hybrid switchport service-class dynamic dot1p power inline auto no shutdown FTOS#sh run | grep strict-priority strict-priority unicast 2

Classifying VOIP traffic and applying QoS policies

Avoid congestion and give precedence to voice and signaling traffic by classifying traffic based on subnet and using strict priority and bandwidth weights on egress, as outlined in the steps below. Figure 30-19 depicts the topology and shows the configuration for a C-Series. The steps are the same on an S-Series. Figure 30-20 on page 598 is a screenshot showing some of the steps and the resulting running-config.

596

Quality of Service

Figure 30-19.

Classifying VOIP Traffic and Applying QoS Policies for an Office VOIP Deployment
FTOS#sh run int gi 6/10 ! interface GigabitEthernet 6/10 description "IP Phone X no ip address portmode hybrid switchport service-policy input phone-pc power inline auto no shutdown FTOS#sh run int gi 6/2 ! interface GigabitEthernet 6/2 description "Uplink to E1200" no ip address switchport service-policy output BW no shutdown

VLAN 300: Voice Signaling VLAN 200: Voice PC data VLAN 100: Data

queue 2, bandwidth weight 64 queue 3, strict priority

6/10 6/2

queue 1, bandwidth weight 8

LLDP-MED advertisements

PC

IP Phone X

FTOS#sh run lldp protocol lldp C-Series advertise med advertise med voice 200 6 46 advertise med voice-signaling 300 5 28 no disable

E-Series

Step 1

Task Create three standard or extended access-lists, one each for voice, voice signaling, and PC data, and place each in its own match-any class-map. Create an input policy-map containing all three class-maps, and assign each class-map a different service queue. Create two input QoS policies, one each for PC data and voice signaling. Assign a different bandwidth weight to each policy. Create an output policy map containing both QoS policies, and assign them to different service queues. Assign a strict priority to unicast traffic in queue 3. Apply the input policy map you created in Step 2 to the interface connected to the phone, and apply the output policy map you created in Step 4 to the interface connected your desired next-hop router.

Command
ip access-list class-map match-any policy-map-input service-queue

Command Mode CONFIGURATION CLASS-MAP CONFIGURATION POLICY-MAP-IN CONFIGURATION QOS-POLICY-IN CONFIGURATION POLICY-MAP-OUT CONFIGURATION INTERFACE

qos-policy-out

bandwidth-weight
policy-map-out service-queue

5 6

strict-priority service-policy

Figure 30-20 on page 598 is a screenshot showing some of the steps, above, and the resulting running-config.

Quality of Service | 597

www.dell.com | support.dell.com

Figure 30-20.

Classifying VOIP Traffic and Applying QoS Policies for an Office VOIP Deployment

FTOS#sh run acl ! ip access-list extended pc-subnet seq 5 permit ip 201.1.1.0/24 any ! ip access-list extended phone-signalling seq 5 permit ip 192.1.1.0/24 host 192.1.1.1 ! ip access-list extended phone-subnet seq 5 permit ip 192.1.1.0/24 any FTOS#sh run class-map ! class-map match-any pc-subnet match ip access-group pc-subnet ! class-map match-any phone-signalling match ip access-group phone-signalling ! class-map match-any phone-subnet match ip access-group phone-subnet FTOS#sh run policy-map-input ! policy-map-input phone-pc service-queue 1 class-map pc-subnet service-queue 2 class-map phone-signalling service-queue 3 class-map phone-subnet FTOS#sh run qos-policy-output ! qos-policy-output data bandwidth-weight 8 ! qos-policy-output signalling bandwidth-weight 64 FTOS#sh run policy-map-output ! policy-map-output BW service-queue 1 qos-policy data service-queue 2 qos-policy signalling FTOS#sh run | grep strict-p strict-priority unicast 3 FTOS#sh run int gi 6/10 ! interface GigabitEthernet 6/10 description "IP Phone X no ip address portmode hybrid switchport service-policy input phone-pc power inline auto no shutdown FTOS#sh run int gi 6/2 ! interface GigabitEthernet 6/2 description "Uplink to E1200" no ip address switchport service-policy output BW no shutdown

598

Quality of Service

31
Routing Information Protocol
Routing Information Protocol is supported only on platforms:

ce s

RIP is supported on the S-Series following the release of FTOS version 7.8.1.0, and on the C-Series with FTOS versions 7.6.1.0 and after. Routing Information Protocol (RIP) is based on a distance-vector algorithm, it tracks distances or hop counts to nearby routers when establishing network connections. Protocol Overview Implementation Information Configuration Information RIP Configuration Example

RIP protocol standards are listed in the Appendix , Standards Compliance chapter.

Protocol Overview
RIP is the oldest interior gateway protocol. There are two versions of RIP: RIP version 1 (RIPv1) and RIP version 2 (RIPv2). These versions are documented in RFCs 1058 and 2453.

RIPv1
RIPv1 learns where nodes in a network are located by automatically constructing a routing data table. The routing table is established after RIP sends out one or more broadcast signals to all adjacent nodes in a network. Hop counts of these signals are tracked and entered into the routing table, which defines where nodes in the network are located. The information that is used to update the routing table is sent as either a request or response message. In RIPv1, automatic updates to the routing table are performed as either one-time requests or periodic responses (every 30 seconds). RIP transports its responses or requests by means of UDP over port 520.

Routing Information Protocol | 599

www.dell.com | support.dell.com

RIP must receive regular routing updates to maintain a correct routing table. Response messages containing a routers full routing table are transmitted every 30 seconds. If a router does not send an update within a certain amount of time, the hop count to that route is changed to unreachable (a route hop metric of 16 hops). Another timer sets the amount of time before the unreachable routes are removed from the routing table. This first RIP version does not support VLSM or CIDR and is not widely used.

RIPv2
RIPv2 adds support for subnet fields in the RIP routing updates, thus qualifying it as a classless routing protocol. The RIPv2 message format includes entries for route tags, subnet masks, and next hop addresses. Another enhancement included in RIPv2 is multicasting for route updates on IP multicast address 224.0.0.9.

Implementation Information
FTOS supports both versions of RIP and allows you to configure one version globally and the other version or both versions on the interfaces. The C-Series and E-Series both support 1,000 RIP routes. Table 31-1 displays the defaults for RIP in FTOS.
Table 31-1. Feature Interfaces running RIP RIP timers RIP Defaults in FTOS Default Listen to RIPv1 and RIPv2 Transmit RIPv1 update timer = 30 seconds invalid timer = 180 seconds holddown timer = 180 seconds flush timer = 240 seconds Enabled 16

Auto summarization ECMP paths supported

Configuration Information
By default, RIP is disabled in FTOS. To configure RIP, you must use commands in two modes: ROUTER RIP and INTERFACE. Commands executed in the ROUTER RIP mode configure RIP globally, while commands executed in the INTERFACE mode configure RIP features on that interface only. RIP is best suited for small, homogeneous networks. All devices within the RIP network must be configured to support RIP if they are to participate in the RIP.

600

Routing Information Protocol

Configuration Task List for RIP

Enable RIP globally (mandatory) Configure RIP on interfaces (optional) Control RIP routing updates (optional) Set send and receive version (optional) Generate a default route (optional) Control route metrics (optional) Summarize routes (optional) Control route metrics Debug RIP

For a complete listing of all commands related to RIP, refer to the FTOS Command Reference.

Enable RIP globally

By default, RIP is not enabled in FTOS. To enable RIP, use the following commands in sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
router rip network ip-address

Command Mode CONFIGURATION ROUTER RIP

Purpose Enter ROUTER RIP mode and enable the RIP process on FTOS. Assign an IP network address as a RIP network to exchange routing information. You can use this command multiple times to exchange RIP information with as many RIP networks as you want.

After designating networks with which the system is to exchange RIP information, ensure that all devices on that network are configured to exchange RIP information. The FTOS default is to send RIPv1, and to receive RIPv1 and RIPv2. To change the RIP version globally, use the version command in the ROUTER RIP mode. When RIP is enabled, you can view the global RIP configuration by using the show running-config command in the EXEC mode or the show config command (Figure ) in the ROUTER RIP mode.
Figure 31-1. show config Command Example in ROUTER RIP mode

FTOS(conf-router_rip)#show config ! router rip network 10.0.0.0 FTOS(conf-router_rip)#

Routing Information Protocol | 601

www.dell.com | support.dell.com

When the RIP process has learned the RIP routes, use the show ip rip database command in the EXEC mode to view those routes (Figure 385).
Figure 31-2. show ip rip database Command Example (Partial)
FTOS#show ip rip database Total number of routes in RIP database: 978 160.160.0.0/16 [120/1] via 29.10.10.12, 00:00:26, Fa 160.160.0.0/16 auto-summary 2.0.0.0/8 [120/1] via 29.10.10.12, 00:01:22, Fa 2.0.0.0/8 auto-summary 4.0.0.0/8 [120/1] via 29.10.10.12, 00:01:22, Fa 4.0.0.0/8 auto-summary 8.0.0.0/8 [120/1] via 29.10.10.12, 00:00:26, Fa 8.0.0.0/8 auto-summary 12.0.0.0/8 [120/1] via 29.10.10.12, 00:00:26, Fa 12.0.0.0/8 auto-summary 20.0.0.0/8 [120/1] via 29.10.10.12, 00:00:26, Fa 20.0.0.0/8 auto-summary 29.10.10.0/24 directly connected,Fa 29.0.0.0/8 auto-summary 31.0.0.0/8 [120/1] via 29.10.10.12, 00:00:26, Fa 31.0.0.0/8 auto-summary 192.162.2.0/24 [120/1] via 29.10.10.12, 00:01:21, Fa 192.162.2.0/24 auto-summary 192.161.1.0/24 [120/1] via 29.10.10.12, 00:00:27, Fa 192.161.1.0/24 auto-summary 192.162.3.0/24 [120/1] via 29.10.10.12, 00:01:22, Fa 192.162.3.0/24 auto-summary

0/0

0/0

0/0

0/0

0/0

0/0 0/0

0/0

0/0

0/0

0/0

To disable RIP globally, use the no router rip command in the CONFIGURATION mode.

Configure RIP on interfaces

When you enable RIP globally on the system, interfaces meeting certain conditions start receiving RIP routes. By default, interfaces that are enabled and configured with an IP address in the same subnet as the RIP network address receive RIPv1 and RIPv2 routes and send RIPv1 routes. Assign IP addresses to interfaces that are part of the same subnet as the RIP network identified in the network command syntax.

602

Routing Information Protocol

Control RIP routing updates

By default, RIP broadcasts routing information out all enabled interfaces, but you can configure RIP to send or to block RIP routing information, either from a specific IP address or a specific interface. To control which devices or interfaces receive routing updates, you must configure a direct update to one router and configure interfaces to block RIP updates from other sources. To control the source of RIP route information, use the following commands, in the ROUTER RIP mode:
Command Syntax
neighbor ip-address

Command Mode ROUTER RIP

Purpose Define a specific router to exchange RIP information between it and the Dell Networking system. You can use this command multiple times to exchange RIP information with as many RIP networks as you want. Disable a specific interface from sending or receiving RIP routing information.

passive-interface interface

ROUTER RIP

Another method of controlling RIP (or any routing protocol) routing information is to filter the information through a prefix list. A prefix lists is applied to incoming or outgoing routes. Those routes must meet the conditions of the prefix list; if not, FTOS drops the route. Prefix lists are globally applied on all interfaces running RIP. Configure the prefix list in the PREFIX LIST mode prior to assigning it to the RIP process. For configuration information on prefix lists, see Chapter 17, IP Access Control Lists, Prefix Lists, and Route-maps, on page 47. To apply prefix lists to incoming or outgoing RIP routes, use the following commands in the ROUTER RIP mode:
Command Syntax
distribute-list prefix-list-name in distribute-list prefix-list-name out

Command Mode ROUTER RIP ROUTER RIP

Purpose Assign a configured prefix list to all incoming RIP routes. Assign a configured prefix list to all outgoing RIP routes.

In addition to filtering routes, you can add routes from other routing instances or protocols to the RIP process. With the redistribute command syntax, you can include OSPF, static, or directly connected routes in the RIP process.

Routing Information Protocol | 603

www.dell.com | support.dell.com

To add routes from other routing instances or protocols, use any of the following commands in the ROUTER RIP mode:
Command Syntax
redistribute {connected | static} [metric metric-value] [route-map map-name]

Command Mode ROUTER RIP

Purpose Include directly connected or user-configured (static) routes in RIP. metric range: 0 to 16 map-name: name of a configured route map. Include IS-IS routes in RIP. metric range: 0 to 16 map-name: name of a configured route map. Note: IS-IS is not supported on the S-Series platform. Include specific OSPF routes in RIP. Configure the following parameters: process-id range: 1 to 65535 metric range: 0 to 16 map-name: name of a configured route map.

redistribute isis [level-1 | level-1-2 | level-2] [metric metric-value] [route-map map-name]

ROUTER RIP

redistribute ospf process-id [match external {1 | 2} | match internal] [metric value] [route-map map-name]

ROUTER RIP

To view the current RIP configuration, use the show running-config command in the EXEC mode or the show config command in the ROUTER RIP mode.

Set send and receive version

To specify the RIP version, use the version command in the ROUTER RIP mode. To set an interface to receive only one or the other version, use the ip rip send version or the ip rip receive version commands in the INTERFACE mode. To change the RIP version globally in FTOS, use the following command in the ROUTER RIP mode:
Command Syntax
version {1 | 2}

Command Mode ROUTER RIP

Purpose Set the RIP version sent and received on the system.

You can set one RIP version globally on the system. This command sets the RIP version for RIP traffic on the interfaces participating in RIP unless the interface was specifically configured for a specific RIP version. Use the show config command in the ROUTER RIP mode to see whether the version command is configured. You can also use the show ip protocols command in the EXEC mode to view the routing protocols configuration.

604

Routing Information Protocol

Figure 31-3 shows an example of the RIP configuration after the ROUTER RIP mode version command is set to RIPv2. When the ROUTER RIP mode version command is set, the interface (GigabitEthernet 0/0) participating in the RIP process is also set to send and receive RIPv2.
Figure 31-3. show ip protocols Command Example
FTOS#show ip protocols Routing Protocols is RIP Sending updates every 30 seconds, next due in 23 Invalid after 180 seconds, hold down 180, flushed after 240 Output delay 8 milliseconds between packets Automatic network summarization is in effect Outgoing filter for all interfaces is Incoming filter for all interfaces is Default redistribution metric is 1 Default version control: receive version 2, send version 2 Interface Recv Send GigabitEthernet 0/0 2 2 Routing for Networks: 10.0.0.0 Routing Information Sources: Gateway Distance Distance: (default is 120) FTOS#

RIPv2 configured globally and on the interface.

Last Update

To configure the interfaces to send or receive different RIP versions from the RIP version configured globally, use either of the following commands in the INTERFACE mode:
Command Syntax
ip rip receive version [1] [2] ip rip send version [1] [2]

Command Mode INTERFACE INTERFACE

Purpose Set the RIP version(s) received on that interface. Set the RIP version(s) sent out on that interface.

To configure an interface to receive or send both versions of RIP, include 1 and 2 in the command syntax. Figure 31-4 displays the command syntax for sending both RIPv1 and RIPv2 and receiving only RIPv2.
Figure 31-4. Configuring an interface to send both versions of RIP
FTOS(conf-if)#ip rip send version 1 2 FTOS(conf-if)#ip rip receive version 2

The show ip protocols command example Figure 31-5 confirms that both versions are sent out that interface. This interface no longer sends and receives the same RIP versions as FTOS does globally.

Routing Information Protocol | 605

www.dell.com | support.dell.com

Figure 31-5. show ip protocols Command Example

FTOS#show ip protocols Routing Protocols is RIP Sending updates every 30 seconds, next due in 11 Invalid after 180 seconds, hold down 180, flushed after 240 Output delay 8 milliseconds between packets Automatic network summarization is in effect Outgoing filter for all interfaces is Incoming filter for all interfaces is Default redistribution metric is 1 Default version control: receive version 2, send version 2 Interface Recv Send Different RIP versions FastEthernet 0/0 2 1 2 Routing for Networks: configured for this 10.0.0.0 interface Routing Information Sources: Gateway Distance Distance: (default is 120) FTOS# Last Update

RIPv2 configured globally

Generate a default route

Traffic is forwarded to the default route when the traffics network is not explicitly listed in the routing table. Default routes are not enabled in RIP unless specified. Use the default-information originate command in the ROUTER RIP mode to generate a default route into RIP. In FTOS, default routes received in RIP updates from other routes are advertised if the default-information originate command is configured. To configure FTOS to generate a default route, use the following command in the ROUTER RIP mode:
Command Syntax
default-information originate [always] [metric value] [route-map route-map-name]

Command Mode ROUTER RIP

Purpose Specify the generation of a default route in RIP. Configure the following parameters: always: enter this keyword to always generate a default route. value range: 1 to 16. route-map-name: name of a configured route map.

Use the show config command in the ROUTER RIP mode to confirm that the default route configuration is completed.

606

Routing Information Protocol

Summarize routes
Routes in the RIPv2 routing table are summarized by default, thus reducing the size of the routing table and improving routing efficiency in large networks. By default, the autosummary command in the ROUTER RIP mode is enabled and summarizes RIP routes up to the classful network boundary. If you must perform routing between discontiguous subnets, disable automatic summarization. With automatic route summarization disabled, subnets are advertised. The command autosummary requires no other configuration commands. To disable automatic route summarization, in the ROUTER RIP mode, enter no autosummary.
Note: If the ip split-horizon command is enabled on an interface, then the system does not advertise the summarized address.

Control route metrics

As a distance-vector protocol, RIP uses hop counts to determine the best route, but sometimes the shortest hop count is a route over the lowest-speed link. To manipulate RIP routes so that the routing protocol prefers a different route, you must manipulate the route by using the offset command. Exercise caution when applying an offset command to routers on a broadcast network, as the router using the offset command is modifying RIP advertisements before sending out those advertisements. The distance command also allows you to manipulate route metrics. Use the command to assign different weights to routes so that the ones with the lower weight or administrative distance assigned are preferred. To set route metrics, use either of the following commands in the ROUTER RIP mode:
Command Syntax
distance weight [ip-address mask [access-list-name]]

Command Mode ROUTER RIP

Purpose Apply a weight to all routes or a specific route and ACL. Configure the following parameters: weight range: 1 to 255 (default is 120) ip-address mask: the IP address in dotted decimal format (A.B.C.D), and the mask in slash format (/x). access-list-name: name of a configured IP ACL. Apply an additional number to the incoming or outgoing route metrics. Configure the following parameters: access-list-name: the name of a configured IP ACL offset range: 0 to 16. interface: the type, slot, and number of an interface.

offset access-list-name {in | out} offset [interface]

ROUTER RIP

Use the show config command in the ROUTER RIP mode to view configuration changes.

Routing Information Protocol | 607

www.dell.com | support.dell.com

Debug RIP
The debug ip rip command enables RIP debugging. When debugging is enabled, you can view information on RIP protocol changes or RIP routes. To enable RIP debugging, use the following command in the EXEC privilege mode:
Command Syntax
debug ip rip [interface | database | events | trigger]

Command Mode EXEC privilege

Purpose Enable debugging of RIP.

Figure 31-6 shows the confirmation when the debug function is enabled.
Figure 31-6. debug ip rip Command Example
FTOS#debug ip rip RIP protocol debug is ON FTOS#

To disable RIP, use the no debug ip rip command.

RIP Configuration Example

The example in this section shows the command sequence to configure RIPv2 on the two routers shown in Figure 31-7 Core 2 and Core 3. The host prompts used in the example screenshots reflect those names. The screenshots are divided into the following groups of command sequences: Configuring RIPv2 on Core 2 Core 2 Output RIP Configuration on Core 3 Core 3 RIP Output RIP Configuration Summary
GigE 3/43 192.168.1.0 /24 GigE 3/44 192.168.2.0 /24

Figure 31-7. RIP Topology Example

GigE 2/41 10.200.10.0 /24 GigE 2/42 10.300.10.0 /24

Core 2

GigE 2/31

GigE 3/21

Core 3

GigE 2/11 10.11.10.1 /24 GigE 2/31 10.11.20.2 /24

GigE 3/11 10.11.30.1 /24 GigE 3/21 10.11.20.1 /24

608

Routing Information Protocol

Configuring RIPv2 on Core 2

Figure 31-8. Configuring RIPv2 on Core 2
Core2(conf-if-gi-2/31)# Core2(conf-if-gi-2/31)#router rip Core2(conf-router_rip)#ver 2 Core2(conf-router_rip)#network 10.200.10.0 Core2(conf-router_rip)#network 10.300.10.0 Core2(conf-router_rip)#network 10.11.10.0 Core2(conf-router_rip)#network 10.11.20.0 Core2(conf-router_rip)#show config ! router rip network 10.0.0.0 version 2 Core2(conf-router_rip)#

Core 2 Output
The screenshots in this section are: Figure 31-9: Using show ip rip database command to display Core 2 RIP database Figure 31-10: Using show ip route command to display Core 2 RIP setup Figure 31-11: Using show ip protocols command to display Core 2 RIP activity

Figure 31-9. Example of RIP Configuration Response from Core 2

Core2(conf-router_rip)#end 00:12:24: %RPM0-P:CP %SYS-5-CONFIG_I: Configured from console by Core2#show ip rip database Total number of routes in RIP database: 7 10.11.30.0/24 [120/1] via 10.11.20.1, 00:00:03, GigabitEthernet 2/31 10.300.10.0/24 directly connected,GigabitEthernet 2/42 10.200.10.0/24 directly connected,GigabitEthernet 2/41 10.11.20.0/24 directly connected,GigabitEthernet 2/31 10.11.10.0/24 directly connected,GigabitEthernet 2/11 10.0.0.0/8 auto-summary 192.168.1.0/24 [120/1] via 10.11.20.1, 00:00:03, GigabitEthernet 2/31 192.168.1.0/24 auto-summary 192.168.2.0/24 [120/1] via 10.11.20.1, 00:00:03, GigabitEthernet 2/31 192.168.2.0/24 auto-summary Core2#

console

Routing Information Protocol | 609

www.dell.com | support.dell.com

Figure 31-10.

Using show ip route Command to Show RIP Configuration on Core 2

Core2#show ip route Codes: C - connected, S - static, R - RIP, B - BGP, IN - internal BGP, EX - external BGP,LO - Locally Originated, O - OSPF, IA - OSPF inter area, N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2, E1 - OSPF external type 1, E2 - OSPF external type 2, i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, IA - IS-IS inter area, * - candidate default, > - non-active route, + - summary route Gateway of last resort is not set Destination Gateway ----------------C 10.11.10.0/24 Direct, Gi 2/11 C 10.11.20.0/24 Direct, Gi 2/31 R 10.11.30.0/24 via 10.11.20.1, C 10.200.10.0/24 Direct, Gi 2/41 C 10.300.10.0/24 Direct, Gi 2/42 R 192.168.1.0/24 via 10.11.20.1, R 192.168.2.0/24 via 10.11.20.1, Core2# R 192.168.1.0/24 via 10.11.20.1, R 192.168.2.0/24 via 10.11.20.1, Core2# Dist/Metric Last Change ----------- ----------0/0 00:02:26 0/0 00:02:02 Gi 2/31 120/1 00:01:20 0/0 00:03:03 0/0 00:02:42 Gi 2/31 120/1 00:01:20 Gi 2/31 120/1 00:01:20 Gi 2/31 Gi 2/31 120/1 120/1 00:05:22 00:05:22

Figure 31-11.

Using show ip protocols Command to Show RIP Configuration Activity on Core 2

Core2#show ip protocols Routing Protocol is "RIP" Sending updates every 30 seconds, next due in 17 Invalid after 180 seconds, hold down 180, flushed after 240 Output delay 8 milliseconds between packets Automatic network summarization is in effect Outgoing filter for all interfaces is Incoming filter for all interfaces is Default redistribution metric is 1 Default version control: receive version 2, send version 2 Interface Recv Send GigabitEthernet 2/42 2 2 GigabitEthernet 2/41 2 2 GigabitEthernet 2/31 2 2 GigabitEthernet 2/11 2 2 Routing for Networks: 10.300.10.0 10.200.10.0 10.11.20.0 10.11.10.0 Routing Information Sources: Gateway Distance 10.11.20.1 120 Distance: (default is 120) Core2#

Last Update 00:00:12

610

Routing Information Protocol

RIP Configuration on Core 3

Figure 31-12. RIP Configuration on Core 3

Core3(conf-if-gi-3/21)#router rip Core3(conf-router_rip)#version 2 Core3(conf-router_rip)#network 192.168.1.0 Core3(conf-router_rip)#network 192.168.2.0 Core3(conf-router_rip)#network 10.11.30.0 Core3(conf-router_rip)#network 10.11.20.0 Core3(conf-router_rip)#show config ! router rip network 10.0.0.0 network 192.168.1.0 network 192.168.2.0 version 2 Core3(conf-router_rip)#

Core 3 RIP Output

The screenshots in this section are: Figure 31-13: Using show ip rip database command to display Core 3 RIP database Figure 31-14: Using show ip route command to display Core 3 RIP setup Figure 31-15: Using show ip protocols command to display Core 3 RIP activity
Using show ip rip database Command for Core 3 RIP Setup

Figure 31-13.

Core3#show ip rip database Total number of routes in RIP database: 7 10.11.10.0/24 [120/1] via 10.11.20.2, 00:00:13, GigabitEthernet 3/21 10.200.10.0/24 [120/1] via 10.11.20.2, 00:00:13, GigabitEthernet 3/21 10.300.10.0/24 [120/1] via 10.11.20.2, 00:00:13, GigabitEthernet 3/21 10.11.20.0/24 directly connected,GigabitEthernet 3/21 10.11.30.0/24 directly connected,GigabitEthernet 3/11 10.0.0.0/8 auto-summary 192.168.1.0/24 directly connected,GigabitEthernet 3/43 192.168.1.0/24 auto-summary 192.168.2.0/24 directly connected,GigabitEthernet 3/44 192.168.2.0/24 auto-summary Core3#

Routing Information Protocol | 611

www.dell.com | support.dell.com

Figure 31-14.

Using show ip routes for Core 3 RIP Setup

Core3#show ip routes Codes: C - connected, S - static, R - RIP, B - BGP, IN - internal BGP, EX - external BGP,LO - Locally Originated, O - OSPF, IA - OSPF inter area, N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2, E1 - OSPF external type 1, E2 - OSPF external type 2, i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, IA - IS-IS inter area, * - candidate default, > - non-active route, + - summary route Gateway of last resort is not set Destination ----------10.11.10.0/24 10.11.20.0/24 10.11.30.0/24 10.200.10.0/24 10.300.10.0/24 192.168.1.0/24 192.168.2.0/24 Gateway ------via 10.11.20.2, Gi 3/21 Direct, Gi 3/21 Direct, Gi 3/11 via 10.11.20.2, Gi 3/21 via 10.11.20.2, Gi 3/21 Direct, Gi 3/43 Direct, Gi 3/44 Dist/Metric Last Change ----------- ----------120/1 00:01:14 0/0 00:01:53 0/0 00:06:00 120/1 00:01:14 120/1 00:01:14 0/0 00:06:53 0/0 00:06:26

R C C R R C C Core3#

Figure 31-15.

Using show ip protocols Command to Show RIP Configuration Activity on Core 3

Core3#show ip protocols Routing Protocol is "RIP" Sending updates every 30 seconds, next due in 6 Invalid after 180 seconds, hold down 180, flushed after 240 Output delay 8 milliseconds between packets Automatic network summarization is in effect Outgoing filter for all interfaces is Incoming filter for all interfaces is Default redistribution metric is 1 Default version control: receive version 2, send version 2 Interface Recv Send GigabitEthernet 3/21 2 2 GigabitEthernet 3/11 2 2 GigabitEthernet 3/44 2 2 GigabitEthernet 3/43 2 2 Routing for Networks: 10.11.20.0 10.11.30.0 192.168.2.0 192.168.1.0 Routing Information Sources: Gateway Distance 10.11.20.2 120 Distance: (default is 120) Core3#

Last Update 00:00:22

612

Routing Information Protocol

RIP Configuration Summary

Figure 31-16. Summary of Core 2 RIP Configuration Using Output of show run Command

! interface GigabitEthernet 2/11 ip address 10.11.10.1/24 no shutdown ! interface GigabitEthernet 2/31 ip address 10.11.20.2/24 no shutdown ! interface GigabitEthernet 2/41 ip address 10.200.10.1/24 no shutdown ! interface GigabitEthernet 2/42 ip address 10.250.10.1/24 no shutdown router rip version 2 10.200.10.0 10.300.10.0 10.11.10.0 10.11.20.0

Figure 31-17.

Summary of Core 3 RIP Configuration Using Output of show run Command

! interface GigabitEthernet 3/11 ip address 10.11.30.1/24 no shutdown ! interface GigabitEthernet 3/21 ip address 10.11.20.1/24 no shutdown ! interface GigabitEthernet 3/43 ip address 192.168.1.1/24 no shutdown ! interface GigabitEthernet 3/44 ip address 192.168.2.1/24 no shutdown

! router rip version 2 network 10.11.20.0 network 10.11.30.0 network 192.168.1.0 network 192.168.2.0

Routing Information Protocol | 613

www.dell.com | support.dell.com |

614

Routing Information Protocol

32
Remote Monitoring
Remote Monitoring is supported on platform

ces

This chapter describes the Remote Monitoring (RMON): Implementation Fault Recovery

Remote Monitoring (RMON) is an industry-standard implementation that monitors network traffic by sharing network monitoring information. RMON provides both 32-bit and 64-bit monitoring facility and long-term statistics collection on Dell Networking Ethernet Interfaces. RMON operates with SNMP and monitors all nodes on a LAN segment. RMON monitors traffic passing through the router and segment traffic not destined for the router. The monitored interfaces may be chosen by using alarms and events with standard MIBs.

Implementation
You must configure SNMP prior to setting up RMON. For a complete SNMP implementation discussion, refer to Chapter 6, Simple Network Management Protocol (SNMP), on page 47. Configuring RMON requires using the RMON CLI and includes the following tasks: Set rmon alarm Configure an RMON event Configure RMON collection statistics Configure RMON collection history Enable an RMON MIB collection history group

RMON implements the following standard RFCs (for details see Appendix , Standards Compliance): RFC-2819 RFC-3273 RFC-3434

Remote Monitoring | 615

www.dell.com | support.dell.com

Fault Recovery
RMON provides the following fault recovery functions: Interface DownWhen an RMON-enabled interface goes down, monitoring continues. However, all data values are registered as 0xFFFFFFFF (32 bits) or ixFFFFFFFFFFFFFFFF (64 bits). When the interface comes back up, RMON monitoring processes resumes.
Note: A Network Management System (NMS) should be ready to interpret a down interface and plot the interface performance graph accordingly.

Line Card DownThe same as Interface Down (see above). RPM Down, RPM FailoverMaster and standby RPMs run the RMON sampling process in the background. Therefore, when an RPM goes down, the other RPM maintains the sampled datathe new master RPM provides the same sampled data as did the old masteras long as the master RPM had been running long enough to sample all the data. NMS backs up all the long-term data collection, and displays the failover downtime from the performance graph. Chassis DownWhen a chassis goes down, all sampled data is lost. But the RMON configurations are saved in the configuration file, and the sampling process continues after the chassis returns to operation. Platform AdaptationRMON supports all Dell Networking chassis and all Dell Networking Ethernet Interfaces.

616

Remote Monitoring

Set rmon alarm

To set an alarm on any MIB object, use the rmon alarm or rmon hc-alarm command in GLOBAL CONFIGURATION mode. To disable the alarm, use the no form of this command:
Command Syntax
[no] rmon alarm number variable interval {delta | absolute} rising-threshold [value event-number] falling-threshold value event-number [owner string]
or

Command Mode CONFIGURATION

Purpose Set an alarm on any MIB object. Use the no form of this command to disable the alarm. Configure the alarm using the following optional parameters: number: Alarm number, should be an integer from 1 to 65,535, the value must be unique in the RMON Alarm Table variable: The MIB object to monitorthe variable must be in the SNMP OID format. For example, 1.3.6.1.2.1.1.3. The object type must be a 32-bit integer for the rmon alarm command and 64 bits for the rmon hc-alarm command. interval: Time in seconds the alarm monitors the MIB variable, the value must be between 1 to 3,600. delta: Tests the change between MIB variables, this is the alarmSampleType in the RMON Alarm table. absolute: Tests each MIB variable directly, this is the alarmSampleType in the RMON Alarm table. rising-threshold value: Value at which the rising-threshold alarm is triggered or reset. For the rmon alarm command this is a 32-bits value, for rmon hc-alarm command this is a 64-bits value. event-number: Event number to trigger when the rising threshold exceeds its limit. This value is identical to the alarmRisingEventIndex in the alarmTable of the RMON MIB. If there is no corresponding rising-threshold event, the value should be zero. falling-threshold value: Value at which the falling-threshold alarm is triggered or reset. For the rmon alarm command, this is a 32-bits value, for rmon hc-alarm command this is a 64bits value. event-number: Event number to trigger when the falling threshold exceeds its limit. This value is identical to the alarmFallingEventIndex in the alarmTable of the RMON MIB. If there is no corresponding falling-threshold event, the value should be zero. owner string: (Optional) Specifies an owner for the alarm, this is the alarmOwner object in the alarmTable of the RMON MIB. Default is a null-terminated string.

[no] rmon hc-alarm number variable interval {delta | absolute} rising-threshold value event-number falling-threshold value event-number [owner string]

The following example configures an RMON alarm using the rmon alarm command.

Remote Monitoring | 617

www.dell.com | support.dell.com

Figure 32-1. rmon alarm Command Example

FTOS(conf)#rmon alarm 10 1.3.6.1.2.1.2.2.1.20.1 20 delta rising-threshold 15 1 falling-threshold 0 owner nms1

Alarm Number

MIB Variable

Monitor Interval

Counter Value Limit

Triggered Event

The above example configures RMON alarm number 10. The alarm monitors the MIB variable 1.3.6.1.2.1.2.2.1.20.1 (ifEntry.ifOutErrors) once every 20 seconds until the alarm is disabled, and checks the rise or fall of the variable. The alarm is triggered when the 1.3.6.1.2.1.2.2.1.20.1 value shows a MIB counter increase of 15 or more (such as from 100000 to 100015). The alarm then triggers event number 1, which is configured with the RMON event command. Possible events include a log entry or a SNMP trap. If the 1.3.6.1.2.1.2.2.1.20.1 value changes to 0 (falling-threshold 0), the alarm is reset and can be triggered again.

Configure an RMON event

To add an event in the RMON event table, use the rmon event command in GLOBAL CONFIGURATION mode. To disable RMON on the interface, use the no form of this command:
Command Syntax
[no] rmon event number [log] [trap community] [description string] [owner string]

Command Mode CONFIGURATION

Purpose number: Assigned event number, which is identical to the eventIndex in the eventTable in the RMON MIB. The value must be an integer from 1 to 65,535, the value must be unique in the RMON Event Table. log: (Optional) Generates an RMON log entry when the event is triggered and sets the eventType in the RMON MIB to log or log-and-trap. Default is no log. trap community: (Optional) SNMP community string used for this trap. Configures the setting of the eventType in the RMON MIB for this row as either snmp-trap or log-and-trap. This value is identical to the eventCommunityValue in the eventTable in the RMON MIB. Default is public. description string: (Optional) Specifies a description of the event, which is identical to the event description in the eventTable of the RMON MIB. Default is a null-terminated string. owner string: (Optional) Owner of this event, which is identical to the eventOwner in the eventTable of the RMON MIB. Default is a null-terminated string.

The following example shows the rmon event command.

618

Remote Monitoring

Figure 32-2. rmon event Command Example

FTOS(conf)#rmon event 1 log trap eventtrap description High ifOutErrors owner nms1

The above configuration example creates RMON event number 1, with the description High ifOutErrors, and generates a log entry when the event is triggered by an alarm. The user nms1 owns the row that is created in the event table by this command. This configuration also generates an SNMP trap when the event is triggered using the SNMP community string eventtrap.

Configure RMON collection statistics

To enable RMON MIB statistics collection on an interface, use the RMON collection statistics command in interface configuration mode. To remove a specified RMON statistics collection, use the no form of this command.
Command Syntax [no] rmon collection statistics {controlEntry integer} [owner ownername] Command Mode CONFIGURATION INTERFACE (config-if) Purpose controlEntry: Specifies the RMON group of statistics using a value. integer: A value from 1 to 65,535 that identifies the RMON Statistics Table. The value must be unique in the RMON Statistic Table. owner: (Optional) Specifies the name of the owner of the RMON group of statistics. ownername: (Optional) Records the name of the owner of the RMON group of statistics. Default is a null-terminated string

The following command enables the RMON statistics collection on the interface, with an ID value of 20 and an owner of john.
Figure 32-3. rmon collection statistics Command Example
FTOS(conf-if-mgmt)#rmon collection statistics controlEntry 20 owner john

Remote Monitoring | 619

www.dell.com | support.dell.com

Configure RMON collection history

To enable the RMON MIB history group of statistics collection on an interface, use the rmon collection history command in interface configuration mode. To remove a specified RMON history group of statistics collection, use the no form of this command.
Command Syntax
[no] rmon collection history {controlEntry integer} [owner ownername] [buckets bucket-number] [interval seconds]

Command Mode CONFIGURATION INTERFACE (config-if)

Purpose controlEntry: Specifies the RMON group of statistics using a value. integer: A value from 1 to 65,535 that identifies the RMON group of statistics. The value must be a unique index in the RMON History Table. owner: (Optional) Specifies the name of the owner of the RMON group of statistics.Default is a null-terminated string. ownername: (Optional) Records the name of the owner of the RMON group of statistics. buckets: (Optional) Specifies the maximum number of buckets desired for the RMON collection history group of statistics. bucket-number: (Optional) A value associated with the number of buckets specified for the RMON collection history group of statistics. The value is limited to from 1 to 1000. Default is 50 (as defined in RFC-2819). interval: (Optional) Specifies the number of seconds in each polling cycle. seconds: (Optional) The number of seconds in each polling cycle. The value is ranged from 5 to 3,600 (Seconds). Default is 1,800 as defined in RFC-2819.

Enable an RMON MIB collection history group

The following command enables an RMON MIB collection history group of statistics with an ID number of 20 and an owner of john, both the sampling interval and the number of buckets use their respective defaults.
Figure 32-4. rmon collection history Command Example

FTOS(conf-if-mgmt)#rmon collection history controlEntry

620

Remote Monitoring

33
Rapid Spanning Tree Protocol
Rapid Spanning Tree Protocol is supported on platforms:

ces

Protocol Overview
Rapid Spanning Tree Protocol (RSTP) is a Layer 2 protocolspecified by IEEE 802.1wthat is essentially the same as Spanning-Tree Protocol (STP) but provides faster convergence and interoperability with switches configured with STP and MSTP. FTOS supports three other variations of Spanning Tree, as shown in Table 33-1.
Table 33-1. FTOS Supported Spanning Tree Protocols IEEE Specification 802.1d 802.1w 802.1s Third Party

Dell Networking Term Spanning Tree Protocol Rapid Spanning Tree Protocol Multiple Spanning Tree Protocol Per-VLAN Spanning Tree Plus

Configuring Rapid Spanning Tree

Configuring Rapid Spanning Tree is a two-step process: 1. Configure interfaces for Layer 2. See page 48. 2. Enable Rapid Spanning Tree Protocol. See page 49.

Related Configuration Tasks

Add and Remove Interfaces Modify Global Parameters Modify Interface Parameters

Rapid Spanning Tree Protocol | 621

www.dell.com | support.dell.com

Configure an EdgePort Preventing Network Disruptions with BPDU Guard Influence RSTP Root Selection Configuring Spanning Trees as Hitless SNMP Traps for Root Elections and Topology Changes Fast Hellos for Link State Detection Flush MAC Addresses after a Topology Change

Important Points to Remember

RSTP is disabled by default. FTOS supports only one Rapid Spanning Tree (RST) instance. All interfaces in VLANs and all enabled interfaces in Layer 2 mode are automatically added to the RST topology. Avoid using the range command to add a large group of ports to a large group of VLANs; adding a group of ports to a range of VLANs sends multiple messages to the RSTP task. When using the range command, Dell Networking recommends limiting the range to 5 ports and 40 VLANs.

Configure Interfaces for Layer 2 Mode

All interfaces on all bridges that will participate in Rapid Spanning Tree must be in Layer 2 and enabled.
Figure 33-1. Configuring Interfaces for Layer 2 Mode

R1
R1(conf)# int range gi 1/1 - 4 R1(conf-if-gi-1/1-4)# switchport R1(conf-if-gi-1/1-4)# no shutdown R1(conf-if-gi-1/1-4)#show config ! interface GigabitEthernet 1/1 no ip address switchport no shutdown ! interface GigabitEthernet 1/2 no ip address switchport no shutdown ! interface GigabitEthernet 1/3 no ip address switchport no shutdown ! interface GigabitEthernet 1/4 no ip address switchport no shutdown

R2
1/3 1/4 2/1 2/2 2/3 2/4

1/1

1/2

3/1

3/2 3/3 3/4

R3

622

Rapid Spanning Tree Protocol

To configure the interfaces for Layer 2 and then enable them:

Step 1 2 3 Task If the interface has been assigned an IP address, remove it. Place the interface in Layer 2 mode. Enable the interface. Command Syntax
no ip address

Command Mode INTERFACE INTERFACE INTERFACE

switchport no shutdown

Verify that an interface is in Layer 2 mode and enabled using the show config command from INTERFACE mode.
Figure 33-2. Verifying Layer 2 Configuration
FTOS(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 no ip address switchport Indicates no shutdown FTOS(conf-if-gi-1/1)#

that the interface is in Layer 2 mode

Enable Rapid Spanning Tree Protocol Globally

Rapid Spanning Tree Protocol must be enabled globally on all participating bridges; it is not enabled by default. To enable Rapid Spanning Tree globally for all Layer 2 interfaces:
Step 1 2 Task Enter the PROTOCOL SPANNING TREE RSTP mode. Enable Rapid Spanning Tree. Command Syntax
protocol spanning-tree rstp

Command Mode CONFIGURATIO N PROTOCOL SPANNING TREE RSTP

no disable

Note: To disable RSTP globally for all Layer 2 interfaces, enter the disable command from PROTOCOL SPANNING TREE RSTP mode.

Verify that Rapid Spanning Tree is enabled using the show config command from PROTOCOL SPANNING TREE RSTP mode.

Rapid Spanning Tree Protocol | 623

www.dell.com | support.dell.com

Figure 33-3. Verifying RSTP is Enabled

FTOS(conf-rstp)#show config ! protocol spanning-tree rstp no disable FTOS(conf-rstp)#

Indicates that Rapid Spanning Tree is enabled

When you enable Rapid Spanning Tree, all physical and port-channel interfaces that are enabled and in Layer 2 mode are automatically part of the RST topology. Only one path from any bridge to any other bridge is enabled. Bridges block a redundant path by disabling one of the link ports.

Figure 33-4. Rapid Spanning Tree Enabled Globally

root R1
1/3 1/4 1/1 1/2
Forwarding

R2
2/1 2/2 2/3 2/4
Blocking

3/1

3/2 3/3 3/4

R3
Port 684 (GigabitEthernet 4/43) is alternate Discarding Port path cost 20000, Port priority 128, Port Identifier 128.684 Designated root has priority 32768, address 0001.e801.cbb4 Designated bridge has priority 32768, address 0001.e801.cbb4 Designated port id is 128.684, designated path cost 20000 Number of transitions to forwarding state 0 BPDU : sent 3, received 219 The port is not in the Edge port mode

View the interfaces participating in Rapid Spanning Tree using the show spanning-tree rstp command from EXEC privilege mode. If a physical interface is part of a port channel, only the port channel is listed in the command output.

624

Rapid Spanning Tree Protocol

Figure 33-5. show spanning-tree rstp Command Example

FTOS#show spanning-tree rstp Root Identifier has priority 32768, Address 0001.e801.cbb4 Root Bridge hello time 2, max age 20, forward delay 15, max hops 0 Bridge Identifier has priority 32768, Address 0001.e801.cbb4 Configured hello time 2, max age 20, forward delay 15, max hops 0 We are the root Current root has priority 32768, Address 0001.e801.cbb4 Number of topology changes 4, last change occurred 00:02:17 ago on Gi 1/26 Port 377 (GigabitEthernet 2/1) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.377 Designated root has priority 32768, address 0001.e801.cbb4 Designated bridge has priority 32768, address 0001.e801.cbb4 Designated port id is 128.377, designated path cost 0 Number of transitions to forwarding state 1 BPDU : sent 121, received 9 The port is not in the Edge port mode Port 378 (GigabitEthernet 2/2) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.378 Designated root has priority 32768, address 0001.e801.cbb4 Designated bridge has priority 32768, address 0001.e801.cbb4 Designated port id is 128.378, designated path cost 0 Number of transitions to forwarding state 1 BPDU : sent 121, received 2 The port is not in the Edge port mode Port 379 (GigabitEthernet 2/3) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.379 Designated root has priority 32768, address 0001.e801.cbb4 Designated bridge has priority 32768, address 0001.e801.cbb4 Designated port id is 128.379, designated path cost 0 Number of transitions to forwarding state 1 BPDU : sent 121, received 5 The port is not in the Edge port mode Port 380 (GigabitEthernet 2/4) is designated Forwarding Port path cost 20000, Port priority 128, Port Identifier 128.380 Designated root has priority 32768, address 0001.e801.cbb4 Designated bridge has priority 32768, address 0001.e801.cbb4 Designated port id is 128.380, designated path cost 0

Number of transitions to forwarding state 1 BPDU : sent 147, received 3 The port is not in the Edge port mode

Confirm that a port is participating in Rapid Spanning Tree using the show spanning-tree rstp brief command from EXEC privilege mode.

Rapid Spanning Tree Protocol | 625

www.dell.com | support.dell.com

Figure 33-6. show spanning-tree rstp brief Command Example

R3#show spanning-tree rstp brief Executing IEEE compatible Spanning Tree Protocol Root ID Priority 32768, Address 0001.e801.cbb4 Root Bridge hello time 2, max age 20, forward delay 15 Bridge ID Priority 32768, Address 0001.e80f.1dad Configured hello time 2, max age 20, forward delay 15 Interface Designated Name PortID Prio Cost Sts Cost Bridge ID PortID ---------- -------- ---- ------- --- ------- -------------------- -------Gi 3/1 128.681 128 20000 BLK 20000 32768 0001.e80b.88bd 128.469 Gi 3/2 128.682 128 20000 BLK 20000 32768 0001.e80b.88bd 128.470 Gi 3/3 128.683 128 20000 FWD 20000 32768 0001.e801.cbb4 128.379 Gi 3/4 128.684 128 20000 BLK 20000 32768 0001.e801.cbb4 128.380 Interface Name Role PortID Prio Cost Sts Cost Link-type Edge ---------- ------ -------- ---- ------- --- ------- --------- ---Gi 3/1 Altr 128.681 128 20000 BLK 20000 P2P No Gi 3/2 Altr 128.682 128 20000 BLK 20000 P2P No Gi 3/3 Root 128.683 128 20000 FWD 20000 P2P No Gi 3/4 Altr 128.684 128 20000 BLK 20000 P2P No R3#

Add and Remove Interfaces

To add an interface to the Rapid Spanning Tree topology, configure it for Layer 2 and it is automatically added. If you previously disabled RSTP on the interface using the command no spanning-tree 0, re-enable it using the command spanning-tree 0. Remove an interface from the Rapid Spanning Tree topology using the command no spanning-tree 0. See also Removing an Interface from the Spanning Tree Group for BPDU Filtering behavior.

Modify Global Parameters

You can modify Rapid Spanning Tree parameters. The root bridge sets the values for forward-delay, hello-time, and max-age and overwrites the values set on other bridges participating in the Rapid Spanning Tree group. Forward-delay is the amount of time an interface waits in the Listening State and the Learning State before it transitions to the Forwarding State. Hello-time is the time interval in which the bridge sends RSTP Bridge Protocol Data Units (BPDUs). Max-age is the length of time the bridge maintains configuration information before it refreshes that information by recomputing the RST topology.

Note: Dell Networking recommends that only experienced network administrators change the Rapid Spanning Tree group parameters. Poorly planned modification of the RSTG parameters can negatively impact network performance.

626

Rapid Spanning Tree Protocol

Table 33-2 displays the default values for RSTP.

Table 33-2. RSTP Parameter Forward Delay Hello Time Max Age Port Cost 100-Mb/s Ethernet interfaces 1-Gigabit Ethernet interfaces 10-Gigabit Ethernet interfaces Port Channel with 100 Mb/s Ethernet interfaces Port Channel with 1-Gigabit Ethernet interfaces Port Channel with 10-Gigabit Ethernet interfaces Port Priority RSTP Default Values

Default Value 15 seconds 2 seconds 20 seconds 200000 20000 2000 180000 18000 1800 128

To change these parameters, use the following commands, on the root bridge:
Task Change the forward-delay parameter. Range: 4 to 30 Default: 15 seconds Change the hello-time parameter. Note: With large configurations (especially those with more ports) Dell Networking recommends that you increase the hello-time. Range: 1 to 10 Default: 2 seconds Change the max-age parameter. Range: 6 to 40 Default: 20 seconds Command Syntax
forward-delay seconds

Command Mode PROTOCOL SPANNING TREE RSTP PROTOCOL SPANNING TREE RSTP

hello-time seconds

max-age seconds

PROTOCOL SPANNING TREE RSTP

View the current values for global parameters using the show spanning-tree rstp command from EXEC privilege mode. See Figure 33-5.

Modify Interface Parameters

On interfaces in Layer 2 mode, you can set the port cost and port priority values. Port cost is a value that is based on the interface type. The default values are listed in Table 33-2. The greater the port cost, the less likely the port will be selected to be a forwarding port.

Rapid Spanning Tree Protocol | 627

www.dell.com | support.dell.com

Port priority influences the likelihood that a port will be selected to be a forwarding port in case that several ports have the same port cost.

To change the port cost or priority of an interface, use the following commands:
Task Change the port cost of an interface. Range: 0 to 65535 Default: see Table 33-2. Change the port priority of an interface. Range: 0 to 15 Default: 128 Command Syntax
spanning-tree rstp cost cost

Command Mode INTERFACE

spanning-tree rstp priority priority-value

INTERFACE

View the current values for interface parameters using the show spanning-tree rstp command from EXEC privilege mode. See Figure 33-5.

Configure an EdgePort
The EdgePort feature enables interfaces to begin forwarding traffic approximately 30 seconds sooner. In this mode an interface forwards frames by default until it receives a BPDU that indicates that it should behave otherwise; it does not go through the Learning and Listening states. The bpduguard shutdown-on-violation option causes the interface hardware to be shutdown when it receives a BPDU. When only bpduguard is implemented, although the interface is placed in an Error Disabled state when receiving the BPDU, the physical interface remains up and spanning-tree will drop packets in the hardware after a BPDU violation. BPDUs are dropped in the software after receiving the BPDU violation. This feature is the same as PortFast mode in Spanning Tree. Caution: Configure EdgePort only on links connecting to an end station. EdgePort can cause loops if it is enabled on an interface connected to a network. To enable EdgePort on an interface, use the following command:
Task Enable EdgePort on an interface. Command Syntax
spanning-tree rstp edge-port [bpduguard | shutdown-on-violation]

Command Mode INTERFACE

Verify that EdgePort is enabled on a port using the show spanning-tree rstp command from the EXEC privilege mode or the show config command from INTERFACE mode; Dell Networking recommends using the show config command, as shown in Figure 33-7.

628

Rapid Spanning Tree Protocol

FTOS Behavior: Regarding bpduguard shutdown-on-violation behavior: 1If the interface to be shutdown is a port channel then all the member ports are disabled in the hardware. 2When a physical port is added to a port channel already in error disable state, the new member port will also be disabled in the hardware. 3When a physical port is removed from a port channel in error disable state, the error disabled state is cleared on this physical port (the physical port will be enabled in the hardware). 4The reset linecard command does not clear the error disabled state of the port or the hardware disabled state. The interface continues to be disables in the hardware. The error disabled state can be cleared with any of the following methods: Perform an shutdown command on the interface. Disable the shutdown-on-violation command on the interface ( no spanning-tree stp-id portfast [bpduguard | [shutdown-on-violation]] ). Disable spanning tree on the interface (no spanning-tree in INTERFACE mode). Disabling global spanning tree (no spanning-tree in CONFIGURATION mode). Figure 33-7. EdgePort Enabled on Interface
FTOS(conf-if-gi-2/0)#show config ! interface GigabitEthernet 2/0 no ip address switchport spanning-tree rstp edge-port shutdown FTOS(conf-if-gi-2/0)#

Indicates the interface is in EdgePort mode

Influence RSTP Root Selection

The Rapid Spanning Tree Protocol determines the root bridge, but you can assign one bridge a lower priority to increase the likelihood that it will be selected as the root bridge. To change the bridge priority, use the following command:
Task Assign a number as the bridge priority or designate it as the primary or secondary root. priority-value range: 0 to 65535. The lower the number assigned, the more likely this bridge will become the root bridge. The default is 32768. Entries must be multiples of 4096. Command Syntax
bridge-priority priority-value

Command Mode PROTOCOL SPANNING TREE RSTP

A console message appears when a new root bridge has been assigned. Figure 33-8 shows the console message after the bridge-priority command is used to make R2 the root bridge.

Rapid Spanning Tree Protocol | 629

www.dell.com | support.dell.com

Figure 33-8. bridge-priority Command Example

FTOS(conf-rstp)#bridge-priority 4096 04:27:59: %RPM0-P:RP2 %SPANMGR-5-STP_ROOT_CHANGE: RSTP root changed. My Bridge ID: 4096:0001.e80b.88bd Old Root: 32768:0001.e801.cbb4 New Root: 4096:0001.e80b.88bd

Old root bridge ID

New root bridge ID

SNMP Traps for Root Elections and Topology Changes

Enable SNMP traps for RSTP, MSTP, and PVST+ collectively using the command snmp-server enable traps xstp.

Fast Hellos for Link State Detection

Fast Hellos for Link State Detection is available only on platform:

Use RSTP Fast Hellos to achieve sub-second link-down detection so that convergence is triggered faster. The standard RSTP link-state detection mechanism does not offer the same low link-state detection speed. RSTP Fast Hellos decrease the hello interval to the order of milliseconds and all timers derived from the hello timer are adjusted accordingly. This feature does not inter-operate with other vendors, and is available only for RSTP.
Task Configure a hello time on the order of milliseconds. Command Syntax
hello-time milli-second interval

Command Mode PROTOCOL RSTP

Range: 50 - 950 milliseconds FTOS(conf-rstp)#do show spanning-tree rstp brief Executing IEEE compatible Spanning Tree Protocol Root ID Priority 0, Address 0001.e811.2233 Root Bridge hello time 50 ms, max age 20, forward delay 15 Bridge ID Priority 0, Address 0001.e811.2233 We are the root Configured hello time 50 ms, max age 20, forward delay 15 Note: The hello time is encoded in BPDUs in increments of 1/256ths of a second. The standard minimum hello time in seconds is 1 second, which is encoded as 256. Millisecond hello times are encoded using values less than 256; the millisecond hello time equals (x/1000)*256. Note: When millisecond hellos are configured, the default hello interval of 2 seconds is still used for edge ports; the millisecond hello interval is not used.

630

Rapid Spanning Tree Protocol

34
Security
Security features are supported on platforms

ces c, e or s icons (as shown below).

This chapter discusses several ways to provide access security to the Dell Networking system. Platform-specific features are identified by the

AAA Accounting AAA Authentication AAA Authorization RADIUS TACACS+ Protection from TCP Tiny and Overlapping Fragment Attacks SCP and SSH Telnet VTY Line and Access-Class Configuration Trace Lists

For details on all commands discussed in this chapter, see the Security Commands chapter in the FTOS Command Reference.

AAA Accounting
AAA Accounting is part of the AAA security model (Accounting, Authentication, and Authorization), which includes services for authentication, authorization, and accounting. For details on commands related to AAA security, refer to the Security chapter in the FTOS Command Reference. AAA Accounting enables tracking of services that users are accessing and the amount of network resources being consumed by those services. When AAA Accounting is enabled, the network server reports user activity to the security server in the form of accounting records. Each accounting record is comprised of accounting AV pairs and is stored on the access control server. As with authentication and authorization, you must configure AAA Accounting by defining a named list of accounting methods, and then apply that list to various interfaces.

Security | 631

www.dell.com | support.dell.com

Configuration Task List for AAA Accounting

The following sections present the AAA Accounting configuration tasks: Enable AAA Accounting (mandatory) Suppress AAA Accounting for null username sessions (optional) Configure Accounting of EXEC and privilege-level command usage (optional) Configure AAA Accounting for terminal lines (optional) Monitor AAA Accounting (optional)

Enable AAA Accounting

The aaa accounting command enables you to create a record for any or all of the accounting functions monitored. To enable AAA accounting, perform the following task in CONFIGURATION mode:
Command Syntax
aaa accounting {system | exec | command level} {default | name} {start-stop | wait-start | stop-only} {tacacs+}

Command Mode CONFIGURATION

Purpose Enable AAA Accounting and create a record for monitoring the accounting function. The variables are: systemsends accounting information of any other AAA configuration execsends accounting information when a user has logged in to the EXEC mode command levelsends accounting of commands executed at the specified privilege level default | nameEnter the name of a list of accounting methods. start-stopUse for more accounting information, to send a start-accounting notice at the beginning of the requested event and a stop-accounting notice at the end. wait-startensures that the TACACS+ security server acknowledges the start notice before granting the user's process request stop-onlyUse for minimal accounting; instructs the TACACS+ server to send a stop record accounting notice at the end of the requested user process. tacacs+ Designate the security service. Currently, FTOS supports only TACACS+

632

Security

Suppress AAA Accounting for null username sessions

When AAA Accounting is activated, the FTOS software issues accounting records for all users on the system, including users whose username string, because of protocol translation, is NULL. An example of this is a user who comes in on a line where the AAA Authentication login method-list none command is applied. To prevent accounting records from being generated for sessions that do not have usernames associated with them, perform the following task in CONFIGURATION mode:
Command Syntax
aaa accounting suppress null-username

Command Mode CONFIGURATION

Purpose Prevent accounting records from being generated for users whose username string is NULL

Configure Accounting of EXEC and privilege-level command usage

The network access server monitors the accounting functions defined in the TACACS+ attribute/value (AV) pairs. In the following sample configuration, AAA accounting is set to track all usage of EXEC commands and commands on privilege level 15.
FTOS(conf)#aaa accounting exec default start-stop tacacs+ FTOS(conf)#aaa accounting command 15 default start-stop tacacs+

System accounting can use only the default method list:

aaa accounting system default start-stop tacacs+

Configure AAA Accounting for terminal lines

Use the following commands to enable accounting with a named method list for a specific terminal line (where com15 and execAcct are the method list names):
FTOS(config-line-vty)# accounting commands 15 com15 FTOS(config-line-vty)# accounting exec execAcct

Monitor AAA Accounting

FTOS does not support periodic interim accounting, because the periodic command can cause heavy congestion when many users are logged in to the network.

Security | 633

www.dell.com | support.dell.com

No specific show command exists for TACACS+ accounting. To obtain accounting records displaying information about users currently logged in, perform the following task in Privileged EXEC mode:
Command Syntax
show accounting

Command Mode CONFIGURATION

Purpose Step through all active sessions and print all the accounting records for the actively accounted functions.

Figure 34-1. show accounting Command Example for AAA Accounting

FTOS#show accounting Active accounted actions on tty2, User admin Priv 1 Task ID 1, EXEC Accounting record, 00:00:39 Elapsed, service=shell Active accounted actions on tty3, User admin Priv 1 Task ID 2, EXEC Accounting record, 00:00:26 Elapsed, service=shell

AAA Authentication
FTOS supports a distributed client/server system implemented through Authentication, Authorization, and Accounting (AAA) to help secure networks against unauthorized access. In the Dell Networking implementation, the Dell Networking system acts as a RADIUS or TACACS+ client and sends authentication requests to a central RADIUS or TACACS+ server that contains all user authentication and network service access information. Dell Networking uses local usernames/passwords (stored on the Dell Networking system) or AAA for login authentication. With AAA, you can specify the security protocol or mechanism for different login methods and different users. In FTOS, AAA uses a list of authentication methods, called method lists, to define the types of authentication and the sequence in which they are applied. You can define a method list or use the default method list. User-defined method lists take precedence over the default method list.

Configuration Task List for AAA Authentication

The following sections provide the configuration tasks: Configure login authentication for terminal lines Configure AAA Authentication login methods Enable AAA Authentication AAA AuthenticationRADIUS

For a complete listing of all commands related to login authentication, refer to the Security chapter in the FTOS Command Reference.

634

Security

Configure login authentication for terminal lines

You can assign up to five authentication methods to a method list. FTOS evaluates the methods in the order in which you enter them in each list. If the first method list does not respond or returns an error, FTOS applies the next method list until the user either passes or fails the authentication. If the user fails a method list, FTOS does not apply the next method list.

Configure AAA Authentication login methods

To configure an authentication method and method list, use these commands in the following sequence in the CONFIGURATION mode:
Step 1 Command Syntax
aaa authentication login {method-list-name | default} method1 [... method4]

Command Mode CONFIGURATION

Purpose Define an authentication method-list (method-list-name) or specify the default. The default method-list is applied to all terminal lines. Possible methods are: enableuse the password defined by the enable secret or enable password command in the CONFIGURATION mode. lineuse the password defined by the password command in the LINE mode. localuse the username/password database defined in the local configuration. noneno authentication. radiususe the RADIUS server(s) configured with the radius-server host command. tacacs+use the TACACS+ server(s) configured with the tacacs-server host command Enter the LINE mode. Assign a method-list-name or the default list to the terminal line.

2 3

line {aux 0 | console 0 | vty number [... end-number]} login authentication {method-list-name | default}

CONFIGURATION LINE

FTOS Behavior: If you use a method list on the console port in which RADIUS or TACACS is the last authentication method, and the server is not reachable, FTOS allows access even though the username and password credentials cannot be verified. Only the console port behaves this way, and does so to ensure that users are not locked out of the system in the event that network-wide issue prevents access to these servers.

Security | 635

www.dell.com | support.dell.com

To view the configuration, use the show config command in the LINE mode or the show running-config in the EXEC Privilege mode.
Note: Dell Networking recommends that you use the none method only as a backup. This method does not authenticate users. The none and enable methods do not work with SSH.

You can create multiple method lists and assign them to different terminal lines.

Enable AAA Authentication

To enable AAA authentication, use the following command in the CONFIGURATION mode:
Command Syntax
aaa authentication enable {method-list-name | default} method1 [... method4]

Command Mode CONFIGURATION

Purpose
defaultUses the listed authentication

methods that follow this argument as the default list of methods when a user logs in. method-list-nameCharacter string used to name the list of enable authentication methods activated when a user logs in. method1 [... method4]Any of the following: RADIUS, TACACS, enable, line, none.

If the default list is not set, only the local enable is checked. This has the same effect as issuing:
aaa authentication enable default enable

AAA AuthenticationRADIUS
To enable authentication from the RADIUS server, and use TACACS as a backup, use the following commands:
Step 1 Command Syntax
aaa authentication enable default radius tacacs radius-server host x.x.x.x key some-password tacacs-server host x.x.x.x key some-password

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION

Purpose To enable RADIUS and to set up TACACS as backup. To establish host address and password. To establish host address and password.

2 3

To get enable authentication from the RADIUS server, and use TACACS as a backup, issue the following commands:
FTOS(config)# aaa authentication enable default radius tacacs Radius and TACACS server has to be properly setup for this. FTOS(config)# radius-server host x.x.x.x key <some-password> FTOS(config)# tacacs-server host x.x.x.x key <some-password>

636

Security

To use local authentication for enable secret on console, while using remote authentication on VTY lines, perform the following steps:
FTOS(config)# aaa authentication enable mymethodlist radius tacacs FTOS(config)# line vty 0 9 FTOS(config-line-vty)# enable authentication mymethodlist

Server-side configuration
TACACS+: When using TACACS+, Dell Networking sends an initial packet with service type SVC_ENABLE, and then, a second packet with just the password. The TACACS server must have an entry for username $enable$. RADIUS: When using RADIUS authentication, FTOS sends an authentication packet with the following:
Username: $enab15$ Password: <password-entered-by-user>

Therefore, the RADIUS server must have an entry for this username.

AAA Authorization
FTOS enables AAA new-model by default.You can set authorization to be either local or remote. Different combinations of authentication and authorization yield different results. By default, FTOS sets both to local.

Privilege Levels Overview

Limiting access to the system is one method of protecting the system and your network. However, at times, you might need to allow others access to the router and you can limit that access to a subset of commands. In FTOS, you can configure a privilege level for users who need limited access to the system. Every command in FTOS is assigned a privilege level of 0, 1 or 15. You can configure up to 16 privilege levels in FTOS. FTOS is pre-configured with 3 privilege levels and you can configure 13 more. The three pre-configured levels are: Privilege level 1is the default level for the EXEC mode. At this level, you can interact with the router, for example, view some show commands and Telnet and ping to test connectivity, but you cannot configure the router. This level is often called the user level. One of the commands available in Privilege level 1 is the enable command, which you can use to enter a specific privilege level. Privilege level 0contains only the end, enable and disable commands. Privilege level 15the default level for the enable command, is the highest level. In this level you can access any command in FTOS.

Security | 637

www.dell.com | support.dell.com

Privilege levels 2 through 14 are not configured and you can customize them for different users and access. After you configure other privilege levels, enter those levels by adding the level parameter after the enable command or by configuring a user name or password that corresponds to the privilege level. Refer to Configure a username and password for more information on configuring user names. By default, commands in FTOS are assigned to different privilege levels. You can access those commands only if you have access to that privilege level. For example, to reach the protocol spanning-tree command, you must log in to the router, enter the enable command for privilege level 15 (this is the default level for the command) and then enter the CONFIGURATION mode. You can configure passwords to control access to the box and assign different privilege levels to users. FTOS supports the use of passwords when you log in to the system and when you enter the enable command. If you move between privilege levels, you are prompted for a password if you move to a higher privilege level.

Configuration Task List for Privilege Levels

The following list has the configuration tasks for privilege levels and passwords. Configure a username and password (mandatory) Configure the enable password command (mandatory) Configure custom privilege levels (mandatory) Specify LINE mode password and privilege (optional) Enable and disabling privilege levels (optional)

For a complete listing of all commands related to FTOS privilege levels and passwords, refer to the Security chapter in the FTOS Command Reference.

Configure a username and password

In FTOS, you can assign a specific username to limit user access to the system.

638

Security

To configure a username and password, use the following command in the CONFIGURATION mode:
Command Syntax
username name [access-class access-list-name] [nopassword | password [encryption-type] password] [privilege level]

Command Mode CONFIGURATION

Purpose Assign a user name and password. Configure the optional and required parameters: name: Enter a text string up to 63 characters long. access-class access-list-name: Enter the name of a configured IP ACL. nopassword: Do not require the user to enter a password. encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a string. privilege level range: 0 to 15.

To view usernames, use the show users command in the EXEC Privilege mode.

Configure the enable password command

To configure FTOS, you must use the enable command to enter the EXEC Privilege level 15. After entering the command, FTOS requests that you enter a password. Privilege levels are not assigned to passwords, rather passwords are assigned to a privilege level. A password for any privilege level can always be changed. To change to a different privilege level, enter the enable command, followed by the privilege level. If you do not enter a privilege level, the default level 15 is assumed. To configure a password for a specific privilege level, use the following command in the CONFIGURATION mode:
Command Syntax
enable password [level level] [encryption-mode] password

Command Mode CONFIGURATION

Purpose Configure a password for a privilege level. Configure the optional and required parameters: level level: Specify a level 0 to 15. Level 15 includes all levels. encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a string. To change only the password for the enable command, configure only the password parameter.

To view the configuration for the enable secret command, use the show running-config command in the EXEC Privilege mode. In custom-configured privilege levels, the enable command is always available. No matter what privilege level you entered FTOS, you can enter the enable 15 command to access and configure all CLI.

Security | 639

www.dell.com | support.dell.com

Configure custom privilege levels

In addition to assigning privilege levels to the user, you can configure the privilege levels of commands so that they are visible in different privilege levels. Within FTOS, commands have certain privilege levels. With the privilege command, the default level can be changed or you can reset their privilege level back to the default. Assign the launch keyword (for example, configure) for the keywords command mode. If you assign only the first keyword to the privilege level, all commands beginning with that keyword are also assigned to the privilege level. If you enter the entire command, the software assigns the privilege level to that command only.

To assign commands and passwords to a custom privilege level, you must be in privilege level 15 and use these commands in the following sequence in the CONFIGURATION mode:
Step 1 Command Syntax
username name [access-class access-list-name] [privilege level] [nopassword | password [encryption-type] password]

Command Mode CONFIGURATION

Purpose Assign a user name and password. Configure the optional and required parameters: name: Enter a text string (up to 63 characters). access-class access-list-name: Enter the name of a configured IP ACL. privilege level range: 0 to 15. nopassword: Do not require the user to enter a password. encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a string. Configure a password for privilege level. Configure the optional and required parameters: level level: Specify a level 0 to 15. Level 15 includes all levels. encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a string up to 25 characters long. To change only the password for the enable command, configure only the password parameter.

enable password [level level] [encryption-mode] password

CONFIGURATION

640

Security

Step

Command Syntax
privilege mode {level level command | reset command}

Command Mode CONFIGURATION

Purpose Configure level and commands for a mode or reset a commands level. Configure the following required and optional parameters: mode: Enter a keyword for the modes (exec, configure, interface, line, route-map, router) level level range: 0 to 15. Levels 0, 1 and 15 are pre-configured. Levels 2 to 14 are available for custom configuration. command: A FTOS CLI keyword (up to 5 keywords allowed). reset: Return the command to its default privilege mode.

To view the configuration, use the show running-config command in the EXEC Privilege mode. Figure 34-2 is an example of a configuration to allow a user john to view only the EXEC mode commands and all snmp-server commands. Since the snmp-server commands are enable level commands and, by default, found in the CONFIGURATION mode, you must also assign the launch command for the CONFIGURATION mode, configure, to the same privilege level as the snmp-server commands.
Figure 34-2. Configuring a Custom Privilege Level
FTOS(conf)#username john privilege 8 password john FTOS(conf)#enable password level 8 notjohn FTOS(conf)#privilege exec level 8 configure FTOS(conf)#privilege config level 8 snmp-server FTOS(conf)#end FTOS#show running-config Current Configuration ... ! hostname FTOS ! enable password level 8 notjohn enable password FTOS ! username admin password 0 admin username john password 0 john privilege 8 !

The user john is assigned privilege level 8 and assigned a password. All other users are assigned a password to access privilege level 8 The command configure is assigned to privilege level 8 since it is needed to reach the CONFIGURATION mode where the snmp-server commands are located. The snmp-server commands, in the CONFIGURATION mode, are assigned to privilege level 8.

Figure 34-3 is a screen shot of the Telnet session for user john. The show privilege command output confirms that john is in privilege level 8. In the EXEC Privilege mode, john can access only the commands listed. In CONFIGURATION mode, john can access only the snmp-server commands.

Security | 641

www.dell.com | support.dell.com

Figure 34-3. User johns Login and the List of Available Commands
apollo% telnet 172.31.1.53 Trying 172.31.1.53... Connected to 172.31.1.53. Escape character is '^]'. Login: john Password: FTOS#show priv Current privilege level is 8 FTOS#? configure Configuring from terminal disable Turn off privileged commands enable Turn on privileged commands exit Exit from the EXEC no Negate a command show Show running system information terminal Set terminal line parameters traceroute Trace route to destination FTOS#confi FTOS(conf)#? end Exit from Configuration mode

Specify LINE mode password and privilege

You can specify a password authentication of all users on different terminal lines. The users privilege level will be the same as the privilege level assigned to the terminal line, unless a more specific privilege level is is assigned to the user. To specify a password for the terminal line, use the following commands, in any order, in the LINE mode:
Command Syntax
privilege level level

Command Mode LINE

Purpose Configure a custom privilege level for the terminal lines. level level range: 0 to 15. Levels 0, 1 and 15 are pre-configured. Levels 2 to 14 are available for custom configuration. Specify either a plain text or encrypted password. Configure the following optional and required parameters: encryption-type: Enter 0 for plain text or 7 for encrypted text. password: Enter a text string up to 25 characters long.

password [encryption-type] password

LINE

To view the password configured for a terminal, use the show config command in the LINE mode.

Enable and disabling privilege levels

Enter the enable or enable privilege-level command in the EXEC Privilege mode to set a users security level. If you do not enter a privilege level, FTOS sets it to 15 by default.

642

Security

To move to a lower privilege level, enter the command disable followed by the level-number you wish to set for the user in the EXEC Privilege mode. If you enter disable without a level-number, your security level is 1.

RADIUS
Remote Authentication Dial-In User Service (RADIUS) is a distributed client/server protocol. This protocol transmits authentication, authorization, and configuration information between a central RADIUS server and a RADIUS client (the Dell Networking system). The system sends user information to the RADIUS server and requests authentication of the user and password. The RADIUS server returns one of the following responses: Access-Acceptthe RADIUS server authenticates the user Access-Rejectthe RADIUS server does not authenticate the user

If an error occurs in the transmission or reception of RADIUS packets, the error can be viewed by enabling the debug radius command. Transactions between the RADIUS server and the client are encrypted (the users passwords are not sent in plain text). RADIUS uses UDP as the transport protocol between the RADIUS server host and the client. For more information on RADIUS, refer to RFC 2865, Remote Authentication Dial-in User Service.

RADIUS Authentication and Authorization

FTOS supports RADIUS for user authentication (text password) at login and can be specified as one of the login authentication methods in the aaa authentication login command. When configuring AAA authorization, you can configure to limit the attributes of services available to a user. When authorization is enabled, the network access server uses configuration information from the user profile to issue the user's session. The users access is limited based on the configuration attributes. RADIUS exec-authorization stores a user-shell profile and that is applied during user login. You may name the relevant named-lists with either a unique name or the default name. When authorization is enabled by the RADIUS server, the server returns the following information to the client: Idle time ACL configuration information Auto-command Privilege level

After gaining authorization for the first time, you may configure these attributes.
Note: RADIUS authentication/authorization is done for every login. There is no difference between first-time login and subsequent logins.

Security | 643

www.dell.com | support.dell.com

Idle Time
Every session line has its own idle-time. If the idle-time value is not changed, the default value of 30 minutes is used. RADIUS specifies idle-time allow for a user during a session before timeout. When a user logs in, the lower of the two idle-time values (configured or default) is used. The idle-time value is updated if both of the following happens: The administrator changes the idle-time of the line on which the user has logged in The idle-time is lower than the RADIUS-returned idle-time

ACL
The RADIUS server can specify an ACL. If an ACL is configured on the RADIUS server, and if that ACL is present, user may be allowed access based on that ACL. If the ACL is absent, authorization fails, and a message is logged indicating the this. RADIUS can specify an ACL for the user if both of the following are true: If an ACL is absent There is a very long delay for an entry, or a denied entry because of an ACL, and a message is logged

Note: The ACL name must be a string. Only standard ACLs in authorization (both RADIUS and TACACS) are supported. Authorization is denied in cases using Extended ACLs.

Auto-command
You can configure the system through the RADIUS server to automatically execute a command when you connect to a specific line. To do this, use the command auto-command. The auto-command is executed when the user is authenticated and before the prompt appears to the user.

Set access to privilege levels through RADIUS

Through the RADIUS server, you can use the command privilege level to configure a privilege level for the user to enter into when they connect to a session.This value is configured on the client system.

Configuration Task List for RADIUS

To authenticate users using RADIUS, at least one RADIUS server must be specified so that the system can communicate with and configure RADIUS as one of your authentication methods. The following list includes the configuration tasks for RADIUS. Define a aaa method list to be used for RADIUS (mandatory) Apply the method list to terminal lines (mandatory except when using default lists) Specify a RADIUS server host (mandatory) Set global communication parameters for all RADIUS server hosts (optional)

644

Security

Monitor RADIUS (optional)

For a complete listing of all FTOS commands related to RADIUS, refer to the Security chapter in the FTOS Command Reference.
Note: RADIUS authentication and authorization are done in a single step. Hence, authorization cannot be used independent of authentication. However, if RADIUS authorization is configured and authentication is not, then a message is logged stating this. During authorization, the next method in the list (if present) is used, or if another method is not present, an error is reported.

To view the configuration, use the show config in the LINE mode or the show running-config command in the EXEC Privilege mode.

Define a AAA method list to be used for RADIUS

To configure RADIUS to authenticate or authorize users on the system, you must create a AAA method list. Default method lists do not need to be explicitly applied to the line, so they are not mandatory. To create a method list, enter one of the following commands in CONFIGURATION mode:
Command Syntax
aaa authentication login method-list-name radius

Command Mode CONFIGURATION

Purpose Enter a text string (up to 16 characters long) as the name of the method list you wish to use with the RADIUS authentication method. Create methodlist with RADIUS and TACACS+ as authorization methods. Typical order of methods: RADIUS, TACACS+, Local, None. If authorization is denied by RADIUS, the session ends (radius should not be the last method specified).

aaa authorization exec {method-list-name | default} radius tacacs+

CONFIGURATION

Apply the method list to terminal lines

To enable RADIUS AAA login authentication for a method list, you must apply it to a terminal line. To configure a terminal line for RADIUS authentication and authorization, enter the following commands:
Command Syntax
line {aux 0 | console 0 | vty number [end-number]} login authentication {method-list-name | default} authorization exec methodlist

Command Mode CONFIGURATION LINE

Purpose Enter the LINE mode. Enable AAA login authentication for the specified RADIUS method list. This procedure is mandatory if you are not using default lists. To use the methodlist.

CONFIGURATION

Security | 645

www.dell.com | support.dell.com

Specify a RADIUS server host

When configuring a RADIUS server host, you can set different communication parameters, such as the UDP port, the key password, the number of retries, and the timeout. To specify a RADIUS server host and configure its communication parameters, use the following command in the CONFIGURATION mode:
Command Syntax
radius-server host {hostname | ip-address} [auth-port port-number] [retransmit retries] [timeout seconds] [key [encryption-type] key]

Command Mode CONFIGURATION

Purpose Enter the host name or IP address of the RADIUS server host. Configure the optional communication parameters for the specific host: auth-port port-number range: 0 to 65335. Enter a UDP port number. The default is 1812. retransmit retries range: 0 to 100. Default is 3. timeout seconds range: 0 to 1000. Default is 5 seconds. key [encryption-type] key: Enter 0 for plain text or 7 for encrypted text, and a string for the key. The key can be up to 42 characters long. This key must match the key configured on the RADIUS server host. If these optional parameters are not configured, the global default values for all RADIUS host are applied.

To specify multiple RADIUS server hosts, configure the radius-server host command multiple times. If multiple RADIUS server hosts are configured, FTOS attempts to connect with them in the order in which they were configured. When FTOS attempts to authenticate a user, the software connects with the RADIUS server hosts one at a time, until a RADIUS server host responds with an accept or reject response. If you want to change an optional parameter setting for a specific host, use the radius-server host command. To change the global communication settings to all RADIUS server hosts, refer to Set global communication parameters for all RADIUS server hosts. To view the RADIUS configuration, use the show running-config radius command in the EXEC Privilege mode. To delete a RADIUS server host, use the no radius-server host {hostname | ip-address} command.

Set global communication parameters for all RADIUS server hosts

You can configure global communication parameters (auth-port, key, retransmit, and timeout parameters) and specific host communication parameters on the same system. However, if both global and specific host parameters are configured, the specific host parameters override the global parameters for that RADIUS server host.

646

Security

To set global communication parameters for all RADIUS server hosts, use any or all of the following commands in the CONFIGURATION mode:
Command Syntax
radius-server deadtime seconds

Command Mode CONFIGURATION

Purpose Set a time interval after which a RADIUS host server is declared dead. seconds range: 0 to 2147483647. Default: 0 seconds Configure a key for all RADIUS communications between the system and RADIUS server hosts. encryption-type: Enter 7 to encrypt the password. Enter 0 to keep the password as plain text. key: Enter a string. The key can be up to 42 characters long. You cannot use spaces in the key. Configure the number of times FTOS retransmits RADIUS requests. retries range: 0 to 100. Default is 3 retries. Configure the time interval the system waits for a RADIUS server host response. seconds range: 0 to 1000. Default is 5 seconds.

radius-server key [encryption-type] key

CONFIGURATION

radius-server retransmit retries

CONFIGURATION

radius-server timeout seconds

CONFIGURATION

To view the configuration of RADIUS communication parameters, use the show running-config command in the EXEC Privilege mode.

Monitor RADIUS
To view information on RADIUS transactions, use the following command in the EXEC Privilege mode:
Command Syntax
debug radius

Command Mode EXEC Privilege

Purpose View RADIUS transactions to troubleshoot problems.

TACACS+
FTOS supports Terminal Access Controller Access Control System (TACACS+ client, including support for login authentication.

Configuration Task List for TACACS+

The following list includes the configuration task for TACACS+ functions:

Security | 647

www.dell.com | support.dell.com

Choose TACACS+ as the Authentication Method Monitor TACACS+ TACACS+ Remote Authentication and Authorization TACACS+ Remote Authentication and Authorization Specify a TACACS+ server host Choose TACACS+ as the Authentication Method

For a complete listing of all commands related to TACACS+, refer to the Security chapter in the FTOS Command Reference.

Choose TACACS+ as the Authentication Method

One of the login authentication methods available is TACACS+ and the users name and password are sent for authentication to the TACACS hosts specified.To use TACACS+ to authenticate users, you must specify at least one TACACS+ server for the system to communicate with and configure TACACS+ as one of your authentication methods. To select TACACS as the login authentication method, use these commands in the following sequence in the CONFIGURATION mode:
Step 1 Command Syntax
tacacs-server host {ip-address | host}

Command Mode CONFIGURATION

Purpose Configure a TACACS+ server host. Enter the IP address or host name of the TACACS+ server. Use this command multiple times to configure multiple TACACS+ server hosts. Enter a text string (up to 16 characters long) as the name of the method list you wish to use with the TACAS+ authentication method The tacacs+ method should not be the last method specified. Enter the LINE mode. Assign the method-list to the terminal line.

aaa authentication login {method-list-name | default} tacacs+ [...method3]

CONFIGURATION

3 4

line {aux 0 | console 0 | vty number [end-number]} login authentication {method-list-name | default}

CONFIGURATION LINE

To view the configuration, use the show config in the LINE mode or the show running-config tacacs+ command in the EXEC Privilege mode. If authentication fails using the primary method, FTOS employs the second method (or third method, if necessary) automatically. For example, if the TACACS+ server is reachable, but the server key is invalid, FTOS proceeds to the next authentication method. In Figure 34-4, the TACACS+ is incorrect, but the user is still authenticated by the secondary method.

648

Security

Figure 34-4. Failed Authentication

FTOS(conf)# FTOS(conf)#do show run aaa ! aaa authentication enable default tacacs+ enable aaa authentication enable LOCAL enable tacacs+ aaa authentication login default tacacs+ local aaa authentication login LOCAL local tacacs+ aaa authorization exec default tacacs+ none aaa authorization commands 1 default tacacs+ none aaa authorization commands 15 default tacacs+ none aaa accounting exec default start-stop tacacs+ aaa accounting commands 1 default start-stop tacacs+ aaa accounting commands 15 default start-stop tacacs+ FTOS(conf)# FTOS(conf)#do show run tacacs+ ! Server key purposely changed to incorrect value tacacs-server key 7 d05206c308f4d35b tacacs-server host 10.10.10.10 timeout 1 FTOS(conf)#tacacs-server key angeline FTOS(conf)#%RPM0-P:CP %SEC-5-LOGIN_SUCCESS: Login successful for user admin on vty0 (10.11.9.209) %RPM0-P:CP %SEC-3-AUTHENTICATION_ENABLE_SUCCESS: Enable password authentication success on vty0 ( 10.11.9.209 ) %RPM0-P:CP %SEC-5-LOGOUT: Exec session is terminated for user admin on line vty0 (10.11.9.209) FTOS(conf)#username angeline password angeline FTOS(conf)#%RPM0-P:CP %SEC-5-LOGIN_SUCCESS: Login successful for user angeline on vty0 (10.11.9.209) %RPM0-P:CP %SEC-3-AUTHENTICATION_ENABLE_SUCCESS: Enable password authentication User authenticated using secondary method success on vty0 ( 10.11.9.209 )

Monitor TACACS+
To view information on TACACS+ transactions, use the following command in the EXEC Privilege mode:
Command Syntax
debug tacacs+

Command Mode EXEC Privilege

Purpose View TACACS+ transactions to troubleshoot problems.

TACACS+ Remote Authentication and Authorization

FTOS takes the access class from the TACACS+ server. Access class is the class of service that restricts Telnet access and packet sizes. If you have configured remote authorization, then FTOS ignores the access class you have configured for the VTY line. FTOS instead gets this access class information from the TACACS+ server. FTOS needs to know the username and password of the incoming user before it can fetch the access class from the server. A user, therefore, will at least see the login prompt. If the access class denies the connection, FTOS closes the Telnet session immediately.

Security | 649

www.dell.com | support.dell.com

Figure 34-5 demonstrates how to configure the access-class from a TACACS+ server. This causes the configured access-class on the VTY line to be ignored. If you have configured a deny10 ACL on the TACACS+ server, FTOS downloads it and applies it. If the user is found to be coming from the 10.0.0.0 subnet, FTOS also immediately closes the Telnet connection. Note, that no matter where the user is coming from, they see the login prompt.
Figure 34-5. Specify a TACACS+ server host

FTOS# FTOS(conf)# FTOS(conf)#ip access-list standard deny10 FTOS(conf-ext-nacl)#permit 10.0.0.0/8 FTOS(conf-ext-nacl)#deny any FTOS(conf)# FTOS(conf)#aaa authentication login tacacsmethod tacacs+ FTOS(conf)#aaa authentication exec tacacsauthorization tacacs+ FTOS(conf)#tacacs-server host 25.1.1.2 key FTOS FTOS(conf)# FTOS(conf)#line vty 0 9 FTOS(config-line-vty)#login authentication tacacsmethod FTOS(config-line-vty)#authorization exec tacauthor FTOS(config-line-vty)# FTOS(config-line-vty)#access-class deny10 FTOS(config-line-vty)#end

When configuring a TACACS+ server host, you can set different communication parameters, such as the the key password. To specify a TACACS+ server host and configure its communication parameters, use the following command in the CONFIGURATION mode:
Command Syntax
tacacs-server host {hostname | ip-address} [port port-number] [timeout seconds] [key key]

Command Mode CONFIGURATION

Purpose Enter the host name or IP address of the TACACS+ server host. Configure the optional communication parameters for the specific host: port port-number range: 0 to 65335. Enter a TCP port number. The default is 49. timeout seconds range: 0 to 1000. Default is 10 seconds. key key: Enter a string for the key. The key can be up to 42 characters long. This key must match a key configured on the TACACS+ server host. This parameter should be the last parameter configured. If these optional parameters are not configured, the default global values are applied.

To specify multiple TACACS+ server hosts, configure the tacacs-server host command multiple times. If multiple TACACS+ server hosts are configured, FTOS attempts to connect with them in the order in which they were configured. To view the TACACS+ configuration, use the show running-config tacacs+ command in the EXEC Privilege mode.

650

Security

To delete a TACACS+ server host, use the no tacacs-server host {hostname | ip-address} command.
freebsd2# telnet 2200:2200:2200:2200:2200::2202 Trying 2200:2200:2200:2200:2200::2202... Connected to 2200:2200:2200:2200:2200::2202. Escape character is '^]'. Login: admin Password: FTOS# FTOS#

Command Authorization
The AAA command authorization feature configures FTOS to send each configuration command to a TACACS server for authorization before it is added to the running configuration. By default, the AAA authorization commands configure the system to check both EXEC mode and CONFIGURATION mode commands. Use the command no aaa authorization config-commands to enable only EXEC mode command checking. If rejected by the AAA server, the command is not added to the running config, and messages similar to Message 1 are displayed.
Message 1 Configuration Command Rejection
04:07:48: %RPM0-P:CP %SEC-3-SEC_AUTHORIZATION_FAIL: Authorization failure authorization failed for user (denyall) on vty0 ( 10.11.9.209 ) Command

Protection from TCP Tiny and Overlapping Fragment Attacks

Tiny and overlapping fragment attack is a class of attack where configured ACL entriesdenying TCP port-specific trafficcan be bypassed, and traffic can be sent to its destination although denied by the ACL. RFC 1858 and 3128 proposes a countermeasure to the problem. This countermeasure is configured into the line cards and enabled by default.

SCP and SSH

Secure Shell (SSH) is a protocol for secure remote login and other secure network services over an insecure network. FTOS is compatible with SSH versions 1.5 and 2, both the client and server modes. SSH sessions are encrypted and use authentication. For details on command syntax, see the Security chapter in the FTOS Command Line Interface Reference.

Security | 651

www.dell.com | support.dell.com

SCP is a remote file copy program that works with SSH and is supported by FTOS.
Note: The Windows-based WinSCP client software is not supported for secure copying between a PC and an FTOS-based system. Unix-based SCP client software is supported.

To use the SSH client, use the following command in the EXEC Privilege mode:
Command Syntax
ssh {hostname} [-l username | -p port-number | -v {1 | 2}

Command Mode EXEC Privilege

Purpose Open an SSH connection specifying the hostname, username, port number, and version of the SSH client. hostname is the IP address or hostname of the remote device. Enter an IPv4 address in dotted decimal format (A.B.C.D).

To enable the SSH server for version 1 and 2, use the following command in the CONFIGURATION mode:
Command Syntax
ip ssh server {enable | port port-number}

Command Mode CONFIGURATION

Purpose

Configure the Dell Networking system as an SCP/SSH server.

To enable the SSH server for version 1 or 2 only, use the following command:
Command Syntax
ip ssh server version {1|2}

Command Mode CONFIGURATION

Purpose

Configure the Dell Networking system as an SSH server that uses only version 1 or 2.

To view the SSH configuration, use the following command in EXEC Privilege mode:
Command Syntax
show ip ssh

Command Mode EXEC Privilege

Purpose Display SSH connection information.

Figure 34-6 on page 653 shows the use of the command ip ssh server version 2 to enable SSH version 2, and the show ip ssh command to confirm the setting.

652

Security

Figure 34-6. Specifying an SSH version

FTOS(conf)#ip ssh server version 2 FTOS(conf)#do show ip ssh SSH server : disabled. SSH server version : v2. Password Authentication : enabled. Hostbased Authentication : disabled. RSA Authentication : disabled.

To disable SSH server functions, enter no ip ssh server enable.

Using SCP with SSH to copy a software image

To use Secure Copy (SCP) to copy a software image through an SSH connection from one switch to another, use the following procedure:
Step 1 2 3 4 Task On Chassis One, set the SSH port number (port 22 by default). On Chassis One, enable SSH. On Chassis Two, invoke SCP. On Chassis Two, in response to prompts, enter the path to the desired file and enter the port number specified in Step 1. Command Syntax
ip ssh server port number

Command Mode CONFIGURATION CONFIGURATION CONFIGURATION EXEC Privilege

ip ssh server enable copy scp: flash:

This example shows the use of SCP and SSH to copy a software image from one switch running SSH Server on UDP port 99 to the local switch:
Figure 34-7. Using SCP to copy from an SSH Server on another Switch
FTOS#copy scp: flash: Address or name of remote host []: 10.10.10.1 Port number of the server [22]: 99 Source file name []: test.cfg User name to login remote host: admin Password to login remote host:

Other SSH-related commands include:

crypto key generate:

Generate keys for the SSH server. debug ip ssh: Enables collecting SSH debug information. ip scp topdir: Identify a location for files used in secure copy transfer. ip ssh authentication-retries: Configure the maximum number of attempts that should be used to authenticate a user.

Security | 653

www.dell.com | support.dell.com

ip ssh connection-rate-limit: Configure

the maximum number of incoming SSH connections per

minute.
ip ssh hostbased-authentication enable: Enable

hostbased-authentication for the SSHv2 server. ip ssh key-size: Configure the size of the server-generated RSA SSHv1 key. ip ssh password-authentication enable: Enable password authentication for the SSH server. ip ssh pub-key-file: Specify the file to be used for host-based authentication. ip ssh rhostsfile: Specify the rhost file to be used for host-based authorization. ip ssh rsa-authentication enable: Enable RSA authentication for the SSHv2 server. ip ssh rsa-authentication: Add keys for the RSA authentication. show crypto: Display the public part of the SSH host-keys. show ip ssh client-pub-keys: Display the client public keys used in host-based authentication. show ip ssh rsa-authentication: Display the authorized-keys for the RSA authentication. ssh-peer-rpm: Open an SSH connection to the peer RPM.

Secure Shell Authentication

Secure Shell (SSH) is disabled by default. Enable it using the command ip ssh server enable. SSH supports three methods of authentication: SSH Authentication by Password RSA Authentication of SSH Host-based SSH Authentication

Important Points to Remember for SSH Authentication

If more than one method is enabled, the order in which the methods are preferred is based on the ssh_config file on the Unix machine. When all the three authentication methods are enabled, password authentication is the backup method when the RSA method fails. The files known_hosts and known_hosts2 are generated when a user tries to SSH using version 1 or version 2, respectively.

SSH Authentication by Password

Authenticate an SSH client by prompting for a password when attempting to connect to the Dell Networking system. This is the simplest methods of authentication and uses SSH version 1. Enable SSH password authentication using the command ip ssh password-authentication enable from CONFIGURATION mode. View your SSH configuration using the command show ip ssh from EXEC Privilege mode.

654

Security

Figure 34-8. Enabling SSH Password Authentication

FTOS(conf)#ip ssh server enable % Please wait while SSH Daemon initializes ... done. FTOS(conf)#ip ssh password-authentication enable FTOS#sh ip ssh SSH server : enabled. Password Authentication : enabled. Hostbased Authentication : disabled. RSA Authentication : disabled.

RSA Authentication of SSH

The following procedure authenticates an SSH client based on an RSA key using RSA authentication. This method uses SSH version 2:
Step 1 Task Command Syntax Command Mode

On the SSH client (Unix machine), generate an RSA key, as shown in Figure 34-9. Figure 34-9. Generating RSA Keys
admin@Unix_client#ssh-keygen -t rsa Generating public/private rsa key pair. Enter file in which to save the key (/home/admin/.ssh/id_rsa): /home/admin/.ssh/id_rsa already exists. Overwrite (y/n)? y Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in /home/admin/.ssh/id_rsa. Your public key has been saved in /home/admin/.ssh/id_rsa.pub.

2 3 4 5

Copy the public key id_rsa.pub to the Dell Networking system. Disable password authentication if enabled. Enable RSA authentication. Bind the public keys to RSA authentication.
no ip ssh password-authentication enable

CONFIGURATION EXEC Privilege EXEC Privilege

ip ssh rsa-authentication enable

ip ssh rsa-authentication my-authorized-keys flash:// public_key

Host-based SSH Authentication

Authenticate a particular host. This method uses SSH version 2.

Security | 655

www.dell.com | support.dell.com

To configure host-based authentication:

Step 1 2 Task Command Syntax Command Mode

Configure RSA Authentication. See RSA Authentication of SSH, above. Create shosts by copying the public RSA key to the to the file shosts in the diretory .ssh, and write the IP address of the host to the file. Figure 34-10. Creating shosts
cp /etc/ssh/ssh_host_rsa_key.pub /.ssh/shosts

admin@Unix_client# cd /etc/ssh admin@Unix_client# ls moduli sshd_config ssh_host_dsa_key.pub ssh_host_key.pub ssh_host_rsa_key.pub ssh_config ssh_host_dsa_key ssh_host_key ssh_host_rsa_key admin@Unix_client# cat ssh_host_rsa_key.pub ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAIEA8K7jLZRVfjgHJzUOmXxuIbZx/ AyWhVgJDQh39k8v3e8eQvLnHBIsqIL8jVy1QHhUeb7GaDlJVEDAMz30myqQbJgXBBRTWgBpLWwL/ doyUXFufjiL9YmoVTkbKcFmxJEMkE3JyHanEi7hg34LChjk9hL1by8cYZP2kYS2lnSyQWk= admin@Unix_client# ls id_rsa id_rsa.pub shosts admin@Unix_client# cat shosts 10.16.127.201, ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAIEA8K7jLZRVfjgHJzUOmXxuIbZx/AyW hVgJDQh39k8v3e8eQvLnHBIsqIL8jVy1QHhUeb7GaDlJVEDAMz30myqQbJgXBBRTWgBpLWwL/ doyUXFufjiL9YmoVTkbKcFmxJEMkE3JyHanEi7hg34LChjk9hL1by8cYZP2kYS2lnSyQWk=

Create a list of IP addresses and usernames that are permitted to SSH in a file called rhosts, as shown in Figure 34-11. Figure 34-11. Creating rhosts

admin@Unix_client# ls id_rsa id_rsa.pub rhosts shosts admin@Unix_client# cat rhosts 10.16.127.201 admin

4 5 6 7

Copy the file shosts and rhosts to the Dell Networking system. Disable password authentication and RSA authentication, if configured Enable host-based authentication. Bind shosts and rhosts to host-based authentication.
no ip ssh password-authentication no ip ssh rsa-authentication

CONFIGURATION EXEC Privilege

ip ssh hostbased-authentication enable ip ssh pub-key-file flash://filename ip ssh rhostsfile flash://filename

CONFIGURATION CONFIGURATION

Client-based SSH Authentication

SSH from the chassis to the SSH client using using the command ssh ip_address. This method uses SSH version 1 or version 2. If the SSH port is a non-default value, use the command ip ssh server port number, to change the default port number. You may only change the port number when SSH is disabled. When must then still use the -p option with the command ssh.

656

Security

Figure 34-12.

Client-based SSH Authentication

FTOS#ssh 10.16.127.201 ? -l User name option -p SSH server port option (default 22) -v SSH protocol version

Troubleshooting SSH
You may not bind id_rsa.pub to RSA authentication while logged in via the console. In this case, Message 2 appears.

Message 2 RSA Authentication Error

%Error: No username set for this term.

Host-based authentication must be enabled on the server (Dell Networking system) and the client (Unix machine). Message 3 appears if you attempt to log in via SSH and host-based is disabled on the client. In this case, verify that host-based authentication is set to Yes in the file ssh_config (root permission is required to edit this file).

Message 3 Host-based Authentication Error

permission denied (host based)

If the IP address in the RSA key does not match the IP address from which you attempt to log in, Message 4 appears. In this case, verify that the name and IP address of the client is contained in the file /etc/hosts.

Message 4 RSA Authentication Error

getname info 8 failed

Telnet
To use Telnet with SSH, you must first enable SSH, as described above. By default, the Telnet daemon is enabled. If you want to disable the Telnet daemon, use the following command, or disable Telnet in the startup config. Use the [no] ip telnet server enable command to enable or disable the Telnet daemon.
FTOS(conf)#ip telnet server enable FTOS(conf)#no ip telnet server enable

Security | 657

www.dell.com | support.dell.com

Trace Lists
The Trace Lists feature is supported only on the E-Series:

You can log packet activity on a port to confirm the source of traffic attacking a system. Once the Trace list is enabled on the system, you view its traffic log to confirm the source address of the attacking traffic. In FTOS, Trace lists are similar to extended IP ACLs, except that Trace lists are not applied to an interface. Instead, Trace lists are enabled for all switched traffic entering the system. The number of entries allowed per trace list is 1K. In the E-Series, you can create a trace filter based on any of the following criteria: Source IP address Destination IP address Source TCP port number Destination TCP port number Source UDP port number Destination UDP port number

For trace lists, you can match criteria on specific or ranges of TCP or UDP ports or established TCP sessions.
Note: If there are unresolved next-hops and a trace-list is enabled, there is a possibility that the traffic hitting the CPU will not be rate-limited.

When creating a trace list, the sequence of the filters is important. You have a choice of assigning sequence numbers to the filters as you enter them, or FTOS assigns numbers in the order the filters were created. For more information on sequence numbering, refer to Chapter 21, IP Access Control Lists, Prefix Lists, and Route-maps, on page 419.

Configuration Tasks for Trace Lists

The following configuration steps include mandatory and optional steps. Creating a trace list (mandatory) Apply trace lists (mandatory)

For a complete listing of all commands related to trace lists, refer to the Security chapter in the FTOS Command Reference.

Creating a trace list

Trace lists filter and log traffic based on source and destination IP addresses, IP host addresses, TCP addresses, TCP host addresses, UDP addresses, and UDP host addresses. When configuring the Trace list filters, include the count and bytes parameters so that any hits to that filter are logged.

658

Security

Since traffic passes through the filter in the order of the filters sequence, you can configure the trace list by first entering the TRACE LIST mode and then assigning a sequence number to the filter. To create a filter for packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip trace-list trace-list-name

Command Mode CONFIGURATION TRACE LIST

Purpose Enter the TRACE LIST mode by creating an trace list. Configure a drop or forward filter. Configure the following required and optional parameters: sequence-number range: 0 to, 4294967290. ip: to specify IP as the protocol to filter for. ip-protocol-number range: 0 to 255. source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported.

seq sequence-number {deny | permit} {ip | ip-protocol-number} {source mask | any | host ip-address} {destination mask | any | host ip-address} [count [byte] | log]

To create a filter for TCP packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip trace-list trace-list-name

Command Mode CONFIGURATION

Purpose Create a trace list and assign it a unique name.

Security | 659

www.dell.com | support.dell.com

Step

Command Syntax
seq sequence-number {deny | permit} tcp {source mask | any | host ip-address} [operator port [port]] {destination mask | any | host ip-address} [operator port [port]] [established] [count [byte] | log]

Command Mode TRACE LIST

Purpose Configure a trace list filter for TCP packets. source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported.

To create a filter for UDP packets with a specified sequence number, use these commands in the following sequence, starting in the CONFIGURATION mode:
Step 1 Command Syntax
ip trace-list access-list-name

Command Mode CONFIGURATION TRACE LIST

Purpose Create a trace list and assign it a unique name. Configure a trace list filter for UDP packets. source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported.

seq sequence-number {deny | permit} udp {source mask | any | host ip-address} [operator port [port]] {destination mask | any | host ip-address} [operator port [port]] [count [byte] | log]

When you create the filters with a specific sequence number, you can create the filters in any order and the filters are placed in the correct order.
Note: When assigning sequence numbers to filters, keep in mind that you might need to insert a new filter. To prevent reconfiguring multiple filters, assign sequence numbers in multiples of five or another number.

Figure 34-13 illustrates how the seq command orders the filters according to the sequence number assigned. In the example, filter 15 was configured before filter 5, but the show config command displays the filters in the correct order.

660

Security

Figure 34-13.

Trace list Using seq Command Example

FTOS(config-trace-acl)#seq 15 deny ip host 12.45.0.0 any log FTOS(config-trace-acl)#seq 5 permit tcp 121.1.3.45 0.0.255.255 any FTOS(config-trace-acl)#show conf ! ip trace-list dilling seq 5 permit tcp 121.1.0.0 0.0.255.255 any seq 15 deny ip host 12.45.0.0 any log

If you are creating a Trace list with only one or two filters, you can let FTOS assign a sequence number based on the order in which the filters are configured. FTOS assigns filters in multiples of 5. To configure a filter for a Trace list without a specified sequence number, use any or all of the following commands in the TRACE LIST mode:
Command Syntax {deny | permit} {ip | ip-protocol-number} {source mask | any | host ip-address} {destination mask | any | host ip-address} [count [byte] | log] Command Mode TRACE LIST Purpose Configure a deny or permit filter to examine IP packets. Configure the following required and optional parameters: ip: to specify IP as the protocol to filter for. ip-protocol-number range: 0 to 255. source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported.

Security | 661

www.dell.com | support.dell.com

Command Syntax {deny | permit} tcp {source mask | any | host

ip-address} [operator port [port]] {destination mask | any | host ip-address} [operator port [port]] [established] [count [byte] | log]

Command Mode TRACE LIST

Purpose Configure a deny or permit filter to examine TCP packets. Configure the following required and optional parameters: source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. precedence precedence range: 0 to 7. tos tos-value range: 0 to 15 count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported. Configure a deny or permit filter to examine UDP packets. Configure the following required and optional parameters: source: An IP address as the source IP address for the filter to match. mask: a network mask any: to match any IP source address host ip-address: to match IP addresses in a host. destination: An IP address as the source IP address for the filter to match. precedence precedence range: 0 to 7. tos tos-value range: 0 to 15 count: count packets processed by the filter. byte: count bytes processed by the filter. log: is supported.

{deny | permit} udp {source mask | any | host

ip-address} [operator port [port]] {destination mask | any | host ip-address} [operator port [port]] | log]

TRACE LIST

Figure 34-14 illustrates a Trace list in which the sequence numbers were assigned by the software. The filters were assigned sequence numbers based on the order in which they were configured (for example, the first filter was given the lowest sequence number). The show config command in the TRACE LIST mode displays the two filters with the sequence numbers 5 and 10.

662

Security

Figure 34-14.

Trace List Example

FTOS(config-trace-acl)#deny tcp host 123.55.34.0 any FTOS(config-trace-acl)#permit udp 154.44.123.34 0.0.255.255 host 34.6.0.0 FTOS(config-trace-acl)#show config ! ip trace-list nimule seq 5 deny tcp host 123.55.34.0 any seq 10 permit udp 154.44.0.0 0.0.255.255 host 34.6.0.0

ip accounting trace-list command

To view all configured Trace lists and the number of packets processed through the Trace list, use the show (Figure 110) in the EXEC Privilege mode.

Apply trace lists

After you create a Trace list, you must enable it. Without enabling the Trace list, no traffic is filtered. You can enable one Trace list. To enable a Trace list, use the following command in the CONFIGURATION mode:
Command Syntax
ip trace-group trace-list-name

Command Mode CONFIGURATION

Purpose Enable a configured Trace list to filter traffic.

To remove a Trace list, use the no ip trace-group trace-list-name command syntax. Once the Trace list is enabled, you can view its log with the show ip accounting trace-list trace-list-name [linecard number] command.
Figure 34-15. show ip accounting trace-list Command Example

FTOS#show ip accounting trace-list dilling Trace List dilling on linecard 0 seq 2 permit ip host 10.1.0.0 any count (0 packets) seq 5 deny ip any any FTOS#

Security | 663

www.dell.com | support.dell.com

VTY Line and Access-Class Configuration

Various methods are available to restrict VTY access in FTOS. These depend on which authentication scheme you use line, local, or remote:
Table 34-1. VTY Access Username VTY access-class access-class support? support? YES NO YES YES NO YES NO NO

Authentication Method Line Local TACACS+ RADIUS

Remote authorization support? NO NO YES (with FTOS 5.2.1.0 and later) YES (with FTOS 6.1.1.0 and later)

FTOS provides several ways to configure access classes for VTY lines, including: VTY Line Local Authentication and Authorization VTY Line Remote Authentication and Authorization

VTY Line Local Authentication and Authorization

FTOS retrieves the access class from the local database. To use this feature: 1. Create a username 2. Enter a password 3. Assign an access class 4. Enter a privilege level Line authentication can be assigned on a per-VTY basis; it is a simple password authentication, using an access-class as authorization. Local authentication is configured globally. You configure access classes on a per-user basis. FTOS can assign different access classes to different users by username. Until users attempt to log in, FTOS does not know if they will be assigned a VTY line. This means that incoming users always see a login prompt even if you have excluded them from the VTY line with a deny-all access class. Once users identify themselves, FTOS retrieves the access class from the local database and applies it. (FTOS also subsequently can close the connection if a user is denied access). Figure 34-16 shows how to allow or deny a Telnet connection to a user. Users will see a login prompt, even if they cannot login. No access class is configured for the VTY line. It defaults from the local database.

664

Security

Figure 34-16.

Example Access-Class Configuration Using Local Database

FTOS(conf)#user gooduser password abc privilege 10 access-class permitall FTOS(conf)#user baduser password abc privilege 10 access-class denyall FTOS(conf)# FTOS(conf)#aaa authentication login localmethod local FTOS(conf)# FTOS(conf)#line vty 0 9 FTOS(config-line-vty)#login authentication localmethod FTOS(config-line-vty)#end

Note: See also the section Chapter 6, Access Control Lists (ACL), Prefix Lists, and Route-maps.

VTY Line Remote Authentication and Authorization

FTOS retrieves the access class from the VTY line. The Dell Networking OS takes the access class from the VTY line and applies it to ALL users. FTOS does not need to know the identity of the incoming user and can immediately apply the access class. If the authentication method is radius, TACACS+, or line, and you have configured an access class for the VTY line, FTOS immediately applies it. If the access-class is deny all or deny for the incoming subnet, FTOS closes the connection without displaying the login prompt. Figure shows how to deny incoming connections from subnet 10.0.0.0 without displaying a login prompt. The example uses TACACS+ as the authentication mechanism.
Figure 34-17. Example Access Class Configuration Using TACACS+ Without Prompt

FTOS(conf)#ip access-list standard deny10 FTOS(conf-ext-nacl)#permit 10.0.0.0/8 FTOS(conf-ext-nacl)#deny any FTOS(conf)# FTOS(conf)#aaa authentication login tacacsmethod tacacs+ FTOS(conf)#tacacs-server host 256.1.1.2 key FTOS FTOS(conf)# FTOS(conf)#line vty 0 9 FTOS(config-line-vty)#login authentication tacacsmethod FTOS(config-line-vty)# FTOS(config-line-vty)#access-class deny10 FTOS(config-line-vty)#end (same applies for radius and line authentication)

VTY MAC-SA Filter Support

FTOS supports MAC access lists which permit or deny users based on their source MAC address. With this approach, you can implement a security policy based on the source MAC address. To apply a MAC ACL on a VTY line, use the same access-class command as IP ACLs (Figure 34-18). Figure 34-18 shows how to deny incoming connections from subnet 10.0.0.0 without displaying a login prompt.

Security | 665

www.dell.com | support.dell.com

Figure 34-18.

Example Access Class Configuration Using TACACS+ Without Prompt

FTOS(conf)#mac access-list standard sourcemac FTOS(config-std-mac)#permit 00:00:5e:00:01:01 FTOS(config-std-mac)#deny any FTOS(conf)# FTOS(conf)#line vty 0 9 FTOS(config-line-vty)#access-class sourcemac FTOS(config-line-vty)#end

666

Security

35
Service Provider Bridging
Service Provider Bridging is supported on platforms: This chapter contains the following major sections: VLAN Stacking VLAN Stacking Packet Drop Precedence Dynamic Mode CoS for VLAN Stacking Layer 2 Protocol Tunneling Provider Backbone Bridging

ces

VLAN Stacking
VLAN Stacking is supported on platforms:

ces

VLAN Stacking, also called Q-in-Q, is defined in IEEE 802.1adProvider Bridges, which is an amendment to IEEE 802.1QVirtual Bridged Local Area Networks. It enables service providers to use 802.1Q architecture to offer separate VLANs to customers with no coordination between customers, and minimal coordination between customers and the provider. Using only 802.1Q VLAN tagging all customers would have to use unique VLAN IDs to ensure that traffic is segregated, and customers and the service provider would have to coordinate to ensure that traffic mapped correctly across the provider network. Even under ideal conditions, customers and the provider would still share the 4094 available VLANs. Instead, 802.1ad allows service providers to add their own VLAN tag to frames traversing the provider network. The provider can then differentiate customers even if they use the same VLAN ID, and providers can map multiple customers to a single VLAN to overcome the 4094 VLAN limitation. Forwarding decisions in the provider network are based on the provider VLAN tag only, so the provider can map traffic through the core independently; the customer and provider need only coordinate at the provider edge. In at the access point of a VLAN-stacking network, service providers add a VLAN tag, the S-Tag, to each frame before the 802.1Q tag. From this point, the frame is double-tagged. The service provider uses the S-Tag, to forward the frame traffic across its network. At the egress edge, the provider removes the S-Tag, so that the customer receives the frame in its original condition (Figure 35-1).

Service Provider Bridging | 667

www.dell.com | support.dell.com

Figure 35-1. VLAN Stacking in a Service Provider Network

TPID (0x9100)

PCP

DEI

VID (VLAN 300)

TPID (0x8100)

PCP

CFI (0)

VID (VLAN Red)

tagged 100
1 00

trunk port

VL

AN

00 N1 VLA

tagged 100
IN

Important Points to Remember

Interfaces that are members of the Default VLAN and are configured as VLAN-Stack access or trunk ports do not switch untagged traffic. To switch traffic, these interfaces must be added to a non-default VLAN-Stack-enabled VLAN. Dell Networking cautions against using the same MAC address on different customer VLANs, on the same VLAN-Stack VLAN. You can ping across a trunk port only if both systems on the link are an E-Series. You cannot ping across the link if one or both of the systems is a C-Series or S-Series. This limitation becomes relevant if you enable the port as a multi-purpose port (carrying single-tagged and double-tagged traffic).

VL AN 0 10

VLAN
VLAN 300

100

VL

AN

20

access port

TE

RN

ET

S E RV I C E P R O V

ID E

Rw / V

LAN

S TA C KI N

Configure VLAN Stacking

Configuring VLAN-Stacking is a three-step process: 1. Create access and trunk ports. See page 669. 2. Assign access and trunk ports to a VLAN. See page 669. 3. Make the VLAN VLAN-stacking capable.

Related Configuration Tasks

Configure the Protocol Type Value for the Outer VLAN Tag FTOS Options for Trunk Ports Debug VLAN Stacking VLAN Stacking in Multi-vendor Networks

668

Service Provider Bridging

Create Access and Trunk Ports

An access port is a port on the service provider edge that directly connects to the customer. An access port may belong to only one service provider VLAN. A trunk port is a port on a service provider bridge that connects to another service provider bridge and is a member of multiple service provider VLANs. Physical ports and port-channels can be access or trunk ports. To create access and trunk ports:
Step 1 2 3 Task Assign the role of access port to a Layer 2 port on a provider bridge that is connected to a customer. Assign the role of trunk port to a Layer 2 port on a provider bridge that is connected to another provider bridge. Assign all access ports and trunk ports to service provider VLANs. Command Syntax
vlan-stack access

Command Mode INTERFACE INTERFACE INTERFACE VLAN

vlan-stack trunk
member

Display the VLAN-Stacking configuration for a switchport using the command show config from INTERFACE mode, as shown in Figure 35-2.
Figure 35-2. Displaying the VLAN-Stack Configuration on a Layer 2 Port
FTOS#show run interface gi 7/0 ! interface GigabitEthernet 7/0 no ip address switchport vlan-stack access no shutdown FTOS#show run interface gi 7/12 ! interface GigabitEthernet 7/12 no ip address switchport vlan-stack trunk no shutdown

Enable VLAN-Stacking for a VLAN

To enable VLAN-Stacking for a VLAN:
Task Enable VLAN-Stacking for the VLAN. Command Syntax INTERFACE VLAN Command Mode
vlan-stack compatible

Service Provider Bridging | 669

www.dell.com | support.dell.com

Display the status and members of a VLAN using the show vlan command from EXEC Privilege mode. Members of a VLAN-Stacking-enabled VLAN are marked with an M in column Q.
Figure 35-3. Display the Members of a VLAN-Stacking-enabled VLAN
FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 4 5 6 Status Active Inactive Inactive Inactive Inactive Active Q Ports U Gi 13/0-5,18

M Po1(Gi 13/14-15)

Configure the Protocol Type Value for the Outer VLAN Tag
The Tag Protocol Identifier (TPID) field of the S-Tag is user-configurable:
Task Select a value for the S-Tag TPID. Default: 9100 Command Syntax CONFIGURATION Command Mode
vlan-stack protocol-type

Display the S-Tag TPID for a VLAN using the command show running-config from EXEC privilege mode. FTOS displays the S-Tag TPID only if it is a non-default value.

670

Service Provider Bridging

FTOS Options for Trunk Ports

802.1ad trunk ports may also be tagged members of a VLAN so that it can carry single and double-tagged traffic. You can enable trunk ports to carry untagged, single-tagged, and double-tagged VLAN traffic by making the trunk port a hybrid port.
Step 1 Task Configure a trunk port to carry untagged, single-tagged, and double-tagged traffic by making it a hybrid port. Note: Note: On the C-Series and S-Series, a trunk port can be added to an 802.1Q VLAN as well as a Stacking VLAN only when the TPID 0x8100. Add the port to a 802.1Q VLAN as tagged or untagged. Command Syntax
portmode hybrid

Command Mode INTERFACE

[tagged | untagged]

INTERFACE VLAN

In Figure 35-4 GigabitEthernet 0/1 a trunk port that is configured as a hybrid port and then added to VLAN 100 as untagged VLAN 101 as tagged, and VLAN 103, which is a stacking VLAN.
Figure 35-4. Hybrid Port as VLAN-Stack Trunk Port and as Member of other VLANs
FTOS(conf)#int gi 0/1 FTOS(conf-if-gi-0/1)#portmode hybrid FTOS(conf-if-gi-0/1)#switchport FTOS(conf-if-gi-0/1)#vlan-stack trunk FTOS(conf-if-gi-0/1)#show config ! interface GigabitEthernet 0/1 no ip address portmode hybrid switchport vlan-stack trunk shutdown FTOS(conf-if-gi-0/1)#interface vlan 100 FTOS(conf-if-vl-100)#untagged gigabitethernet 0/1 FTOS(conf-if-vl-100)#interface vlan 101 FTOS(conf-if-vl-101)#tagged gigabitethernet 0/1 FTOS(conf-if-vl-101)#interface vlan 103 FTOS(conf-if-vl-103)#vlan-stack compatible FTOS(conf-if-vl-103-stack)#member gigabitethernet 0/1 FTOS(conf-if-vl-103-stack)#do show vlan Codes: Q: U x G * - Default VLAN, G - GVRP VLANs Untagged, T - Tagged Dot1x untagged, X - Dot1x tagged GVRP tagged, M - Vlan-stack Status Inactive Inactive Inactive Inactive Description Q Ports U Gi 0/1 T Gi 0/1 M Gi 0/1

NUM 1 100 101 103

Service Provider Bridging | 671

www.dell.com | support.dell.com

Debug VLAN Stacking

To debug the internal state and membership of a VLAN and its ports, use the debug member command, as shown in Figure 35-5. The port notations in Figure 35-5 are as follows: MT stacked trunk MU stacked access port T 802.1Q trunk port U 802.1Q access port NU Native VLAN (untagged)

Figure 35-5. Example of Output of debug member vlan and debug member port
FTOS# debug member vlan 603 vlan id : 603 ports : Gi 2/47 (MT), Gi 3/1(MU), Gi 3/25(MT), Gi 3/26(MT), Gi 3/27(MU) FTOS#debug member port gigabitethernet 2/47 vlan id : 603 (MT), 100(T), 101(NU) FTOS#

VLAN Stacking in Multi-vendor Networks

The first field in the VLAN tag is the Tag Protocol Identifier (TPID), which is two bytes. In a VLAN-stacking network, once the frame is double tagged, the outer tag TPID must match the TPID of the next-hop system. While 802.1Q requires that the inner tag TPID is 0x8100, it does not require a specific value for the outer tag TPID. Systems may use any two-byte value; FTOS uses 0x9100 (Figure 35-6) while non-Dell Networking systems might use a different value. If the next-hop systems TPID does not match the outer-tag TPID of the incoming frame, the system drops the frame. For example, in Figure 35-6, the frame originating from Building A is tagged VLAN RED, and then double-tagged VLAN PURPLE on egress at R4. The TPID on the outer tag is 0x9100. R2s TPID must also be 0x9100, and it is, so R2 forwards the frame. Given the matching-TPID requirement, there are limitations when you employ Dell Networking systems at network edges, at which, frames are either double tagged on ingress (R4) or the outer tag is removed on egress (R3).

VLAN Stacking with E-Series TeraScale Systems

The default TPID for the outer VLAN tag is 0x9100. Although the TPID field is 16 bits, E-Series TeraScale only uses the first eight to make forwarding decisions, and as such makes no distinction between 0x8100 and any other TPID beginning with 0x81, for example, 0x8181. You can configure the first eight bits of the TPID using the command vlan-stack protocol-type command. In Figure 35-6, the frame originating from Building C is tagged 0x9191 on egress of R1. R2s TPID is 0x9100, but it its an E-Series TeraScale system and makes no distinction between 0x9191 and 0x9100, so it forwards the frame.

672

Service Provider Bridging

Figure 35-6. TPID Match and First-byte Match on the E-Series TeraScale

Building D
INTE RN ET
VLA

CE PROVIDER RVI SE
TPID 0x9191

NB

LUE

VLAN GREEN

VLAN GREEN, VLAN

PURPLE
R3-E-Series TeraScale TPID: 0x9100

VLAN RE

UE N BL VLA
R1-E-Series TeraScale TPID: 0x9191

R2-E-Series TeraScale TPID: 0x9100

VL

AN

PU

RP LE
TPID (0x9100) PCP CFI (0)

Building B
VID (VLAN Purple) TPID (0x8100) PCP CFI (0) VID (VLAN Red)

Building C
R4-Non-Force10 System TPID: 0x9100

R4

TPID 0x8100 on E-Series TeraScale Systems

E-Series TeraScale treats TPID 0x8100 as a normal VLAN even when on the outer tag. E-Series TeraScale makes forwarding decisions based strictly on the protocol type, without regard for whether the port is an access port. Therefore, when the outer tag has TPID 0x8100, the system does not remove it from frames egressing an access port. Still, although the frames cannot be decapsulated, the system is able to switch them. In Figure 35-7, the frame originating from Building A is double tagged on egress at R4 and is switched towards Building B, but is not decapsulated on egress at R2 because its TPID is 0x8181.
FTOS Behavior: The E-Series ExaScale and TeraScale forward frames with TPID 0x8100 even when its own TPID is not 0x8100. This behavior is required to service ARP and PVST packets, which use TPID 0x8100.

VL

TPID (0x8100)

PCP

CFI (0)

VID (VLAN Red)

Building A

AN D RE

Service Provider Bridging | 673

www.dell.com | support.dell.com

Figure 35-7. TPID Mismatch and 0x8100 Match on the E-Series TeraScale

INTE RN ET

TPID 0x9100

VLA

CE PROVIDER RVI SE

LUE

Building D
TPID 0x8100

NB

VLAN GREEN
UE N BL VLA

R1-E-Series TeraScale TPID: 0x9100

R2-E-Series TeraScale TPID: 0x8181

VLAN GREEN, VLAN

VL

PURPLE
R3-E-Series TeraScale TPID: 0x8181

VLAN R

ED

X
Building B
VID (VLAN Purple) TPID (0x8100) PCP CFI (0) VID (VLAN Red)

AN

PU

RP LE
TPID (0x8100) PCP CFI (0)

Building C
R4-Non-Force10 System TPID: 0x8100
VL
TPID (0x8100) PCP CFI (0) VID (VLAN Red)

Building A

VLAN Stacking with E-Series ExaScale Systems

E-Series ExaScale, beginning with FTOS version 8.2.1.0, allows you to configure both bytes of the 2-byte TPID. TeraScale systems allow you to configure the first byte only and thus, the system did not differentiate between TPIDs with a common first byte. For example 0x9100 and 0x91A8 were treated as the same TPID. In Figure 35-6, R2 forwards the frame with TPID 0x9191 which originated from Building C. In contrast, R2 drops the frame with TPID 0x9191 originating from Building C in Figure 35-8 because the frames TPID does not match both bytes of its own TPID.
FTOS Behavior: The E-Series ExaScale and TeraScale forwards frames with TPID 0x8100 even when its own TPID is not 0x8100. This behavior is required to service ARP and PVST packets, which use TPID 0x8100.

AN D RE

674

Service Provider Bridging

Figure 35-8. First-byte TPID Match on the E-Series ExaScale

INTE RN ET

TPID 0x9191
PURPLE

VLA

CE PROVIDER RVI SE

NB

LUE

Building D

VLAN GREEN

UE N BL VLA

R1-E-Series TeraScale TPID: 0x9191

R2-E-Series ExaScale TPID: 0x9100

VLAN GREEN, VLAN

VLAN R

ED

VL

AN

PU

RP LE

Building C
VL AN D RE

Table 35-1 details the outcome of matched and mis-matched TPIDs in a VLAN-stacking network with the E-Series.
Table 35-1. Network Position Core E-Series Behaviors for Mis-matched TPID Incoming Packet TPID 0xUVWX 0xUVWZ Egress Access Point 0xUVWX 0x81WX 0xUVWZ System TPID 0xUVYZ 0xQRST 0xUVYZ 0x81YZ 0xQRST

Match Type 1st-byte match mismatch 1st-byte match 1st-byte match mismatch

TeraScale Behavior switch as 0xUVYZ drop switch as 0xUVYZ switch as is (no decapsulation) drop

ExaScale Behavior drop drop drop drop drop

VLAN Stacking with C-Series and S-Series

The default TPID for the outer VLAN tag is 0x9100. Beginning with FTOS version 8.2.1.0, both the C-Series and S-Series allow you to configure both bytes of the 2-byte TPID. Previous versions allowed you to configure the first byte only, and thus, the systems did not differentiate between TPIDs with a common first byte. For example 0x8100 and any other TPID beginning with 0x81 were treated as the same TPID, as shown in Figure 35-9. Versions 8.2.1.0 and later differentiate between 0x9100 and 0x91XY, as shown in Figure 35-11.

Service Provider Bridging | 675

www.dell.com | support.dell.com

You can configure the first eight bits of the TPID using the command vlan-stack protocol-type. The TPID on the C-Series and S-Series systems is global. Ingress frames that do not match the system TPID are treated as untagged. This rule applies for both the outer tag TPID of a double-tagged frame and the TPID of a single-tagged frame. For example, if you configure TPID 0x9100, then the system treats 0x8100 and untagged traffic the same and maps both types to the default VLAN, as shown by the frame originating from Building C in Figure 35-11. For the same traffic types, if you configure TPID 0x8100, then the system is able to differentiate between 0x8100 and untagged traffic and maps each to the appropriate VLAN, as shown by the packet originating from Building A in Figure 35-11. Therefore, a mismatched TPID results in the port not differentiating between tagged and untagged traffic.
Figure 35-9. Single and Double-tag TPID Match on the C-Series and S-Series

Building C
UE N BL VLA

TPID 0x8100 R2-C-Series TPID: 0x8100

VLAN GREEN
VL

VLA

NB

LUE

Building D

VLAN GREEN, VLAN

PURPLE

VLAN R

ED

R1-C-Series TPID: 0x8100

AN

R3-C-Series TPID: 0x8100

PU RP LE
TPID (0x8100) PCP CFI (0)

Building B
VID (VLAN Purple) TPID (0x8100) PCP CFI (0) VID (VLAN Red)

TE

R4-Non-Force10 System TPID: 0x8100

IN

TPID (0x8100)

PCP

RN

CFI (0)

VID (VLAN Red)

ET S

E R V I C E P R O VID

ER
VL AN D RE

Building A

676

Service Provider Bridging

Figure 35-10.

Single and Double-tag First-byte TPID Match on C-Series and S-Series

DEFAULT VLAN

TPID 0x8181 R2-C-Series w/ FTOS <8.2.1.0 ED TPID: 0x8181 VLAN R PURPLE VLAN GREEN, VLAN EN GRE VLAN UE DEFAULT VLAN N BL R3-C-Series w/ FTOS >=8.2.1.0 VL VLA TPID: 0x8181 AN PU R1-C-Series w/ FTOS <8.2.1.0 RP TPID: 0x8181 LE

VLA

NB

LUE

Building B

TE

R4-Non-Force10 System TPID: 0x8100

IN

TPID (0x8100)

PCP

RN

CFI (0)

VID (VLAN Red)

ET S

E R V I C E P R O VID

ER
VL AN D RE

Building A

Figure 35-11.

Single and Double-tag TPID Mismatch on the C-Series and S-Series

DEFAULT VLAN

DEFAULT VLAN VLA

TPID 0x9100

TPID 0x8181 R2-E-Series w/ FTOS version <8.2.1.0 TPID: 0x8181 PURPLE VLAN GREEN, VLAN VLAN GREEN
VL AN

Building C
UE N BL VLA

VLA

NB

LUE

Building D

VLAN R

ED

R3-C-Series TPID: 0x9100

PU RP LE
TPID (0x8181) PCP CFI (0)

R1-C-Series TPID: 0x9100

Building B
VID (VLAN Purple) TPID (0x8100) PCP CFI (0) VID (VLAN Red)

TE

R4-Non-Force10 System TPID: 0x8181

IN

TPID (0x8100)

PCP

RN

CFI (0)

VID (VLAN Red)

ET S

E R V I C E P R O VID

ER
VL AN D RE

R4

Building A

Service Provider Bridging | 677

www.dell.com | support.dell.com

Table 35-2 details the outcome of matched and mismatched TPIDs in a VLAN-stacking network with the C-Series and S-Series.
Table 35-2. Network Position C-Series and S-Series Behaviors for Mis-matched TPID Incoming Packet TPID System TPID 0xUVWX 0xUVWX 0x8100 0x81XY Core untagged double-tag 0xUVWX 0xUVWX 0xUVWX 0xUVYZ 0xQRST Egress Access Point untagged double-tag 0xUVWX 0xUVWX 0xUVWX 0xUVYZ 0xQRST

Match Type single-tag mismatch single-tag match

Pre-8.2.1.0 switch to default VLAN switch to default VLAN switch to VLAN

8.2.1.0+ switch to default VLAN switch to default VLAN switch to VLAN switch to default VLAN switch to default VLAN switch to VLAN switch to default VLAN switch to default VLAN switch to default VLAN switch to VLAN switch to default VLAN switch to default VLAN

Ingress Access Point untagged single-tag (0x8100)

single-tag first-byte switch to VLAN match double-tag match double-tag first-byte match double-tag mismatch double-tag match double-tag first-byte match double-tag mismatch switch to default VLAN switch to VLAN switch to VLAN switch to default VLAN switch to default VLAN switch to VLAN switch to VLAN switch to default VLAN

VLAN Stacking Packet Drop Precedence

VLAN Stacking Packet Drop Precedence is available only on platform:

cs

The Drop Eligible Indicator (DEI) bit in the S-Tag indicates to a service provider bridge which packets it should prefer to drop when congested.

678

Service Provider Bridging

Enable Drop Eligibility

You must enable Drop Eligibility globally before you can honor or mark the DEI value.
Task Make packets eligible for dropping based on their DEI value. By default, packets are colored green, and DEI is marked 0 on egress. Command Syntax
dei enable

Command Mode CONFIGURATION

When Drop Eligibility is enabled, DEI mapping or marking takes place according to the defaults. In this case, the CFI is affected according to Table 35-3.
Table 35-3. Ingress Normal Port Trunk Port Drop Eligibility Behavior Egress Normal Port Trunk Port DEI Disabled Retain CFI Retain inner tag CFI Retain outer tag CFI Access Port Trunk Port Retain inner tag CFI Set outer tag CFI to 0 DEI Enabled Set CFI to 0 Retain inner tag CFI Set outer tag CFI to 0 Retain inner tag CFI Set outer tag CFI to 0

Honor the Incoming DEI Value

To honor the incoming DEI value, you must explicitly map the DEI bit to an FTOS drop precedence; precedence can have one of three colors:
Precedence Green Yellow Red Description High priority packets that are the least preferred to be dropped. Lower priority packets that are treated as best-effort. Lowest priority packets that are always dropped (regardless of congestion status).

Task Honor the incoming DEI value by mapping it to an FTOS drop precedence. You may enter the command once for 0 and once for 1. Packets with an unmapped DEI value are colored green. Display the DEI-honoring configuration.

Command Syntax
dei honor {0 | 1} {green | red | yellow}

Command Mode INTERFACE

show interface dei-honor [interface slot/ port | linecard number port-set number]

EXEC Privilege

Service Provider Bridging | 679

www.dell.com | support.dell.com

Task
FTOS#show interface dei-honor

Command Syntax

Command Mode

Default Drop precedence: Green Interface CFI/DEI Drop precedence ------------------------------------------------------------Gi 0/1 0 Green Gi 0/1 1 Yellow Gi 8/9 1 Red Gi 8/40 0 Yellow

Mark Egress Packets with a DEI Value

On egress, you can set the DEI value according to a different mapping than ingress (see Honor the Incoming DEI Value).
Task Set the DEI value on egress according to the color currently assigned to the packet. Display the DEI-marking configuration.
FTOS#show interface dei-mark Default CFI/DEI Marking: 0 Interface Drop precedence CFI/DEI -----------------------------------------------Gi 0/1 Green 0 Gi 0/1 Yellow 1 Gi 8/9 Yellow 0 Gi 8/40 Yellow 0

Command Syntax
dei mark {green | yellow} {0 | 1}

Command Mode INTERFACE EXEC Privilege

show interface dei-mark [interface slot/ port | linecard number port-set number]

Dynamic Mode CoS for VLAN Stacking

Dynamic Mode CoS for VLAN Stacking is available only on platforms:

cs

One of the ways to ensure quality of service for customer VLAN-tagged frames is to use the 802.1p priority bits in the tag to indicate the level of QoS desired. When an S-Tag is added to incoming customer frames, the 802.1p bits on the S-Tag may be configured statically for each customer or derived from the C-Tag using Dynamic Mode CoS. Dynamic Mode CoS maps the C-Tag 802.1p value to a S-Tag 802.1p value.

680

Service Provider Bridging

Figure 35-12.

Statically and Dynamically Assigned dot1p for VLAN Stacking

Untagged

S-Tag with statically-assigned dot1p

S-Tag

DATA

0x0800

SA

DA

DATA

0x0800

400

0x9100

SA

DA

C-Tag

C-Tag

S-Tag

100

0x8100

SA

DA

100

0x8100

400

0x9100

SA

DA

C-Tagged

S-Tag with mapped dot1p

When configuring Dynamic Mode CoS, you have two options: a b mark the S-Tag dot1p and queue the frame according to the original C-Tag dot1p. In this case, you must have other dot1p QoS configurations; this option is classic dot1p marking. mark the S-Tag dot1p and queue the frame according to the S-Tag dot1p. For example, if frames with C-Tag dot1p values 0, 6 and 7 are mapped to an S-Tag dot1p value 0, then all such frames are sent to the queue associated with the S-Tag 802.1p value 0. This option requires two different CAM entries, each in a different Layer 2 ACL FP block.

Note: The ability to map incoming C-Tag dot1p to any S-Tag dot1p requires up to 8 entries to be installed in the Layer 2 QoS and Layer 2 ACL table for each configured customer VLAN. The scalability of this feature is limited by the impact of the 1:8 expansion in these CAM tables.

Service Provider Bridging | 681

www.dell.com | support.dell.com

FTOS Behavior: For Option A above, when there is a conflict between the queue selected by Dynamic Mode CoS (vlan-stack dot1p-mapping) and a QoS configuration, the queue selected by Dynamic Mode CoS takes precedence. However, rate policing for the queue is determined by QoS configuration. For example, the following access-port configuration maps all traffic to Queue 0: vlan-stack dot1p-mapping c-tag-dot1p 0-7 sp-tag-dot1p 1 However, if the following QoS configuration also exists on the interface, traffic is queued to Queue 0 but will be rate policed at 40Mbps (qos-policy-input for queue 3) since class-map "a" of Queue 3 also matches the traffic. This behavior is expected. policy-map-input in layer2 service-queue 3 class-map a qos-policy 3 ! class-map match-any a layer2 match mac access-group a ! mac access-list standard a seq 5 permit any ! qos-policy-input 3 layer2 rate-police 40

Likewise, in the configuration below, packets with dot1p priority 0 3 are marked as dot1p 7 in the outer tag and queued to Queue 3. Rate policing is according to qos-policy-input 3. All other packets will have outer dot1p 0 and hence are queued to Queue 1. They are therefore policed according to qos-policy-input 1. A policy map output with rate shape for different queues can also be used.
policy-map-input in layer2 service-queue 1 qos-policy 1 service-queue 3 qos-policy 3 ! qos-policy-input 1 layer2 rate-police 10 ! qos-policy-input 3 layer2 rate-police 30 ! interface GigabitEthernet 0/21 no ip address switchport vlan-stack access vlan-stack dot1p-mapping c-tag-dot1p 0-3 sp-tag-dot1p 7 service-policy input in layer2 no shutdown

682

Service Provider Bridging

To map C-Tag dot1p values to S-Tag dot1p values and mark the frames accordingly:
Step 1 Task Allocate CAM space to enable queuing frames according to the C-Tag or the S-Tag. vman-qos: mark the S-Tag dot1p and queue the frame according to the original C-Tag dot1p. This method requires half as many CAM entries as vman-qos-dual-fp. vman-qos-dual-fp: mark the S-Tag dot1p and queue the frame according to the S-Tag dot1p. This method requires twice as many CAM entries as vman-qos and FP blocks in multiples of 2. The new CAM configuration is stored in NVRAM and takes effect only after a save and reload. Map C-Tag dot1p values to a S-Tag dot1p value. C-Tag values may be separated by commas, and dashed ranges are permitted. Dynamic Mode CoS overrides any Layer 2 QoS configuration in case of conflicts. Command Syntax
cam-acl l2acl number ipv4acl number ipv6acl number ipv4qos number l2qos number l2pt number ipmacacl number ecfmacl number {vman-qos | vman-qos-dual-fp} number

Command Mode CONFIGURATION

Default: 0 FP blocks for vman-qos and vman-qos-dual-fp

copy running-config startup-config reload

EXEC Privilege

vlan-stack dot1p-mapping c-tag-dot1p values sp-tag-dot1p value

INTERFACE

Note: Since dot1p-mapping marks and queues packets, the only remaining applicable QoS configuration is rate metering. You may use Rate Shaping or Rate Policing.

Layer 2 Protocol Tunneling

Layer 2 Protocol Tunneling (L2PT) is supported on platforms:

ces

Spanning Tree BPDUs use a reserved destination MAC address called the Bridge Group Address, which is 01-80-C2-00-00-00. Only spanning-tree bridges on the LAN recognize this address and process the BPDU. When VLAN stacking is used to connect physically separate regions of a network, BPDUs attempting to traverse the intermediate network might be consumed and subsequently dropped because the intermediate network itself might be using Spanning Tree (Figure 35-13).

Service Provider Bridging | 683

www.dell.com | support.dell.com

Figure 35-13.

VLAN Stacking without L2PT

SPANNI NG TR

EE

EE TR

no spanning-tree
ETWORK EN RE

SPAN NIN G

INTE RN E

ING TREE ANN SP CE PROVIDER w/ I V R SE

Building B
no spanning-tree

X
BPDU w/ destination MAC address: 01-80-C2-00-00-00

Building A

You might need to transport control traffic transparently through the intermediate network to the other region. Layer 2 Protocol Tunneling enables BPDUs to traverse the intermediate network by identifying frames with the Bridge Group Address, rewriting the destination MAC to a user-configured non-reserved address, and forwarding the frames. Since the frames now use a unique MAC address, BPDUs are treated as normal data frames by the switches in the intermediate network core. On egress edge of the intermediate network, the MAC address rewritten to the original MAC address and forwarded to the opposing network region (Figure 35-14).
FTOS Behavior: In FTOS versions prior to 8.2.1.0, the MAC address that Dell Networking systems use to overwrite the Bridge Group Address on ingress was non-configurable. The value of the L2PT MAC address was the Dell Networking-unique MAC address, 01-01-e8-00-00-00. As such, with these FTOS versions, Dell Networking systems are required at the egress edge of the intermediate network because only FTOS could recognize the significance of the destination MAC address and rewrite it to the original Bridge Group Address. In FTOS version 8.2.1.0 and later, the L2PT MAC address is user-configurable, so you can specify an address that non-Dell Networking systems can recognize and rewrite the address at egress edge.

684

Service Provider Bridging

Figure 35-14.

VLAN Stacking with L2PT

SPANNI NG TR

EE

EE TR

no spanning-tree
ETWORK EN RE

SPAN NIN G

INTE RN E

NING TREE PAN /S CE PROVIDER w I V R SE BPDU w/ destination T MAC address: 01-01-e8-00-00-00

Building B

BPDU w/ destination MAC address: 01-80-C2-00-00-00 no spanning-tree

R1-E-Series BPDU w/ destination MAC address: 01-80-C2-00-00-00

R2 Non-Dell Force10 System

R3 Non-Dell Force10 System

Building A

Implementation Information
L2PT is available for STP, RSTP, MSTP, and PVST+ BPDUs. No protocol packets are tunneled when VLAN Stacking is enabled. L2PT requires the default CAM profile.

Enable Layer 2 Protocol Tunneling

Step 1 2 3 Task Verify that the system is running the default CAM profile; you must use this CAM profile for L2PT. Enable protocol tunneling globally on the system. Tunnel BPDUs the VLAN. Command Syntax
show cam-profile

Command Mode EXEC Privilege CONFIGURATION INTERFACE VLAN

protocol-tunnel enable protocol-tunnel stp

Service Provider Bridging | 685

www.dell.com | support.dell.com

Specify a Destination MAC Address for BPDUs

By default, FTOS uses a Dell Networking-unique MAC address for tunneling BPDUs. You can configure another value.
Task Overwrite the BPDU with a user-specified destination MAC address when BPDUs are tunneled across the provider network. Default: 01:01:e8:00:00:00 Command Syntax
protocol-tunnel destination-mac

Command Mode CONFIGURATION

Rate-limit BPDUs on the E-Series

In order to rewrite the destination MAC address on BPDUs, they are forwarded to the RPM. You can rate-limit BPDUs to protect the RPM, in which case the system drops BPDUs when the threshold is reached.
Task Set a maximum rate at which the RPM will process BPDUs for L2PT. Default: 75 pps E-Series Range: 75 to 3000 pps Command Syntax
protocol-tunnel rate-limit

Command Mode CONFIGURATION

Rate-limit BPDUs on the C-Series and S-Series

CAM space is allocated in sections called Field Processor (FP) blocks. There are total 13 user-configurable FP blocks on the C-Series and S-Series. The default number of blocks for L2PT is 0; you must allocate at least one to enable BPDU rate-limiting.
Step 1 Task Create at least one FP group for L2PT. See CAM Allocation for details on this command. Save the running-config to the startup-config. Reload the system. Set a maximum rate at which the RPM will process BPDUs for L2PT. Default: no rate limiting C-Series Range: 64 to 640 kbps S-Series Range: 64 to 320 kbps Command Syntax
cam-acl l2acl

Command Mode CONFIGURATION

2 3 4

copy running-config startup-config

EXEC Privilege EXEC Privilege VLAN STACKING

reload
protocol-tunnel rate-limit

686

Service Provider Bridging

Debug Layer 2 Protocol Tunneling

Task Display debugging information for L2PT. Command Syntax
debug protocol-tunnel

Command Mode EXEC Privilege

Provider Backbone Bridging

Provider Backbone Bridging is supported only on platforms:

cs

IEEE 802.1adProvider Bridges amends 802.1QVirtual Bridged Local Area Networks so that service providers can use 802.1Q architecture to offer separate VLANs to customers with no coordination between customers, and minimal coordination between customers and the provider. 802.1ad specifies that provider bridges operating Spanning Tree use a reserved destination MAC address called the Provider Bridge Group Address, 01-80-C2-00-00-08, to exchange BPDUs instead of the Bridge Group Address, 01-80-C2-00-00-00, originally specified in 802.1Q. Only bridges in the service provider network use this destination MAC address so these bridges treat BPDUs originating from the customer network as normal data frames, rather than consuming them. The same is true for GVRP. 802.1ad specifies that provider bridges participating in GVRP use a reserved destination MAC address called the Provider Bridge GVRP Address, 01-80-C2-00-00-0D, to exchange GARP PDUs instead of the GVRP Address, 01-80-C2-00-00-21, specified in 802.1Q. Only bridges in the service provider network use this destination MAC address so these bridges treat GARP PDUs originating from the customer network as normal data frames, rather than consuming them. Provider Backbone Bridging through IEEE 802.1ad eliminates the need for tunneling BPDUs with L2PT and increases the reliability of provider bridge networks as the network core need only learn the MAC addresses of core switches, as opposed to all MAC addresses received from attached customer devices.
Task Use the Provider Bridge Group address as the destination MAC address in BPDUs. The xstp keyword applies this functionality to STP, RSTP, and MSTP; this functionality is not available for PVST+. Command Syntax
bpdu-destination-mac-address [stp | gvrp] provider-bridge-group

Command Mode CONFIGURATION

Service Provider Bridging | 687

688 www.dell.com | support.dell.com

Service Provider Bridging

36
sFlow
Configuring sFlow is supported on platforms Enable and Disable sFlow sFlow Show Commands Specify Collectors Polling Intervals Sampling Rate Back-off Mechanism sFlow on LAG ports Extended sFlow

ces

Overview
FTOS supports sFlow version 5. sFlow is a standard-based sampling technology embedded within switches and routers which is used to monitor network traffic. It is designed to provide traffic monitoring for high speed networks with many switches and routers. sFlow uses two types of sampling: Statistical packet-based sampling of switched or routed packet flows Time-based sampling of interface counters

The sFlow monitoring system consists of an sFlow Agent (embedded in the switch/router) and an sFlow collector. The sFlow Agent resides anywhere within the path of the packet, and combines the flow samples and interface counters into sFlow datagrams and forwards them to the sFlow Collector at regular intervals. The datagrams consists of information on, but not limited to, packet header, ingress and egress interfaces, sampling parameters, and interface counters. Packet sampling is typically done by the ASIC. sFlow Collector analyses the sFlow datagrams received from different devices and produces a network-wide view of traffic flows.

sFlow | 689

www.dell.com | support.dell.com

Figure 36-1. sFlow Traffic Monitoring System sFlow Collector sFlow Datagrams

Switch/Router

sFlow Agent Poll Interface Counters

Interface Counters Switch ASIC

Flow Samples

Implementation Information
Dell Networking sFlow is designed so that the hardware sampling rate is per line card port-pipe and is decided based upon all the ports in that port-pipe. If sFlow is not enabled on any port specifically, then the global sampling rate is downloaded to that port and is to calculate the port-pipes lowest sampling rate. This design supports, then, the possibility that sFlow might be configured on that port in the future. Back-off is triggered based on the port-pipes hardware sampling rate. For example, if port 1 in a the port-pipe has sFlow configured with a 16384 sampling rate while port 2 in the port-pipe has sFlow configured but no sampling rate set, FTOS applies a global sampling rate of 512 to port 2. The hardware sampling rate on the port-pipe is then set at 512 because that is the lowest configured rate on the port-pipe. When a high traffic situation occurs, a back-off is triggered and the hardware sampling rate is backed-off from 512 to 1024. Note that port 1 maintains its sampling rate of 16384; port 1 is unaffected because it maintains its configured sampling rate of 16484. To avoid the back-off, either increase the global sampling rate or configure all the line card ports with the desired sampling rate even if some ports have no sFlow configured.

Important Points to Remember

The FTOS implementation of the sFlow MIB supports sFlow configuration via snmpset. Collection through management interface is supported on E-Series only Dell Networking recommends that the sFlow Collector be connected to the Dell Networking chassis through a line card port rather than the RPM Management Ethernet port. E-Series TeraScale sFlow sampling is done on a per-port-pipe basis. E-Series ExaScale, C-Series , and S-Series sFlow sampling is done on a per-port basis.

690

sFlow

FTOS exports all sFlow packets to the collector. A small sampling rate can equate to a large number of exported packets. A backoff mechanism will automatically be applied to reduce this amount. Some sampled packets may be dropped when the exported packet rate is high and the backoff mechanism is about to or is starting to take effect. The dropEvent counter, in the sFlow packet, will always be zero. Community list and local preference fields are not filled in extended gateway element in sFlow datagram. 802.1P source priority field is not filled in extended switch element in sFlow datagram. Only Destination and Destination Peer AS number are packed in the dst-as-path field in extended gateway element If packet being sampled is redirected using PBR (Policy-Based Routing), sFlow datagram may contain incorrect extended gateway/router information. Source VLAN field in the extended switch element will not be packed in case of routed packet. Destination VLAN field in the extended switch element will not be packed in case of Multicast packet. On the S-Series, up to 700 packets can be sampled and processed per second. On the C-Series up to 1000 packets can be sampled and processed per second. On the E-Series, the maximum number of packets that can be sampled and processed per second is: 7500 packets when no extended information packing is enabled. 1000 packets when only extended-switch information packing is enabled. 1600 packets when extended-router and/or extended-gateway information packing is enabled.

Enable and Disable sFlow

By default, sFlow is disabled globally on the system. To enable sFlow globally, use the sflow enable command in CONFIGURATION mode. Use the no version of this command to disable sFlow globally.
Command Syntax [no] sflow enable Command Mode CONFIGURATION Usage Enable sFlow globally.

Enable and Disable on an Interface

By default, sFlow is disabled on all interfaces. To enable sFlow on a specific interface, use the sflow enable command in INTERFACE mode. Use the no version of this command to disable sFlow on an interface. This CLI is supported on physical ports and LAG ports.
Command Syntax [no] sflow enable Command Mode INTERFACE Usage Enable sFlow on an interface.

sFlow | 691

www.dell.com | support.dell.com

sFlow Show Commands

FTOS includes the following sFlow display commands: Show sFlow Globally on page 49 Show sFlow on an Interface on page 50 Show sFlow on a Line Card on page 50

Show sFlow Globally

Use the following command to view sFlow statistics:
Command Syntax show sflow Command Mode EXEC Purpose Display sFlow configuration information and statistics.

Figure 36-2 is a sample output from the show sflow command:

Figure 36-2. Command Example: show sflow
FTOS#show sflow Indicates sFlow is globally enabled sFlow services are enabled Global default sampling rate: 32768 Global default counter polling interval: 20 1 collectors configured Collector IP addr: 133.33.33.53, Agent IP addr: 133.33.33.116, UDP port: 6343 77 UDP packets exported 0 UDP packets dropped 165 sFlow samples collected 69 sFlow samples dropped due to sub-sampling Indicates sFlow is enabled on linecards Gi 1/16 and Gi 1/17 Linecard 1 Port set 0 H/W sampling rate 8192 Gi 1/16: configured rate 8192, actual rate 8192, sub-sampling rate 1 Gi 1/17: configured rate 16384, actual rate 16384, sub-sampling rate 2

Show sFlow on an Interface

Use the following command to view sFlow information on a specific interface:
Command Syntax show sflow interface interface-name Command Mode EXEC Purpose Display sFlow configuration information and statistics on a specific interface.

Figure 36-3 is a sample output from the show sflow interface command.

692

sFlow

Figure 36-3. Command Example: show sflow interface

FTOS#show sflow interface gigabitethernet 1/16 Gi 1/16 Configured sampling rate :8192 Actual sampling rate :8192 Sub-sampling rate :2 Counter polling interval :15 Samples rcvd from h/w :33 Samples dropped for sub-sampling :6

The configuration, shown in Figure 36-2, is also displayed in the running configuration (Figure 36-4):
Figure 36-4. Command Example: show running-config interface
FTOS#show running-config interface gigabitethernet 1/16 ! interface GigabitEthernet 1/16 no ip address mtu 9252 ip mtu 9234 switchport sflow enable sflow sample-rate 8192 no shutdown

Show sFlow on a Line Card

Use the following command to view sFlow statistitics on a specified line card:
Command Syntax show sflow linecard slot-number Command Mode EXEC Purpose Display sFlow configuration information and statistics on the specified interface.

Figure 36-5 is a sample output from the show sflow linecard command:
Figure 36-5. Command Example: show sflow linecard
FTOS#show sflow linecard 1 Linecard 1 Samples rcvd from h/w Samples dropped for sub-sampling Total UDP packets exported UDP packets exported via RPM UDP packets dropped

:165 :69 :77 :77 :

sFlow | 693

www.dell.com | support.dell.com

Specify Collectors
The sflow collector command allows identification of sFlow Collectors to which sFlow datagrams are forwarded. The user can specify up to two sFlow collectors. If two Collectors are specified, the samples are sent to both. Collection through Management interface is supported on platform:
Command Syntax
sflow collector ip-address agent-addr ip-address [number [max-datagram-size number] ] | [max-datagram-size number ]

e.

Command Mode CONFIGURATION

Usage Identify sFlow collectors to which sFlow datagrams are forwarded. Default UDP port: 6343 Default max-datagram-size: 1400

Polling Intervals
The sflow polling-interval command configures the polling interval for an interface in the maximum number of seconds between successive samples of counters to be sent to the collector. This command changes the global default counter polling (20 seconds) interval. You can configure an interface to use a different polling interval. The polling interval can be configured globally (in CONFIGURATION mode) or by interface (in INTERFACE mode) by executing the interval command:
.

Command Syntax
sflow polling-interval
interval value

Command Mode CONFIGURATION or INTERFACE

Usage Change the global default counter polling interval.

interval valuein seconds.

Range: 15 to 86400 seconds Default: 20 seconds

Sampling Rate
Sampling Rate is supported on platform

e t.

The sFlow sampling rate is the number of packets that are skipped before the next sample is taken. sFlow does not have time-based packet sampling.

694

sFlow

The sflow sample-rate command, when issued in CONFIGURATION mode, changes the default sampling rate. By default, the sampling rate of an interface is set to the same value as the current global default sampling rate.If the value entered is not a correct power of 2, the command generates an error message with the previous and next power-of-2 value. Select one of these two number and re-enter the command. (For more information on values in power-of-2, see Sub-sampling.) The sample rate can be configured globally or by interface using the sample rate command:
Command Syntax [no] sflow sample-rate sample-rate Command Mode CONFIGURATION or INTERFACE Usage Change the global or interface sampling rate. Rate must be entered in factors of 2 (eg, 4096, 8192). sample-rate range: 256-8388608 for C-Series and S-Series 2-8388608 for E-Series

Sub-sampling
Sub-sampling is available only on platform:

et

The sFlow sample rate is not the frequency of sampling, but the number of packets that are skipped before the next sample is taken. Although a sampling rate can be configured for each port, TeraScale line cards can support only a single sampling rate per port-pipe. Therefore, sFlow Agent uses sub-sampling to create multiple sampling rates per port-pipe. To achieve different sampling rates for different ports in a port-pipe, sFlow Agent takes the lowest numerical value of the sampling rate of all the ports within the port-pipe, and configures all ports to this value. sFlow Agent is then able to skip samples on ports where you require a larger sampling rate value. Sampling rates are configurable in powers of two. This allows the smallest sampling rate possible to be configured on the hardware, and also allows all other sampling rates to be available through sub-sampling. For example, if Gig 1/0 and 1/1 are in a port-pipe, and they are configured with a sampling rate of 4096 on interface Gig 1/0, and 8192 on Gig 1/1, sFlow Agent does the following: 1. Configures the hardware to a sampling rate of 4096 for all ports with sFlow enabled on that port-pipe. 2. Configure interface Gig 1/0 to a sub-sampling rate of 1 to achieve an actual rate of 4096. 3. Configure interface Gig 1/1 to a sub-sampling rate of 2 to achieve an actual rate of 8192.
Note: Sampling rate backoff can change the sampling rate value that is set in the hardware. This equation shows the relationship between actual sampling rate, sub-sampling rate, and the hardware sampling rate for an interface: Actual sampling rate = sub-sampling rate * hardware sampling rate

Note the absence of a configured rate in the equation. That is because when the hardware sampling rate value on the port-pipe exceeds the configured sampling rate value for an interface, the actual rate changes to the hardware rate. The sub-sampling rate never goes below a value of one.

sFlow | 695

www.dell.com | support.dell.com

Back-off Mechanism
If the sampling rate for an interface is set to a very low value, the CPU can get overloaded with flow samples under high-traffic conditions. In such a scenario, a binary back-off mechanism gets triggered, which doubles the sampling-rate (halves the number of samples per second) for all interfaces. The backoff mechanism continues to double the sampling-rate until CPU condition is cleared. This is as per sFlow version 5 draft. Once the back-off changes the sample-rate, users must manually change the sampling rate to the desired value. As a result of back-off, the actual sampling-rate of an interface may differ from its configured sampling rate. The actual sampling-rate of the interface and the configured sample-rate can be viewed by using the show sflow command.

sFlow on LAG ports

When a physical port becomes a member of a LAG, it inherits the sFlow configuration from the LAG port.

Extended sFlow
Extended sFlow is supported fully on platform Platforms

c and s support extended-switch information processing only.

Extended sFlow packs additional information in the sFlow datagram depending on the type of sampled packet. The following options can be enabled:
extended-switch 802.1Q VLAN ID and 802.1p priority information extended-router Next-hop and source and destination mask length. extended-gateway Source and destination AS number and the BGP next-hop.

Note: The entire AS path is not included. BGP community-list and local preference information are not included. These fields are assigned default values and are not interpreted by the collector.

Use the command sflow [extended-switch] [extended-router] [extended-gateway] enable command. By default packing of any of the extended information in the datagram is disabled. Use the command show sflow to confirm that extended information packing is enabled, as shown in Figure 36-6.

696

sFlow

Figure 36-6. Confirming that Extended sFlow is Enabled

FTOS#show sflow sFlow services are enabled Extended sFlow settings Global default sampling rate: 4096 show all 3 types are enabled Global default counter polling interval: 15 Global extended information enabled: gateway, router, switch 1 collectors configured Collector IP addr: 10.10.10.3, Agent IP addr: 10.10.0.0, UDP port: 6343 77 UDP packets exported 0 UDP packets dropped 165 sFlow samples collected 69 sFlow samples dropped due to sub-sampling Linecard 1 Port set 0 H/W sampling rate 8192 Gi 1/16: configured rate 8192, actual rate 8192, sub-sampling rate 1 Gi 1/17: configured rate 16384, actual rate 16384, sub-sampling rate 2 Linecard 3 Port set 1 H/W sampling rate 16384 Gi 3/40: configured rate 16384, actual rate 16384, sub-sampling rate 1

If none of the extended information is enabled, the show output is as shown in Figure 36-7.
Figure 36-7. Confirming that Extended sFlow is Disabled
FTOS#show sflow sFlow services are disabled Global default sampling rate: 32768 Global default counter polling interval: 20 Global extended information enabled: none 0 collectors configured 0 UDP packets exported 0 UDP packets dropped 0 sFlow samples collected 0 sFlow samples dropped due to sub-sampling

Indicates no Extended sFlow types enabled.

Important Points to Remember

The IP destination address has to be learned via BGP in order to export extended-gateway data, prior to FTOS version 7.8.1.0. If the IP destination address is not learned via BGP the Dell Networking system does not export extended-gateway data, prior to FTOS version 7.8.1.0. FTOS 7.8.1.0 and later enhances the sFlow implementation for real time traffic analysis on the E-Series to provide extended gateway information in cases where the destination IP addresses are learned by different routing protocols, and for cases where the destination is reachable over ECMP. If the IP source address is learned via IGP then srcAS and srcPeerAS are zero. The srcAS and srcPeerAS might be zero even though the IP source address is learned via BGP. The Dell Networking system packs the srcAS and srcPeerAS information only if the route is learned via BGP and it is reachable via the ingress interface of the packet.

The previous points are summarized in following table.

sFlow | 697

www.dell.com | support.dell.com

Table 36-1.

Extended Gateway Summary srcAS and srcPeerAS dstAS and dstPeerAS

IP SA static/connected/IGP

IP DA static/connected/IGP

Description Extended gateway data is not exported because there is no AS information. src_as & src_peer_as are zero because there is no AS information for IGP. Prior to FTOS version 7.8.1.0, extended gateway data is not be exported because IP DA is not learned via BGP. 7.8.1.0 allows extended gateway information in cases where the source and destination IP addresses are learned by different routing protocols, and for cases where is source is reachable over ECMP.

static/connected/IGP

BGP

Exported

BGP

static/connected/IGP

Exported

Exported

BGP

BGP

Exported

Exported

Extended gateway data is packed.

698

sFlow

37
Simple Network Management Protocol
Simple Network Management Protocol is supported on platforms

ces

Protocol Overview
Network management stations use Simple Network Management Protocol (SNMP) to retrieve or alter management data from network elements. A datum of management information is called a managed object; the value of a managed object can be static or variable. Network elements store managed objects in a database called a Management Information Base (MIB). MIBs are hierarchically structured and use object identifiers to address managed objects, but managed objects also have a textual name called an object descriptor.

Implementation Information
FTOS supports SNMP version 1 as defined by RFC 1155, 1157, and 1212, SNMP version 2c as defined by RFC 1901, and SNMP version 3 as defined by RFC 2571. FTOS supports up to 15 trap receivers. The FTOS implementation of the sFlow MIB supports sFlow configuration via SNMP sets. SNMP traps for STP and MSTP state changes are based on BRIDGE MIB (RFC 1483) for STP and IEEE 802.1 draft ruzin-mstp-mib-02 for MSTP.

Configure Simple Network Management Protocol

Note: The configurations in this chapter use a Unix environment with net-snmp version 5.4. This is only one of many RFC-compliant SNMP utilities you can use to manage your Dell Networking system using SNMP. Also, these configurations use SNMP version 2c.

Configuring SNMP requires only a single step: 1. Create a community. See page 700.

Simple Network Management Protocol | 699

www.dell.com | support.dell.com

Related Configuration Tasks

The following list contains configuration tasks for SNMP: Read Managed Object Values Write Managed Object Values Subscribe to Managed Object Value Updates using SNMP Copy Configuration Files Using SNMP Manage VLANs using SNMP Enable and Disable a Port using SNMP Fetch Dynamic MAC Entries using SNMP Deriving Interface Indices Viewing the System Image Viewing Alarm LED Status Using SNMP Using SNMP for Entity MIB Queries

Important Points to Remember

Typically, 5-second timeout and 3-second retry values on an SNMP server are sufficient for both LAN and WAN applications. If you experience a timeout with these values, increase the timeout value to greater than 3 seconds, and increase the retry value to greater than 2 on your SNMP server. Group ACLs override user ACLs in SNMPv3 configurations when both are configured and the user is part of the group.

Create a Community
The management station generates requests to either retrieve or alter the value of a management object and is called the SNMP manager. A network element that processes SNMP requests is called an SNMP agent. An SNMP community is a group of SNMP agents and managers that are allowed to interact. Communities are necessary to secure communication between SNMP managers and agents; SNMP agents do not respond to requests from management stations that are not part of the community. FTOS enables SNMP automatically when you create an SNMP community and displays Message 1. You must specify whether members of the community may only retrieve values (read), or retrieve and alter values (read-write). To create an SNMP community:
Task Choose a name for the community. Command
snmp-server community name {ro | rw}

Command Mode CONFIGURATION

700

Simple Network Management Protocol

Message 1 SNMP Enabled

22:31:23: %RPM1-P:CP %SNMP-6-SNMP_WARM_START: Agent Initialized - SNMP WARM_START.

View your SNMP configuration, using the command show running-config snmp from EXEC Privilege mode, as shown in Figure 37-1.
Figure 37-1. Creating an SNMP Community
FTOS#snmp-server community my-snmp-community ro 22:31:23: %RPM1-P:CP %SNMP-6-SNMP_WARM_START: Agent Initialized - SNMP WARM_START. FTOS#do show running-config snmp ! snmp-server community mycommunity ro FTOS#

Read Managed Object Values

You may only retrieve (read) managed object values if your management station is a member of the same community as the SNMP agent. There are several Unix SNMP commands that read data:
Task Read the value of a single managed object, as shown in Figure 37-2. Command
snmpget -v version -c community agent-ip {identifier.instance | descriptor.instance}

Figure 37-2. Reading the Value of a Managed Object

> snmpget -v 2c -c mycommunity 10.11.131.161 sysUpTime.0 DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (32852616) 3 days, 19:15:26.16 > snmpget -v 2c -c mycommunity 10.11.131.161 .1.3.6.1.2.1.1.3.0 DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (32856932) 3 days, 19:16:09.32

Read the value of the managed object directly below the specified object, as shown in Figure 37-3.

snmpgetnext -v version -c community agent-ip {identifier.instance | descriptor.instance}

Figure 37-3. Reading the Value of the Next Managed Object in the MIB
> snmpgetnext -v 2c -c mycommunity 10.11.131.161 .1.3.6.1.2.1.1.3.0 SNMPv2-MIB::sysContact.0 = STRING: > snmpgetnext -v 2c -c mycommunity 10.11.131.161 sysContact.0 SNMPv2-MIB::sysName.0 = STRING: S50V_7.7

Read the value of many objects at once, as shown in Figure 37-4.

snmpwalk -v version -c community agent-ip {identifier.instance | descriptor.instance}

Simple Network Management Protocol | 701

www.dell.com | support.dell.com

Task

Command

Figure 37-4. Reading the Value of Many Managed Objects at Once

> snmpwalk -v 2c -c mycommunity 10.11.131.161 .1.3.6.1.2.1.1 SNMPv2-MIB::sysDescr.0 = STRING: Force10 Networks Real Time Operating System Software Force10 Operating System Version: 1.0 Force10 Application Software Version: E_MAIN4.7.6.350 Copyright (c) 1999-2007 by Force10 Networks, Inc. Build Time: Mon May 12 14:02:22 PDT 2008 SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.6027.1.3.1 DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (32920954) 3 days, 19:26:49.54 SNMPv2-MIB::sysContact.0 = STRING: SNMPv2-MIB::sysName.0 = STRING: S50V_7.7 SNMPv2-MIB::sysLocation.0 = STRING: SNMPv2-MIB::sysServices.0 = INTEGER: 4

Write Managed Object Values

You may only alter (write) a managed object value if your management station is a member of the same community as the SNMP agent, and the object is writable. To write or write-over the value of a managed object:
Task To write or write-over the value of a managed object, as shown in Figure 37-5. Command
snmpset -v version -c community agent-ip {identifier.instance | descriptor.instance}

Figure 37-5. Writing over the Current Value of a Managed Object

> snmpset -v 2c -c mycommunity 10.11.131.161 sysName.0 s "R5" SNMPv2-MIB::sysName.0 = STRING: R5

Configure Contact and Location Information using SNMP

You may configure system contact and location information from the Dell Networking system or from the management station using SNMP. To configure system contact and location information from the Dell Networking system:
Task Identify the system manager along with this persons contact information (e.g E-mail address or phone number). You may use up to 55 characters. Default: None Command
snmp-server contact text

Command Mode CONFIGURATION

702

Simple Network Management Protocol

Task Identify the physical location of the system. For example, San Jose, 350 Holger Way, 1st floor lab, rack A1-1. You may use up to 55 characters. Default: None

Command
snmp-server location text

Command Mode CONFIGURATION

To configure the system from the manumitting station using SNMP:

Task Identify the system manager along with this persons contact information (e.g E-mail address or phone number). You may use up to 55 characters. Default: None Identify the physical location of the system. For example, San Jose, 350 Holger Way, 1st floor lab, rack A1-1. You may use up to 55 characters. Default: None Command
snmpset -v version -c community agent-ip sysContact.0 s contact-info

Command Mode CONFIGURATION

snmpset -v version -c community agent-ip sysLocation.0 s location-info

CONFIGURATION

Subscribe to Managed Object Value Updates using SNMP

This feature is supported on S60 in FTOS version 8.3.3.9. By default, the Dell Networking system displays some unsolicited SNMP messages (traps) upon certain events and conditions. You can also configure the system to send the traps to a management station. Traps cannot be saved on the system. FTOS supports the following three sets of traps: RFC 1157-defined traps: coldStart, warmStart, linkDown, linkUp, authenticationFailure, egpNeighbborLoss Force10 enterpriseSpecific environment traps: fan, supply, temperature Force10 enterpriseSpecific protocol traps: bgp, ecfm, stp, xstp,

To configure the system to send SNMP notifications:

Step 1 Task Configure the Dell Networking system send notifications to an SNMP server. Command
snmp-server host ip-address

Command Mode CONFIGURATION

Simple Network Management Protocol | 703

www.dell.com | support.dell.com

Step

Task Specify which traps the Dell Networking system sends to the trap receiver. Enable all Dell Networking enterpriseSpecific and RFC-defined traps using the command snmp-server enable traps from CONFIGURATION mode. Enable all of the RFC-defined traps using the command snmp-server enable traps snmp from CONFIGURATION mode. Specify the interfaces out of which FTOS sends SNMP traps.

Command
snmp-server enable traps

Command Mode CONFIGURATION

snmp-server trap-source

CONFIGURATION

Table 37-1 lists the traps the RFC-defined SNMP traps and the command used to enable each. Note that the coldStart and warmStart traps are enabled using a single command.
Table 37-1. RFC 1157 Defined SNMP Traps on FTOS Trap
SNMP_AUTH_FAIL:SNMP Authentication failed.Request with invalid community string. SNMP_COLD_START: Agent Initialized - SNMP COLD_START. SNMP_WARM_START: Agent Initialized - SNMP WARM_START. PORT_LINKDN:changed interface state to down:%d PORT_LINKUP:changed interface state to up:%d

Command Option
snmp authentication snmp coldstart snmp linkdown snmp linkup

Enable a subset of Dell Networking enterpriseSpecific SNMP traps using one of the listed command options Table 37-2 with the command snmp-server enable traps. Note that the envmon option enables all environment traps including those that are enabled with the envmon supply, envmon temperature, and envmon fan options.

704

Simple Network Management Protocol

Table 37-2.

Dell Networking Enterprise-specific SNMP Traps Trap

CARD_SHUTDOWN: %sLine card %d down - %s CARD_DOWN: %sLine card %d down - %s LINECARDUP: %sLine card %d is up CARD_MISMATCH: Mismatch: line card %d is type %s - type %s required. RPM_STATE: RPM1 is in Active State RPM_STATE: RPM0 is in Standby State RPM_DOWN: RPM 0 down - hard reset RPM_DOWN: RPM 0 down - card removed HOT_FAILOVER: RPM Failover Completed SFM_DISCOVERY: Found SFM 1 SFM_REMOVE: Removed SFM 1 MAJOR_SFM: Major alarm: Switch fabric down MAJOR_SFM_CLR: Major alarm cleared: Switch fabric up MINOR_SFM: MInor alarm: No working standby SFM MINOR_SFM_CLR: Minor alarm cleared: Working standby SFM present TASK SUSPENDED: SUSPENDED - svce:%d - inst:%d - task:%s RPM0-P:CP %CHMGR-2-CARD_PARITY_ERR ABNORMAL_TASK_TERMINATION: CRASH - task:%s %s CPU_THRESHOLD: Cpu %s usage above threshold. Cpu5SecUsage (%d) CPU_THRESHOLD_CLR: Cpu %s usage drops below threshold. Cpu5SecUsage (%d) MEM_THRESHOLD: Memory %s usage above threshold. MemUsage (%d) MEM_THRESHOLD_CLR: Memory %s usage drops below threshold. MemUsage (%d) DETECT_STN_MOVE: Station Move threshold exceeded for Mac %s in vlan %d

Command Option
envmon

envmon supply

PEM_PRBLM: Major alarm: problem with power entry module %s PEM_OK: Major alarm cleared: power entry module %s is good MAJOR_PS: Major alarm: insufficient power %s MAJOR_PS_CLR: major alarm cleared: sufficient power MINOR_PS: Minor alarm: power supply non-redundant MINOR_PS_CLR: Minor alarm cleared: power supply redundant

envmon temperature

MINOR_TEMP: Minor alarm: chassis temperature MINOR_TEMP_CLR: Minor alarm cleared: chassis temperature normal (%s %d temperature is within threshold of %dC) MAJOR_TEMP: Major alarm: chassis temperature high (%s temperature reaches or exceeds threshold of %dC) MAJOR_TEMP_CLR: Major alarm cleared: chassis temperature lower (%s %d temperature is within threshold of %dC)

envmon fan

FAN_TRAY_BAD: Major alarm: fantray %d is missing or down FAN_TRAY_OK: Major alarm cleared: fan tray %d present FAN_BAD: Minor alarm: some fans in fan tray %d are down FAN_OK: Minor alarm cleared: all fans in fan tray %d are good

Simple Network Management Protocol | 705

www.dell.com | support.dell.com

Table 37-2.

Dell Networking Enterprise-specific SNMP Traps Trap

%SPANMGR-5-STP_NEW_ROOT: New Spanning Tree Root, Bridge ID Address 0001.e801.fc35. Priority 32768,

Command Option
xstp

%SPANMGR-5-STP_TOPOLOGY_CHANGE: Bridge port GigabitEthernet 11/38 transitioned from Forwarding to Blocking state. %SPANMGR-5-MSTP_NEW_ROOT_BRIDGE: Elected root bridge for instance 0. %SPANMGR-5-MSTP_NEW_ROOT_PORT: MSTP root changed to port Gi 11/38 for instance 0. My Bridge ID: 40960:0001.e801.fc35 Old Root: 40960:0001.e801.fc35 New Root: 32768:00d0.038a.2c01. %SPANMGR-5-MSTP_TOPOLOGY_CHANGE: Topology change BridgeAddr: 0001.e801.fc35 Mstp Instance Id 0 port Gi 11/38 transitioned from forwarding to discarding state.

ecfm

%ECFM-5-ECFM_XCON_ALARM: Cross connect fault detected by MEP 1 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_ERROR_ALARM: Error CCM Defect detected by MEP 1 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_MAC_STATUS_ALARM: MAC Status Defect detected by MEP 1 in Domain provider at Level 4 VLAN 3000 %ECFM-5-ECFM_REMOTE_ALARM: Remote CCM Defect detected by MEP 3 in Domain customer1 at Level 7 VLAN 1000 %ECFM-5-ECFM_RDI_ALARM: RDI Defect detected by MEP 3 in Domain customer1 at Level 7 VLAN 1000

<cr>

SNMP Copy Config Command Completed %RPM0-P:CP %SNMP-4-RMON_RISING_THRESHOLD: RMON rising threshold alarm from SNMP OID <oid> %RPM0-P:CP %SNMP-4-RMON_FALLING_THRESHOLD: RMON falling threshold alarm from SNMP OID <oid> %RPM0-P:CP %SNMP-4-RMON_HC_RISING_THRESHOLD: RMON high-capacity rising threshold alarm from SNMP OID <oid>

Copy Configuration Files Using SNMP

This feature is supported on S60 in FTOS version 8.3.3.9. Use SNMP from a remote client to: copy the running-config file to the startup-config file, or copy configuration files from the Dell Networking system to a server copy configuration files from a server to the Dell Networking system

706

Simple Network Management Protocol

The relevant MIBs for these functions are:

Table 37-3. MIB Object copySrcFileType MIB Objects for Copying Configuration Files via SNMP OID .1.3.6.1.4.1.6027.3.5.1.1.1.2 Object Values 1 = FTOS file 2 = running-config 3 = startup-config Description Specifies the type of file to copy from. Valid values are: If the copySrcFileType is running-config or startup-config, the default copySrcFileLocation is flash. If the copySrcFileType is a binary file, the copySrcFileLocation and copySrcFileName must also be specified. Specifies the location of source file. If the copySrcFileLocation is FTP or SCP, copyServerAddress, copyUserName, and copyUserPassword must be specified.

copySrcFileLocation

.1.3.6.1.4.1.6027.3.5.1.1.1.3

1 = flash 2 = slot0 3 = tftp 4 = ftp 5 = scp

copySrcFileName

.1.3.6.1.4.1.6027.3.5.1.1.1.4

Path (if file is not in Specifies name of the file. current directory) If copySourceFileType is set to and filename. running-config or startup-config, copySrcFileName is not required. 1 = FTOS file 2 = running-config 3 = startup-config Specifies the type of file to copy to. If the copySourceFileType is running-config or startup-config, the default copyDestFileLocation is flash. If the copyDestFileType is a binary the copyDestFileLocation and copyDestFileName must be specified. Specifies the location of destination file. If the copyDestFileLocation is FTP or SCP, copyServerAddress, copyUserName, and copyUserPassword must be specified.

copyDestFileType

.1.3.6.1.4.1.6027.3.5.1.1.1.5

copyDestFileLocation

.1.3.6.1.4.1.6027.3.5.1.1.1.6

1 = flash 2 = slot0 3 = tftp 4 = ftp 5 = scp

copyDestFileName

.1.3.6.1.4.1.6027.3.5.1.1.1.7

Path (if file is not in Specifies the name of destination file. default directory) and filename. IP Address of the server The IP address of the server. If the copyServerAddress is specified so must copyUserName, and copyUserPassword. Username for for the FTP, TFTP, or SCP server. If the copyUserName is specified so must copyUserPassword. Password for the FTP, TFTP, or SCP server.

copyServerAddress

.1.3.6.1.4.1.6027.3.5.1.1.1.8

copyUserName

.1.3.6.1.4.1.6027.3.5.1.1.1.9

Username for the server.

copyUserPassword

.1.3.6.1.4.1.6027.3.5.1.1.1.10

Password for the server.

Simple Network Management Protocol | 707

www.dell.com | support.dell.com

To copy a configuration file:

Step 1 2 3 Task Create an SNMP community string with read/ write privileges. Command Syntax
snmp-server community community-name rw

Command Mode CONFIGURATION

Copy the f10-copy-config.mib MIB from the Dell Networking iSupport webpage to the server to which you are copying the configuration file. On the server, use the command snmpset as shown:
snmpset -v snmp-version -c community-name -m mib_path/f10-copy-config.mib force10system-ip-address mib-object.index {i | a | s} object-value...

Every specified object must have an object value, which must be preceded by the keyword i. See Table 6 for valid values. index must be unique to all previously executed snmpset commands. If an index value has been used previously, a message like the one in Message 3 appears. In this case, increment the index value and enter the command again. Use as many MIB Objects in the command as required by the MIB Object descriptions in Table 6 to complete the command. See Table 7 or examples.

Note: You can use the entire OID rather than the object name. Use the form: OID.index i object-value, as shown in Figure 57.

Message 2 snmpset Index Value Error

Error in packet. Reason: notWritable (that object does not support modification) Failed object: FORCE10-COPY-CONFIG-MIB::copySrcFileType.101

Table 7 shows examples of using the command snmpset to copy a configuration. These examples assume that: the server OS is Unix you are using SNMP version 2c the community name is public, and the file f10-copy-config.mib is in the current directory or in the snmpset tool path.
Note: In Unix, enter the command snmpset for help using this command. Place the file f10-copy-config.mib the directory from which you are executing the snmpset command or in the

snmpset tool path.

Table 37-4. Task

Copying Configuration Files via SNMP

Copy the running-config to the startup-config using the following command from the Unix machine:

708

Simple Network Management Protocol

Table 37-4. Task

Copying Configuration Files via SNMP

snmpset -v 2c -c public -m ./f10-copy-config.mib force10system-ip-address copySrcFileType.index i 2 copyDestFileType.index i 3

Figure 56 show the command syntax using MIB object names, and Figure 57 shows the same command using the object OIDs. In both cases, the object is followed by a unique index number. Figure 37-6. Copying Configuration Files via SNMP using Object-Name Syntax
> snmpset -v 2c -r 0 -t 60 -c public -m ./f10-copy-config.mib 10.10.10.10 copySrcFileType.101 i 2 copyDestFileType.101 i 3 FORCE10-COPY-CONFIG-MIB::copySrcFileType.101 = INTEGER: runningConfig(2) FORCE10-COPY-CONFIG-MIB::copyDestFileType.101 = INTEGER: startupConfig(3)

Figure 37-7. Copying Configuration Files via SNMP using OID Syntax
> snmpset -v 2c -c public -m ./f10-copy-config.mib 10.10.10.10 .1.3.6.1.4.1.6027.3.5.1.1.1.1.2.100 i 2 .1.3.6.1.4.1.6027.3.5.1.1.1.1.5.100 i 3 FORCE10-COPY-CONFIG-MIB::copySrcFileType.100 = INTEGER: runningConfig(2) FORCE10-COPY-CONFIG-MIB::copyDestFileType.100 = INTEGER: startupConfig(3)

Copy the startup-config to the running-config using the following command from a Unix machine: snmpset -c private -v 2c force10system-ip-address copySrcFileType.index i 3 copyDestFileType.index i 2 Figure 37-8. Copying Configuration Files via SNMP using Object-Name Syntax
> snmpset -c public -v 2c -m ./f10-copy-config.mib 10.11.131.162 copySrcFileType.7 i 3 copyDestFileType.7 i 2 FORCE10-COPY-CONFIG-MIB::copySrcFileType.7 = INTEGER: runningConfig(3) FORCE10-COPY-CONFIG-MIB::copyDestFileType.7 = INTEGER: startupConfig(2)

Figure 37-9. Copying Configuration Files via SNMP using OID Syntax
>snmpset -c public -v 2c 10.11.131.162 .1.3.6.1.4.1.6027.3.5.1.1.1.1.2.8 i 3 .1.3.6.1.4.1.6027.3.5.1.1.1.1.5.8 i 2 SNMPv2-SMI::enterprises.6027.3.5.1.1.1.1.2.8 = INTEGER: 3 SNMPv2-SMI::enterprises.6027.3.5.1.1.1.1.5.8 = INTEGER: 2

Copy the startup-config to the server via FTP using the following command from the Unix machine:
snmpset -v 2c -c public -m ./f10-copy-config.mib force10system-ip-address copySrcFileType.index i 2 copyDestFileName.index s filepath/filename copyDestFileLocation.index i 4 copyServerAddress.index a server-ip-address copyUserName.index s server-login-id copyUserPassword.index s server-login-password

Simple Network Management Protocol | 709

www.dell.com | support.dell.com

Table 37-4. Task

Copying Configuration Files via SNMP

server-ip-address must be preceded by the keyword a. values for copyUsername and copyUserPassword must be preceded by the keyword s.

Figure 37-10.

Copying Configuration Files via SNMP and FTP to a Remote Server

> snmpset -v 2c -c private -m ./f10-copy-config.mib 10.10.10.10 copySrcFileType.110 i 2 copyDestFileName.110 s /home/startup-config copyDestFileLocation.110 i 4 copyServerAddress.110 a 11.11.11.11 copyUserName.110 s mylogin copyUserPassword.110 s mypass FORCE10-COPY-CONFIG-MIB::copySrcFileType.110 = INTEGER: runningConfig(2) FORCE10-COPY-CONFIG-MIB::copyDestFileName.110 = STRING: /home/startup-config FORCE10-COPY-CONFIG-MIB::copyDestFileLocation.110 = INTEGER: ftp(4) FORCE10-COPY-CONFIG-MIB::copyServerAddress.110 = IpAddress: 11.11.11.11 FORCE10-COPY-CONFIG-MIB::copyUserName.110 = STRING: mylogin FORCE10-COPY-CONFIG-MIB::copyUserPassword.110 = STRING: mypass

Copy the startup-config to the server via TFTP using the following command from the Unix machine: Note: Verify that the file exists and its permissions are set to 777, and specify the relative path to the TFTP root directory.
snmpset -v 2c -c public -m ./f10-copy-config.mib force10system-ip-address copySrcFileType.index i 3 copyDestFileType.index i 1 copyDestFileName.index s filepath/filename copyDestFileLocation.index i 3 copyServerAddress.index a server-ip-address

Figure 37-11.

Copying Configuration Files via SNMP and TFTP to a Remote Server

.snmpset -v 2c -c private -m ./f10-copy-config.mib 10.10.10.10 copySrcFileType.4 i 3 copyDestFileType.4 i 1 copyDestFileLocation.4 i 3 copyDestFileName.4 s /home/myfilename copyServerAddress.4 a 11.11.11.11

Copy a binary file from the server to the startup-configuration on the Dell Networking system via FTP using the following command from the Unix server:
snmpset -v 2c -c public -m ./f10-copy-config.mib force10system-ip-address copySrcFileType.index i 1 copySrcFileLocation.index i 4 copySrcFileName.index s filepath/filename copyDestFileType.index i 3 copyServerAddress.index a server-ip-address copyUserName.index s server-login-id copyUserPassword.index s server-login-password

Figure 37-12.

Copying Configuration Files via SNMP and FTP from a Remote Server

> snmpset -v 2c -c private -m ./f10-copy-config.mib 10.10.10.10 copySrcFileType.10 i 1 copySrcFileLocation.10 i 4 copyDestFileType.10 i 3 copySrcFileName.10 s /home/myfilename copyServerAddress.10 a 172.16.1.56 copyUserName.10 s mylogin copyUserPassword.10 s mypass

710

Simple Network Management Protocol

Dell Networking provides additional MIB Objects to view copy statistics. These are provided in Table 8.
Table 37-5. MIB Object copyState MIB Objects for Copying Configuration Files via SNMP OID .1.3.6.1.4.1.6027.3.5.1.1.1.11 Values 1= running 2 = successful 3 = failed Time value Time value Description Specifies the state of the copy operation.

copyTimeStarted copyTimeCompleted copyFailCause

.1.3.6.1.4.1.6027.3.5.1.1.1.12 .1.3.6.1.4.1.6027.3.5.1.1.1.13 .1.3.6.1.4.1.6027.3.5.1.1.1.14

Specifies the point in the up-time clock that the copy operation started. Specifies the point in the up-time clock that the copy operation completed.

Specifies the reason the copy request 1 = bad file name 2 = copy in progress failed. 3 = disk full 4 = file exists 5 = file not found 6 = timeout 7 = unknown Row status Specifies the state of the copy operation. Uses CreateAndGo when you are performing the copy. The state is set to active when the copy is completed.

copyEntryRowStatus

.1.3.6.1.4.1.6027.3.5.1.1.1.15

To obtain a value for any of the MIB Objects in Table 8:

Step 1 Task Get a copy-config MIB object value.
snmpset -v 2c -c public -m ./f10-copy-config.mib force10system-ip-address [OID.index | mib-object.index

index is the index value used in the snmpset command used to complete the copy operation.

Note: You can use the entire OID rather than the object name. Use the form: OID.index, as shown in Figure 62.

Figure 61 and Figure 62 are examples of using the command snmpget to obtain a MIB object value. These examples assume that: the server OS is Unix you are using SNMP version 2c the community name is public, and the file f10-copy-config.mib is in the current directory.
Note: In Unix, enter the command snmpset for help using this command.

Simple Network Management Protocol | 711

www.dell.com | support.dell.com

Figure 61 shows the command syntax using MIB object names, and Figure 62 shows the same command using the object OIDs. In both cases, the object is followed by same index number used in the snmpset command.
Figure 37-13. Obtaining MIB Object Values for a Copy Operation using Object-name Syntax

> snmpget -v 2c -c private -m ./f10-copy-config.mib 10.11.131.140 copyTimeCompleted.110 FORCE10-COPY-CONFIG-MIB::copyTimeCompleted.110 = Timeticks: (1179831) 3:16:38.31

Figure 37-14.

Obtaining MIB Object Values for a Copy Operation using OID Syntax

> snmpget -v 2c -c private 10.11.131.140 .1.3.6.1.4.1.6027.3.5.1.1.1.1.13.110 SNMPv2-SMI::enterprises.6027.3.5.1.1.1.1.13.110 = Timeticks: (1179831) 3:16:38.31

Manage VLANs using SNMP

The qBridgeMIB managed objects in the Q-BRIDGE-MIB, defined in RFC 2674, enable you to use SNMP manage VLANs.

Create a VLAN
Use the dot1qVlanStaticRowStatus object to create a VLAN. The snmpset operation in Figure 37-15 creates VLAN 10 by specifying a value of 4 for instance 10 of the dot1qVlanStaticRowStatus object.
Figure 37-15. Creating a VLAN using SNMP

> snmpset -v2c -c mycommunity 123.45.6.78 .1.3.6.1.2.1.17.7.1.4.3.1.5.10 i 4 SNMPv2-SMI::mib-2.17.7.1.4.3.1.5.10 = INTEGER: 4

Assign a VLAN Alias

Write a character string to the dot1qVlanStaticName object to assign a name to a VLAN, as shown in Figure 37-16.

712

Simple Network Management Protocol

Figure 37-16.

Assign a VLAN Alias using SNMP

[Unix system output] > snmpset -v2c -c mycommunity 10.11.131.185 .1.3.6.1.2.1.17.7.1.4.3.1.1.1107787786 s "My VLAN" SNMPv2-SMI::mib-2.17.7.1.4.3.1.1.1107787786 = STRING: "My VLAN" [FTOS system output] FTOS#show int vlan 10 Vlan 10 is down, line protocol is down Vlan alias name is: My VLAN Address is 00:01:e8:cc:cc:ce, Current address is 00:01:e8:cc:cc:ce Interface index is 1107787786 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed auto ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 01:01:00 Queueing strategy: fifo

Display the Ports in a VLAN

FTOS identifies VLAN interfaces using an interface index number that is displayed in the output of the command show interface vlan, as shown in Figure 37-17.
Figure 37-17. Identifying the VLAN Interface Index Number

FTOS(conf)#do show interface vlan id 10 % Error: No such interface name. R5(conf)#do show interface vlan 10 Vlan 10 is down, line protocol is down Address is 00:01:e8:cc:cc:ce, Current address is 00:01:e8:cc:cc:ce Interface index is 1107787786 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed auto ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:12:42 Queueing strategy: fifo Time since last interface status change: 00:12:42

To display the ports in a VLAN, send an snmpget request for the object dot1qStaticEgressPorts using the interface index as the instance number, as shown for an S-Series in Figure 37-18.
Figure 37-18. Display the Ports in a VLAN in SNMP

> snmpget -v2c -c mycommunity 10.11.131.185 .1.3.6.1.2.1.17.7.1.4.3.1.2.1107787786 SNMPv2-SMI::mib-2.17.7.1.4.3.1.2.1107787786 = Hex-STRING: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Simple Network Management Protocol | 713

www.dell.com | support.dell.com

The table that the Dell Networking system sends in response to the snmpget request is a table that contains hexadecimal (hex) pairs, each pair representing a group of eight ports. On the E-Series, 12 hex pairs represents a line card. Twelve pairs accommodates the greatest currently available line card port density, 96 ports. On the C-Series, 28 hex pairs represents a line card. Twenty-eight pairs accommodates the greatest currently available line card port density, 28 ports per port-pipe, and any remaining hex pairs are unused. On the S-Series, 7 hex pairs represents a stack unit. Seven pairs accommodates the greatest number of ports available on an S-Series, 56 ports. The last stack unit is assigned 8 pairs; the eighth pair is unused.

The first hex pair, 00 in Figure 37-18, represents ports 1-7 in Stack Unit 0. The next pair to the right represents ports 8-15. To resolve the hex pair into a representation of the individual ports, convert the hex pair to binary. Consider the first hex pair 00, which resolves to 0000 0000 in binary: On the E-Series and C-Series each position in the 8-character string is for one port, starting with Port 0 at the left end of the string, and ending with Port 7 at the right end. A 0 indicates that the port is not a member of the VLAN; a 1 indicates VLAN membership. On the S-Series, each position in the 8-character string is for one port, starting with Port 1 at the left end of the string, and ending with Port 8 at the right end. A 0 indicates that the port is not a member of the VLAN; a 1 indicates VLAN membership.

Figure 37-18 shows the output for an S-Series. All hex pairs are 00, indicating that no ports are assigned to VLAN 10. In Figure 37-19, Port 0/2 is added to VLAN 10 as untagged. And the first hex pair changes from 00 to 04.
Figure 37-19. Displaying Ports in a VLAN using SNMP

[FTOS system output] R5(conf)#do show vlan id 10 Codes: Q: U x G NUM 10 * - Default VLAN, G - GVRP VLANs Untagged, T - Tagged Dot1x untagged, X - Dot1x tagged GVRP tagged, M - Vlan-stack Status Inactive Description Q Ports U Gi 0/2

[Unix system output] > snmpget -v2c -c mycommunity 10.11.131.185 .1.3.6.1.2.1.17.7.1.4.3.1.2.1107787786 SNMPv2-SMI::mib-2.17.7.1.4.3.1.2.1107787786 = Hex-STRING: 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

714

Simple Network Management Protocol

The value 40 is in the first set of 7 hex pairs, indicating that these ports are in Stack Unit 0. The hex value 40 is 0100 0000 in binary. As described above, the left-most position in the string represents Port 1. The next position from the left represents Port 2 and has a value of 1, indicating that Port 0/2 is in VLAN 10. The remaining positions are 0, so those ports are not in the VLAN. Note that the table contains none of the other information provided by the show vlan command, such as port speed or whether the ports are tagged or untagged.

Add Tagged and Untagged Ports to a VLAN

The value dot1qVlanStaticEgressPorts object is an array of all VLAN members. The dot1qVlanStaticUntaggedPorts object is an array of only untagged VLAN members. All VLAN members that are not in dot1qVlanStaticUntaggedPorts are tagged. To add a tagged port to a VLAN, write the port to the dot1qVlanStaticEgressPorts object, as shown in Figure 37-20. To add an untagged port to a VLAN, write the port to the dot1qVlanStaticEgressPorts and dot1qVlanStaticUntaggedPorts objects, as shown in Figure 37-21.
Note: Whether adding a tagged or untagged port, you must specify values for both dot1qVlanStaticEgressPorts and dot1qVlanStaticUntaggedPorts.

In Figure 37-20, Port 0/2 is added as an untagged member of VLAN 10.

Figure 37-20. Adding Untagged Ports to a VLAN using SNMP
00 00 00 00 00

>snmpset -v2c -c mycommunity 10.11.131.185 .1.3.6.1.2.1.17.7.1.4.3.1.2.1107787786 x "40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00" .1.3.6.1.2.1.17.7.1.4.3.1.4.1107787786 x "40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00" SNMPv2-SMI::mib-2.17.7.1.4.3.1.2.1107787786 = Hex-STRING: 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 SNMPv2-SMI::mib-2.17.7.1.4.3.1.4.1107787786 = Hex-STRING: 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

00

In Figure 37-21, Port 0/2 is added as a tagged member of VLAN 10.

Simple Network Management Protocol | 715

www.dell.com | support.dell.com

Figure 37-21.

Adding Tagged Ports to a VLAN using SNMP

00 00 00 00 00

>snmpset -v2c -c mycommunity 10.11.131.185 .1.3.6.1.2.1.17.7.1.4.3.1.2.1107787786 x "40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00" .1.3.6.1.2.1.17.7.1.4.3.1.4.1107787786 x "00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00" SNMPv2-SMI::mib-2.17.7.1.4.3.1.2.1107787786 = Hex-STRING: 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 SNMPv2-SMI::mib-2.17.7.1.4.3.1.4.1107787786 = Hex-STRING: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

00

Enable and Disable a Port using SNMP

Step 1 2 Task Create an SNMP community on the Dell Networking system. From the Dell Networking system, identify the interface index of the port for which you want to change the admin status. Or, from the management system, use the snmpwwalk command to identify the interface index. Command Syntax
snmp-server community
show interface

Command Mode CONFIGURATION EXEC Privilege

Enter the command snmpset to change the admin status using either the object descriptor or the OID. Choose integer 1 to change the admin status to Up, or 2 to change the admin status to Down. snmpset with descriptor: snmpset -v version -c community agent-ip ifAdminStatus.ifindex i {1 | 2} snmpset with OID: snmpset -v version -c community agent-ip .1.3.6.1.2.1.2.2.1.7.ifindex i {1 | 2}

Fetch Dynamic MAC Entries using SNMP

Table 37-6. MIB Object
dot1dTpFdbTable dot1qTpFdbTable dot3aCurAggFdbTable

MIB Objects for Fetching Dynamic MAC Entries in the Forwarding Database OID
.1.3.6.1.2.1.17.4.3 .1.3.6.1.2.1.17.7.1.2.2 .1.3.6.1.4.1.6027.3.2.1.1.5

Description
List the learned unicast MAC addresses on the default VLAN. List the learned unicast MAC addresses on non-default VLANs. List the learned MAC addresses of aggregated links (LAG).

MIB
Q-BRIDGE MIB

F10-LINK-AGGREGATION-MIB

In Figure 37-22, R1 has one dynamic MAC address, learned off of port GigabitEthernet 1/21, which a member of the default VLAN, VLAN 1. The SNMP walk returns the values for dot1dTpFdbAddress, dot1dTpFdbPort, and dot1dTpFdbStatus.

716

Simple Network Management Protocol

Each object is comprised an OID concatenated with an instance number. In the case of these objects, the instance number is the decimal equivalent of the MAC address; derive the instance number by converting each hex pair to its decimal equivalent. For example, the decimal equivalent of E8 is 232, and so the instance number for MAC address 00:01:e8:06:95:ac is .0.1.232.6.149.172. The value of dot1dTpFdbPort is the port number of the port off which the system learns the MAC address. In this case, of GigabitEthernet 1/21, the manager returns the integer 118. The maximum line card port density on the E-Series is 96 ports, and line card numbering begins with 0; GigabitEthernet 1/21 is the 21st port on Line Card 1, and 96 + 21 yields 118.
Figure 37-22. Fetching Dynamic MAC Addresses on the Default VLAN

---------------------------MAC Addresses on FTOS System----------------------R1_E600#show mac-address-table VlanId Mac Address Type Interface State 1 00:01:e8:06:95:ac Dynamic Gi 1/21 Active ------------------------------Query from Management Station------------------------------->snmpwalk -v 2c -c techpubs 10.11.131.162 .1.3.6.1.2.1.17.4.3.1 SNMPv2-SMI::mib-2.17.4.3.1.1.0.1.232.6.149.172 = Hex-STRING: 00 01 E8 06 95 AC SNMPv2-SMI::mib-2.17.4.3.1.2.0.1.232.6.149.172 = INTEGER: 118 SNMPv2-SMI::mib-2.17.4.3.1.3.0.1.232.6.149.172 = INTEGER: 3

In Figure 37-23, GigabitEthernet 1/21 is moved to VLAN 1000, a non-default VLAN. Use the objects dot1qTpFdbTable to fetch the MAC addresses learned on non-default VLANs. The instance number is the VLAN number concatenated with the decimal conversion of the MAC address.
Figure 37-23. Fetching Dynamic MAC Addresses on Non-default VLANs

-------------------------MAC Addresses on FTOS System--------------------------R1_E600#show mac-address-table VlanId Mac Address Type Interface State 1000 00:01:e8:06:95:ac Dynamic Gi 1/21 Active ------------------------------Query from Management Station------------------------------->snmpwalk -v 2c -c techpubs 10.11.131.162 .1.3.6.1.2.1.17.7.1.2.2.1 SNMPv2-SMI::mib-2.17.7.1.2.2.1.2.1000.0.1.232.6.149.172 = INTEGER: 118 SNMPv2-SMI::mib-2.17.7.1.2.2.1.3.1000.0.1.232.6.149.172 = INTEGER: 3

Use dot3aCurAggFdbTable to fetch the learned MAC address of a port-channel. The instance number is the decimal conversion of the MAC address concatenated with the port-channel number.
Figure 37-24. Fetching Dynamic MAC Addresses on the Default VLAN

-----------------------------MAC Addresses on FTOS System------------------------------R1_E600(conf)#do show mac-address-table VlanId Mac Address Type Interface State 1000 00:01:e8:06:95:ac Dynamic Po 1 Active ------------------------------Query from Management Station------------------------------->snmpwalk -v 2c -c techpubs 10.11.131.162 .1.3.6.1.4.1.6027.3.2.1.1.5 SNMPv2-SMI::enterprises.6027.3.2.1.1.5.1.1.1000.0.1.232.6.149.172.1 = INTEGER: 1000 SNMPv2-SMI::enterprises.6027.3.2.1.1.5.1.2.1000.0.1.232.6.149.172.1 = Hex-STRING: 00 01 E8 06 95 AC SNMPv2-SMI::enterprises.6027.3.2.1.1.5.1.3.1000.0.1.232.6.149.172.1 = INTEGER: 1 SNMPv2-SMI::enterprises.6027.3.2.1.1.5.1.4.1000.0.1.232.6.149.172.1 = INTEGER: 1

Simple Network Management Protocol | 717

www.dell.com | support.dell.com

Deriving Interface Indices

FTOS assigns an interface number to each (configured or unconfigured) physical and logical interface. Display the interface index number using the command show interface from EXEC Privilege mode, as shown in Figure 37-25.
Figure 37-25. Display the Interface Index Number

FTOS#show interface gig 1/21 GigabitEthernet 1/21 is up, line protocol is up Hardware is Force10Eth, address is 00:01:e8:0d:b7:4e Current address is 00:01:e8:0d:b7:4e Interface index is 72925242 [output omitted] FTOS#show linecard all | grep 1 1 online online E48TF E48TF 7.7.1.1

48

The interface index is a binary number with bits that indicate the slot number, port number, interface type, and card type of the interface. FTOS converts this binary index number to decimal, and displays it in the output of the show interface command.
Figure 37-26. Interface Index Binary Calculations

1 bit

1 bit

5 bits

7 bits

4 bits

14 bits

Unused P/L Flag Slot Number

Port Number

Interface Type

Card Type

Starting from the least significant bit (LSB): the first 14 bits represent the card type the next 4 bits represent the interface type the next 7 bits represent the port number the next 5 bits represent the slot number the next 1 bit is 0 for a physical interface and 1 for a logical interface the next 1 bit is unused

For example, the index 72925242 is 100010110001100000000111010 in binary. The binary interface index for GigabitEthernet 1/21 of a 48-port 10/100/1000Base-T line card with RJ-45 interface is shown in Figure 37-27. Notice that the physical/logical bit and the final, unused bit are not given. The interface is physical, so this must be represented by a 0 bit, and the unused bit is always 0. These two bits are not given because they are the most significant bits, and leading zeros are often omitted.

718

Simple Network Management Protocol

Figure 37-27.
2 bits

Binary Representation of Interface Index

7 bits 4 bits 14 bits

10

0010110

0011 Interface Type

00000000111010 Card Type

Slot Port Number Number

For interface indexing, slot and port numbering begins with the binary one. If the Dell Networking system begins slot and port numbering from 0, then the binary 1 represents slot and port 0. For example, the index number in Figure 37-27 gives the binary 2 for the slot number, though interface GigabitEthernet 1/21 belongs to Slot 1. This is because the port for this example is on an E-Series which begins numbering slots from 0. You must subtract 1 from the slot number 2, which yields 1, the correct slot number for interface 1/21. Note that the interface index does not change if the interface reloads or fails over. On the S-Series, if the unit is renumbered (for any reason) the interface index will change during a reload.

Viewing the System Image

Use the chSysSwInPartitionAImgVers object to view the system image on Flash Partition A or use the chSysSwInPartitionBImgVers object to view the system image on Flash Partition B.
Table 37-7. MIB Object MIB Objects for Viewing the System Image on Flash Partitions OID Description MIB

chSysSwInPartitionAImgVers 1.3.6.1.4.1.6027.3.10.1.2.8.1.11 chSysSwInPartitionBImgVers 1.3.6.1.4.1.6027.3.10.1.2.8.1.12

List the version string of the Chassis MIB system image in Flash Partition A List the version string of the Chassis MIB system image in Flash Partition B.

The system image can also be retrieved by performing an SNMP walk on the following OID:
MIB Object chSysSwModuleTable OID 1.3.6.1.4.1.6027.3.10.1.2.8

Simple Network Management Protocol | 719

www.dell.com | support.dell.com

Viewing Alarm LED Status Using SNMP

The following table specifies the alarm notification settings (minor, major, and off). An example of a minor alarm event would be a rise in temperature above the minimum threshold. An example of a major alarm event would be a rise in temperature above the maximum threshold.
Table 37-8. MIB Object chStackUnitAlarmLEDStatus MIB Objects for Viewing System Alarm LED Status OID 1.3.6.1.4.1.6027.3.10.1.2.2.1.35 Description LED status of system alarms on unit. Values: 1 = off 2 = minor 3 = major MIB Chassis MIB

The system alarm LED status can be retrieved using an SNMP query as shown in the following output example:
Figure 37-28. SNMP Alarm Query

>snmpwalk -v2c -c techpubs 10.11.131.162 .1.3.6.1.4.1.6027.3.10.1.2.2.1.35 SNMPv2-SMI::enterprises.6027.3.10.1.2.2.1.35.1 = INTEGER: 1

720

Simple Network Management Protocol

Using SNMP for Entity MIB Queries

The Entity MIB can be used for SNMP queries such as snmpget, snmpgetnext, and snmpwalk. Refer to the following table for OIDs and variables. To verify the results of the SNMP query, use the show inventory command for stack ID, chassis, and optional module details or the show inventory media command for information on base or optional modules, such as SFP or SFP+.
Table 37-9. MIB Object entity entityMIBObjects entityPhysical entPhysicalTable entPhysicalEntry entPhysicalIndex entPhysicalDescr entPhysicalVendorType entPhysicalContainedIn entPhysicalClass entPhysicalParentRelPos entPhysicalName entPhysicalHardwareRev entPhysicalFirmwareRev entPhysicalSoftwareRev entPhysicalSerialNum entPhysicalMfgName entPhysicalModelName entPhysicalAlias entPhysicalAssetID entPhysicalIsFRU entPhysicalMfgDate entPhysicalUris entityMapping entPhysicalContainsTable entPhysicalContainsEntry entPhysicalChildIndex entityGeneral entLastChangeTime entityMIBTraps MIB Objects for Entity MIB Queries OID 1.3.6.1.2.1.47 1.3.6.1.2.1.47.1 1.3.6.1.2.1.47.1.1 1.3.6.1.2.1.47.1.1.1 1.3.6.1.2.1.47.1.1.1.1 1.3.6.1.2.1.47.1.1.1.1.1 1.3.6.1.2.1.47.1.1.1.1.2 1.3.6.1.2.1.47.1.1.1.1.3 1.3.6.1.2.1.47.1.1.1.1.4 1.3.6.1.2.1.47.1.1.1.1.5 1.3.6.1.2.1.47.1.1.1.1.6 1.3.6.1.2.1.47.1.1.1.1.7 1.3.6.1.2.1.47.1.1.1.1.8 1.3.6.1.2.1.47.1.1.1.1.9 1.3.6.1.2.1.47.1.1.1.1.10 1.3.6.1.2.1.47.1.1.1.1.11 1.3.6.1.2.1.47.1.1.1.1.12 1.3.6.1.2.1.47.1.1.1.1.13 1.3.6.1.2.1.47.1.1.1.1.14 1.3.6.1.2.1.47.1.1.1.1.14 1.3.6.1.2.1.47.1.1.1.1.16 1.3.6.1.2.1.47.1.1.1.1.17 1.3.6.1.2.1.47.1.1.1.1.18 1.3.6.1.2.1.47.1.3 1.3.6.1.2.1.47.1.3.3 1.3.6.1.2.1.47.1.3.3.1 1.3.6.1.2.1.47.1.3.3.1.1 1.3.6.1.2.1.47.1.4 1.3.6.1.2.1.47.1.4.1 1.3.6.1.2.1.47.2 Variable NODE NODE NODE NODE NODE LEAF PhysicalIndex LEAF SnmpAdminString LEAF AutonomousType LEAF INTEGER LEAF PhysicalClass LEAF INTEGER LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF SnmpAdminString LEAF TruthValue LEAF DateAndTime LEAF OCTET STRING NODE NODE NODE LEAF PhysicalIndex NODE LEAF TimeStamp NODE MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB Entity MIB

Simple Network Management Protocol | 721

www.dell.com | support.dell.com

Table 37-9. MIB Object

MIB Objects for Entity MIB Queries OID 1.3.6.1.2.1.47.2.0 1.3.6.1.2.1.47.2.0.1 Variable NODE NODE MIB Entity MIB Entity MIB

entityMIBTrapPrefix entConfigChange

722

Simple Network Management Protocol

38
Spanning Tree Protocol
Spanning Tree Protocol is supported on platforms:

ces

Protocol Overview
Spanning Tree Protocol (STP) is a Layer 2 protocolspecified by IEEE 802.1dthat eliminates loops in a bridged topology by enabling only a single path through the network. By eliminating loops, the protocol improves scalability in a large network and enables you to implement redundant paths, which can be activated upon the failure of active paths. Layer 2 loops, which can occur in a network due to poor network design and without enabling protocols like xSTP, can cause unnecessarily high switch CPU utilization and memory consumption. FTOS supports three other variations of Spanning Tree, as shown here:
Table 38-1. FTOS Supported Spanning Tree Protocols IEEE Specification 802.1d 802.1w 802.1s Third Party

Dell Networking Term Spanning Tree Protocol Rapid Spanning Tree Protocol Multiple Spanning Tree Protocol Per-VLAN Spanning Tree Plus

Configuring Spanning Tree

Configuring Spanning Tree is a two-step process: 1. Configure interfaces for Layer 2. See page 49. 2. Enable Spanning Tree Protocol. See page 726.

Related Configuration Tasks

Adding an Interface to the Spanning Tree Group Removing an Interface from the Spanning Tree Group

Spanning Tree Protocol | 723

www.dell.com | support.dell.com

Modifying Global Parameters Modifying Interface STP Parameters Enabling PortFast Preventing Network Disruptions with BPDU Guard STP Root Selection SNMP Traps for Root Elections and Topology Changes Configuring Spanning Trees as Hitless

Important Points to Remember

In FTOS software version 8.3.3.9, Spanning Tree Protocol (STP) is enabled by default on all 1G ports. In all other software versions, Spanning Tree Protocol (STP) is disabled by default. FTOS supports only one Spanning Tree instance (0). For multiple instances, you must enable MSTP, or PVST+. You may only enable one flavor of Spanning Tree at any one time. All ports in VLANs and all enabled interfaces in Layer 2 mode are automatically added to the Spanning Tree topology at the time you enable the protocol. To add interfaces to the Spanning Tree topology after STP is enabled, enable the port and configure it for Layer 2 using the command switchport. The IEEE Standard 802.1D allows eight bits for port ID and eight bits for priority. However, the eight bits for port ID provide port IDs for only 256 ports and the C-Series can contain 336 ports. To accommodate the increased number of ports, FTOS uses four bits from priority field in the port ID field.This implementation affects the Bridge MIB (RFC 1493), and you must interpret objects such as dot1dStpPortDesignatedPort object by using the first four bits as the priority and the last 12 bits as the port ID.

724

Spanning Tree Protocol

Configuring Interfaces for Layer 2 Mode

All interfaces on all switches that will participate in Spanning Tree must be in Layer 2 mode and enabled.
Figure 38-1. Example of Configuring Interfaces for Layer 2 Mode

R1
R1(conf)# int range gi 1/1 - 4 R1(conf-if-gi-1/1-4)# switchport R1(conf-if-gi-1/1-4)# no shutdown R1(conf-if-gi-1/1-4)#show config ! interface GigabitEthernet 1/1 no ip address switchport no shutdown ! interface GigabitEthernet 1/2 no ip address switchport no shutdown ! interface GigabitEthernet 1/3 no ip address switchport no shutdown ! interface GigabitEthernet 1/4 no ip address switchport no shutdown

R2
1/3 1/4 2/1 2/2 2/3 2/4

1/1

1/2

3/1

3/2 3/3 3/4

R3

To configure the interfaces for Layer 2 and then enable them:

Step 1 2 3 Task If the interface has been assigned an IP address, remove it. Place the interface in Layer 2 mode. Enable the interface. Command Syntax
no ip address

Command Mode INTERFACE INTERFACE INTERFACE

switchport no shutdown

Verify that an interface is in Layer 2 mode and enabled using the show config command from INTERFACE mode.
Figure 38-2. Verifying Layer 2 Configuration
FTOS(conf-if-gi-1/1)#show config ! interface GigabitEthernet 1/1 no ip address switchport Indicates no shutdown FTOS(conf-if-gi-1/1)#

that the interface is in Layer 2 mode

Spanning Tree Protocol | 725

www.dell.com | support.dell.com

Enabling Spanning Tree Protocol Globally

Spanning Tree Protocol must be enabled globally; it is not enabled by default. To enable Spanning Tree globally for all Layer 2 interfaces:
Step 1 2 Task Enter the PROTOCOL SPANNING TREE mode. Enable Spanning Tree. Command Syntax
protocol spanning-tree 0 no disable

Command Mode CONFIGURATION PROTOCOL SPANNING TREE

Note: To disable STP globally for all Layer 2 interfaces, enter the disable command from PROTOCOL SPANNING TREE mode.

Verify that Spanning Tree is enabled using the show config command from PROTOCOL SPANNING TREE mode.
Figure 38-3. Verifying STP is Enabled
FTOS(conf)#protocol spanning-tree 0 FTOS(config-span)#show config ! protocol spanning-tree 0 no disable Indicates FTOS#

that Spanning Tree is enabled

When you enable Spanning Tree, all physical, VLAN, and port-channel interfaces that are enabled and in Layer 2 mode are automatically part of the Spanning Tree topology. Only one path from any bridge to any other bridge participating in STP is enabled. Bridges block a redundant path by disabling one of the link ports.

726

Spanning Tree Protocol

Figure 38-4. Spanning Tree Enabled Globally

root R1
1/3 1/4 1/1 1/2
Forwarding

R2
2/1 2/2 2/3 2/4
Blocking

3/1

3/2 3/3 3/4

R3

Port 290 (GigabitEthernet 2/4) is Blocking Port path cost 4, Port priority 8, Port Identifier 8.290 Designated root has priority 32768, address 0001.e80d.2462 Designated bridge has priority 32768, address 0001.e80d.2462 Designated port id is 8.497, designated path cost 0 Timers: message age 1, forward delay 0, hold 0 Number of transitions to forwarding state 1 BPDU: sent 21, received 486 The port is not in the portfast mode

View the Spanning Tree configuration and the interfaces that are participating in STP using the show spanning-tree 0 command from EXEC privilege mode. If a physical interface is part of a port channel, only the port channel is listed in the command output.
Figure 38-5. show spanning-tree 0 Command Example
R2#show spanning-tree 0 Executing IEEE compatible Spanning Tree Protocol Bridge Identifier has priority 32768, address 0001.e826.ddb7 Configured hello time 2, max age 20, forward delay 15 Current root has priority 32768, address 0001.e80d.2462 Root Port is 289 (GigabitEthernet 2/1), cost of root path is 4 Topology change flag not set, detected flag not set Number of topology changes 3 last change occurred 0:16:11 ago from GigabitEthernet 2/3 Timers: hold 1, topology change 35 hello 2, max age 20, forward delay 15 Times: hello 0, topology change 0, notification 0, aging Normal Port 289 (GigabitEthernet 2/1) is Forwarding Port path cost 4, Port priority 8, Port Identifier 8.289 Designated root has priority 32768, address 0001.e80d.2462 Designated bridge has priority 32768, address 0001.e80d.2462 Designated port id is 8.496, designated path cost 0 Timers: message age 1, forward delay 0, hold 0 Number of transitions to forwarding state 1 BPDU: sent 21, received 486 The port is not in the portfast mode Port 290 (GigabitEthernet 2/2) is Blocking Port path cost 4, Port priority 8, Port Identifier 8.290 --More--

Spanning Tree Protocol | 727

www.dell.com | support.dell.com

Confirm that a port is participating in Spanning Tree using the show spanning-tree 0 brief command from EXEC privilege mode.
Figure 38-6. show spanning-tree brief Command Example
FTOS#show spanning-tree 0 brief Executing IEEE compatible Spanning Tree Protocol Root ID Priority 32768, Address 0001.e80d.2462 We are the root of the spanning tree Root Bridge hello time 2, max age 20, forward delay 15 Bridge ID Priority 32768, Address 0001.e80d.2462 Configured hello time 2, max age 20, forward delay 15 Interface Designated Name PortID Prio Cost Sts Cost Bridge ID -------------- ------ ---- ---- --- --------------------Gi 1/1 8.496 8 4 DIS 0 32768 0001.e80d.2462 Gi 1/2 8.497 8 4 DIS 0 32768 0001.e80d.2462 Gi 1/3 8.513 8 4 FWD 0 32768 0001.e80d.2462 Gi 1/4 8.514 8 4 FWD 0 32768 0001.e80d.2462 FTOS#

PortID -----8.496 8.497 8.513 8.514

Adding an Interface to the Spanning Tree Group

To add a Layer 2 interface to the Spanning Tree topology:
Task Enable Spanning Tree on a Layer 2 interface. Command Syntax
spanning-tree 0

Command Mode INTERFACE

Removing an Interface from the Spanning Tree Group

To remove a Layer 2 interface from the Spanning Tree topology:
Task Disable Spanning Tree on a Layer 2 interface. Command Syntax
no spanning-tree 0

Command Mode INTERFACE

In FTOS versions prior to 7.6.1.0, the command no spanning tree disables Spanning Tree on the interface, however, BPDUs are still forwarded to the RPM, where they are dropped. Beginning in FTOS version 7.6.1.0, the command no spanning tree disables Spanning Tree on the interface, and incoming BPDUs are dropped at the line card instead of at the RPM, which frees processing resources. This behavior is called Layer 2 BPDU filtering and is available for STP, RSTP, PVST+, and MSTP.

728

Spanning Tree Protocol

Modifying Global Parameters

You can modify Spanning Tree parameters. The root bridge sets the values for forward-delay, hello-time, and max-age and overwrites the values set on other bridges participating in Spanning Tree.
Note: Dell Networking recommends that only experienced network administrators change the Spanning Tree parameters. Poorly planned modification of the Spanning Tree parameters can negatively impact network performance.

Table 38-2 displays the default values for Spanning Tree.

Table 38-2. STP Default Values Default Value 15 seconds 2 seconds 20 seconds 100-Mb/s Ethernet interfaces 1-Gigabit Ethernet interfaces 10-Gigabit Ethernet interfaces Port Channel with 100 Mb/s Ethernet interfaces Port Channel with 1-Gigabit Ethernet interfaces Port Channel with 10-Gigabit Ethernet interfaces Port Priority 19 4 2 18 3 1 8

STP Parameter Forward Delay Hello Time Max Age Port Cost

To change STP global parameters:

Task Change the forward-delay parameter (the wait time before the interface enters the forwarding state). Range: 4 to 30 Default: 15 seconds Change the hello-time parameter (the BPDU transmission interval). Note: With large configurations (especially those with more ports) Dell Networking recommends that you increase the hello-time. Range: 1 to 10 Default: 2 seconds Change the max-age parameter (the refresh interval for configuration information that is generated by recomputing the Spanning Tree topology). Range: 6 to 40 Default: 20 seconds Command Syntax
forward-delay seconds

Command Mode PROTOCOL SPANNING TREE

hello-time seconds

PROTOCOL SPANNING TREE

max-age seconds

PROTOCOL SPANNING TREE

Spanning Tree Protocol | 729

www.dell.com | support.dell.com

View the current values for global parameters using the show spanning-tree 0 command from EXEC privilege mode. See Figure 38-5.

Modifying Interface STP Parameters

You can set the port cost and port priority values of interfaces in Layer 2 mode. Port cost is a value that is based on the interface type. The greater the port cost, the less likely the port will be selected to be a forwarding port. Port priority influences the likelihood that a port will be selected to be a forwarding port in case that several ports have the same port cost.

The default values are listed in Table 38-2. To change the port cost or priority of an interface:
Task Change the port cost of an interface. Range: 0 to 65535 Default: see Table 38-2. Change the port priority of an interface. Range: 0 to 15 Default: 8 Command Syntax
spanning-tree 0 cost cost

Command Mode INTERFACE

spanning-tree 0 priority priority-value

INTERFACE

View the current values for interface parameters using the show spanning-tree 0 command from EXEC privilege mode. See Figure 38-5.

Enabling PortFast
The PortFast feature enables interfaces to begin forwarding traffic approximately 30 seconds sooner. Interfaces forward frames by default until they receive a BPDU that indicates that they should behave otherwise; they do not go through the Learning and Listening states. The bpduguard shutdown-on-violation option causes the interface hardware to be shutdown when it receives a BPDU. When only bpduguard is implemented, although the interface is placed in an Error Disabled state when receiving the BPDU, the physical interface remains up and spanning-tree will drop packets in the hardware after a BPDU violation. BPDUs are dropped in the software after receiving the BPDU violation. Caution: Enable PortFast only on links connecting to an end station. PortFast can cause loops if it is enabled on an
interface connected to a network.

730

Spanning Tree Protocol

To enable PortFast on an interface:

Task Enable PortFast on an interface. Command Syntax
spanning-tree stp-id portfast [bpduguard | [shutdown-on-violation]]

Command Mode INTERFACE

Verify that PortFast is enabled on a port using the show spanning-tree command from the EXEC privilege mode or the show config command from INTERFACE mode; Dell Networking recommends using the show config command, as shown in Figure 38-7.
Figure 38-7. PortFast Enabled on Interface
FTOS#(conf-if-gi-1/1)#show conf ! interface GigabitEthernet 1/1 no ip address switchport spanning-tree 0 portfast no shutdown FTOS#(conf-if-gi-1/1)#

Indicates that the interface is in PortFast mode

Preventing Network Disruptions with BPDU Guard

The Portfast (and Edgeport, in the case of RSTP, PVST+, and MSTP) feature should be configured on ports that connect to end stations. End stations do not generate BPDUs, so ports configured with Portfast/ Edgport (edgeports) do not expect to receive BDPUs. If an edgeport does receive a BPDU, it likely means that it is connected to another part of the network, which can negatively effect the STP topology. The BPDU Guard feature blocks an edgeport upon receiving a BPDU to prevent network disruptions, and FTOS displays Message 1. Enable BPDU Guard using the option bpduguard when enabling PortFast or EdgePort. The bpduguard shutdown-on-violation option causes the interface hardware to be shutdown when it receives a BPDU. Otherwise, although the interface is placed in an Error Disabled state when receiving the BPDU, the physical interface remains up and spanning-tree will only drop packets after a BPDU violation. Figure 38-8 shows a scenario in which an edgeport might unintentionally receive a BPDU. The port on the Dell Networking system is configured with Portfast. If the switch is connected to the hub, the BPDUs that the switch generates might trigger an undesirable topology change. If BPDU Guard is enabled, when the edge port receives the BPDU, the BPDU will be dropped, the port will be blocked, and a console message will be generated.
Message 1 BPDU Guard Error
3w3d0h: %RPM0-P:RP2 %SPANMGR-5-BPDU_GUARD_RX_ERROR: Received Spanning Tree BPDU on BPDU guard port. Disable GigabitEthernet 3/41.

Spanning Tree Protocol | 731

www.dell.com | support.dell.com

Note: Note that unless the shutdown-on-violation option is enabled, spanning-tree only drops packets after a BPDU violation; the physical interface remains up, as shown below.
FTOS(conf-if-gi-0/7)#do show spanning-tree rstp brief Executing IEEE compatible Spanning Tree Protocol Root ID Priority 32768, Address 0001.e805.fb07 Root Bridge hello time 2, max age 20, forward delay 15 Bridge ID Priority 32768, Address 0001.e85d.0e90 Configured hello time 2, max age 20, forward delay 15 Interface Name ---------Gi 0/6 Gi 0/7 Designated Cost Bridge ID PortID ------- -------------------- -------20000 32768 0001.e805.fb07 128.653 20000 32768 0001.e85d.0e90 128.264

PortID -------128.263 128.264

Prio ---128 128

Cost ------20000 20000

Sts --FWD EDS

Interface Name Role PortID Prio Cost Sts Cost ---------- ------ -------- ---- ------- --- ------Gi 0/6 Root 128.263 128 20000 FWD 20000 Gi 0/7 ErrDis 128.264 128 20000 EDS 20000 FTOS(conf-if-gi-0/7)#do show ip int br gi 0/7 Interface IP-Address OK Method GigabitEthernet 0/7 unassigned YES Manual

Link-type --------P2P P2P Status up

Edge ---No No Protocol up

FTOS Behavior: Regarding bpduguard shutdown-on-violation behavior: 1If the interface to be shutdown is a port channel then all the member ports are disabled in the hardware. 2When a physical port is added to a port channel already in error disable state, the new member port will also be disabled in the hardware. 3When a physical port is removed from a port channel in error disable state, the error disabled state is cleared on this physical port (the physical port will be enabled in the hardware). 4The reset linecard command does not clear the error disabled state of the port or the hardware disabled state. The interface continues to be disables in the hardware. The error disabled state can be cleared with any of the following methods: Perform an shutdown command on the interface. Disable the shutdown-on-violation command on the interface ( no spanning-tree stp-id portfast [bpduguard | [shutdown-on-violation]] ). Disable spanning tree on the interface (no spanning-tree in INTERFACE mode). Disabling global spanning tree (no spanning-tree in CONFIGURATION mode).

732

Spanning Tree Protocol

Figure 38-8. Enabling BPDU Guard

FTOS(conf-if-gi-3/41)# spanning-tree 0 portfast bpduguard shutdown-on-violation FTOS(conf-if-gi-3/41)#show config ! interface GigabitEthernet 3/41 no ip address switchport spanning-tree 0 portfast bpduguard shutdown-on-violation no shutdown 3/41 Hub

Switch with Spanning Tree Enabled

FTOS Behavior: BPDU Guard and BPDU filtering (see Removing an Interface from the Spanning Tree Group) both block BPDUs, but are two separate features. BPDU Guard: is used on edgeports and blocks all traffic on edgeport if it receives a BPDU drops the BPDU after it reaches the RPM and generates a console message BPDU Filtering: disables Spanning Tree on an interface drops all BPDUs at the line card without generating a console message

STP Root Selection

The Spanning Tree Protocol determines the root bridge, but you can assign one bridge a lower priority to increase the likelihood that it will be selected as the root bridge. You can also specify that a bridge is the root or the secondary root. To change the bridge priority or specify that a bridge is the root or secondary root:
Task Assign a number as the bridge priority or designate it as the root or secondary root. priority-value range: 0 to 65535. The lower the number assigned, the more likely this bridge will become the root bridge. The default is 32768. The primary option specifies a bridge priority of 8192. The secondary option specifies a bridge priority of 16384. Command Syntax
bridge-priority {priority-value | primary | secondary}

Command Mode PROTOCOL SPANNING TREE

Spanning Tree Protocol | 733

www.dell.com | support.dell.com

View only the root information using the show spanning-tree root command (see Figure 38-9) from EXEC privilege mode.
Figure 38-9. show spanning-tree root Command Example
FTOS#show spanning-tree 0 root Root ID Priority 32768, Address 0001.e80d.2462 We are the root of the spanning tree Root Bridge hello time 2, max age 20, forward delay 15 FTOS#

SNMP Traps for Root Elections and Topology Changes

Enable SNMP traps for Spanning Tree state changes using the command snmp-server enable traps stp. Enable SNMP traps for RSTP, MSTP, and PVST+ collectively using the command snmp-server enable traps xstp.

Configuring Spanning Trees as Hitless

Configuring Spanning Trees as Hitless is supported only on platforms:

ce

You can configure Spanning Tree (STP), Rapid Spanning Tree (RSTP), Multiple Spanning Tree (MSTP), and Per-Vlan Spanning Tree (PVST+) to be hitless (all or none must be configured as hitless). When configured as hitless, critical protocol state information is synchronized between RPMs so that RPM failover is seamless, and no topology change is triggered. Configure all Spanning Tree types to be hitless using the command redundancy protocol xstp from CONFIGURATION mode, as shown in Figure 38-10.
Figure 38-10. Configuring all Spanning Tree Types to be Hitless

FTOS(conf)#redundancy protocol xstp FTOS#show running-config redundancy ! redundancy protocol xstp FTOS#

734

Spanning Tree Protocol

39
Stacking S-Series Switches
Stacking S-Series Switches is supported on platform

Using the FTOS stacking feature, multiple S-Series switch units can be interconnected with stacking interfaces. The stack becomes manageable as a single switch through the stack management unit. This chapter contains the following sections: S-Series Stacking Overview Important Points to Remember S-Series Stacking Configuration Tasks

S-Series Stacking Overview

The S60 supports up to 12 stack members with FTOS version 8.3.3.4. A stack is analogous to an E-Series or C-Series system with redundant RPMs and multiple line cards. FTOS elects a primary and secondary management unit, and all other units are member units. The forwarding database resides on the primary, and all other units maintain a sychnronized local copy. Each unit in the stack makes forwarding decisions based on their local copy. FTOS presents all of the units like line cards; for example, to access GigabitEthernet Port 1 on Stack Unit 0, enter interface gigabitethernet 0/1 from CONFIGURATION mode.

High Availability on S-Series Stacks

S-Series stacks have primary and secondary management units analogous to Dell Networking Route Processor Modules (Message 39-1). The management units synchronize the running configuration and protocol states so that the system fails over in the event of a hardware or software fault on the primary. In such an event, or when the primary is removed, the secondary unit becomes the stack manager and FTOS elects a new secondary. FTOS resets the failed management unit, and once online, it becomes a member unit; the remaining members remain online.

Stacking S-Series Switches | 735

www.dell.com | support.dell.com

Figure 39-1. S-Series Stack Manager Redundancy

Stack#show redundancy -- Stack-unit Status ------------------------------------------------Mgmt ID: 0 Stack-unit ID: 1 Stack-unit Redundancy Role: Primary Stack-unit State: Active Stack-unit SW Version: 7.8.1.0 Link to Peer: Up -- PEER Stack-unit Status ------------------------------------------------Stack-unit State: Standby Peer stack-unit ID: 2 Stack-unit SW Version: 7.8.1.0 -- Stack-unit Redundancy Configuration ------------------------------------------------Primary Stack-unit: mgmt-id 0 Auto Data Sync: Full Failover Type: Hot Failover Auto reboot Stack-unit: Enabled Auto failover limit: 3 times in 60 minutes -- Stack-unit Failover Record ------------------------------------------------Failover Count: 0 Last failover timestamp: None Last failover Reason: None Last failover type: None -- Last Data Block Sync Record: ------------------------------------------------Stack Unit Config: succeeded Mar 24 2009 20:35:14 Start-up Config: failed Mar 24 2009 20:35:14 Runtime Event Log: succeeded Mar 24 2009 20:35:14 Running Config: succeeded Mar 24 2009 20:35:14 ACL Mgr: succeeded Mar 24 2009 20:35:14

Management Unit Selection on S-Series Stacks

FTOS has a selection algorithm to decide which stack units will be the primary and secondary management units. During the bootup of a single unit or the stack, FTOS compares the priority values of all of the units in the stack and elects the unit with the numerically highest priority the primary, and the next highest priority the secondary. For example, if you add a powered standalone unit to a stack, either the standalone unit or the stack reloads (excluding the new unit), depending on which has the higher priority, the new unit or the existing stack manager. If the new unit has the higher priority, it becomes the new stack manager after the stack reloads. All switches have a default priority of 0; if a priority tie occurs, the system with the highest MAC address supersedes, as shown in Figure 39-2.

736

Stacking S-Series Switches

Figure 39-2. Electing the Stack Manager

Stack>show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present Stack#show system stack-unit 0 | grep priority Master priority : 0 Stack#show system stack-unit 1 | grep priority Master priority : 0 Stack#show system stack-unit 2 | grep priority Master priority : 0 Stack#show system stack-unit 0| grep "Burned In MAC" Burned In MAC : 00:01:e8:d5:ef:81 Stack#show system stack-unit 0 | grep "Burned In MAC" Burned In MAC : 00:01:e8:d5:ef:81 Stack#show system stack-unit 1 | grep "Burned In MAC" Burned In MAC : 00:01:e8:d5:f9:6f Stack#show system stack-unit 2 | grep "Burned In MAC" Burned In MAC : 00:01:e8:cc:cc:cc

MAC Addressing on S-Series Stacks

The S-Series has three MAC addressees: the chassis MAC, interface MAC, and null interface MAC. All interfaces in the stack use the interface MAC address of the management unit (stack manager), and the chassis MAC for the stack is the masters chassis MAC. The stack continues to use the masters chassis MAC address even after a failover. The MAC address is not refreshed until the stack is reloaded and a different unit becomes the stack manager.
Note: If the removed management unit is brought up as a standalone unit or as part of a different stack, there is a possibility of MAC address collisions.

In Figure 39-3 and Figure 39-4, a standalone is added to a stack. The standalone and the stack master have the same priority, but the standalone has a lower MAC address, so the standalone reboots. In Figure 39-4 and Figure 39-5, a standalone is added to a stack. The standalone has a higher priority than the stack, so the stack (excluding the new unit) reloads.

Stacking S-Series Switches | 737

www.dell.com | support.dell.com

Figure 39-3. Adding a Standalone with a Lower MAC Address to a Stack Before
-------------------------------STANDALONE BEFORE CONNECTION---------------------------------Standalone#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Management online S50V S50V 7.8.1.0 52 1 Member not present 2 Member not present 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] Standalone#show system | grep priority Master priority : 0 ------------------------------------STACK BEFORE CONNECTION---------------------------------Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member not present 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] Stack#show system stack-unit 0 | grep priority Master priority : 0 Stack#show system stack-unit 1 | grep priority Master priority : 0

738

Stacking S-Series Switches

Figure 39-4. Adding a Standalone with a Lower MAC Address and Equal Priority to a StackAfter
-------------------------------STANDALONE AFTER CONNECTION---------------------------------Standalone#%STKUNIT0-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit is present 00:20:20: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 1 present 00:20:22: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present Going for reboot. Reason is Stack merge [bootup messages omitted] ------------------------------------STACK AFTER CONNECTION---------------------------------Stack# 3w1d14h: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present 3w1d14h: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 2 down - card removed 3w1d14h: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present 3w1d14h: %STKUNIT1-M:CP %CHMGR-5-CHECKIN: Checkin from Stack unit 2 (type S50V, 52 ports) 3w1d14h: %S50V:2 %CHMGR-0-PS_UP: Power supply 0 in unit 2 is up 3w1d14h: %STKUNIT1-M:CP %CHMGR-5-STACKUNITUP: Stack unit 2 is up Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

Stacking S-Series Switches | 739

www.dell.com | support.dell.com

Figure 39-5. Adding a Standalone with a Lower MAC Address but Higher Priority to a Stack Before
-------------------------------STANDALONE BEFORE CONNECTION---------------------------------Standalone#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present S50V 1 Member not present S50N 2 Management online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] Stack#show system | grep priority Master priority : 1 ------------------------------------STACK BEFORE CONNECTION---------------------------------Stack##show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member not present 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present Stack#show system stack-unit 0 | grep priority Master priority : 0 Stack#show system stack-unit 1 | grep priority Master priority : 0

740

Stacking S-Series Switches

Figure 39-6.

Adding a Standalone with a Lower MAC Address but Higher Priority to a StackAfter

-------------------------------STANDALONE AFTER CONNECTION---------------------------------Standalone#00:18:27: %STKUNIT2-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 00:18:27: %STKUNIT2-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 1 present 00:18:40: %STKUNIT2-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 0 down - card removed 00:18:40: %STKUNIT2-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 1 down - card removed 00:19:30: %STKUNIT2-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit i s present 00:19:30: %STKUNIT2-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 00:19:30: %STKUNIT2-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 1 present [remaining bootup messages omitted] ------------------------------------STACK AFTER CONNECTION---------------------------------Stack#3w1d15h: %STKUNIT1-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit is not present 3w1d15h: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present Going for reboot. Reason is Stack merge 3w1d15h: %STKUNIT1-M:CP %CHMGR-2-STACK_UNIT_DOWN: Stack-unit 0 down - card removed [bootup messages omitted] Stack#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member online S50V S50V 7.8.1.0 52 1 Standby online S50N S50N 7.8.1.0 52 2 Management online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

Management Access on S-Series Stacks

You can access the stack via the console port or VTY line. Console access: You may access the stack through the console port of the stack manager only. Like a standby RPM, the console port of the standby unit does not provide management capability; only a limited number of commands are available. Member units provide a severely limited set of commands, as shown in Figure 39-7. Remote access: You may access the stack with SNMP, SSH, or Telnet through any enabled, Layer 3 interface on any stack unit. There is no dedicated management port or management route table.

Stacking S-Series Switches | 741

www.dell.com | support.dell.com

Figure 39-7. Accessing Non-Master Units on a Stack via the Console Port
-------------------------------CONSOLE ACCESS ON THE STANDBY---------------------------------Stack(standby)>? disable Turn off privileged commands enable Turn on privileged commands exit Exit from the EXEC show Show running system information ssh-peer-stack-unit Open a SSH connection to the peer Stack-unit telnet-peer-stack-unit Open a telnet connection to the peer Stack-unit terminal Set terminal line parameters Stack(standby)>show ? calendar Display the hardware calendar clock Display the system clock command-history CLI command history redundancy Current Stack unit HA status version Software version -------------------------------CONSOLE ACCESS ON A MEMBER------------------------------------Stack(stack-member-0)#? reset-self Reset this unit alone show Show running system information

Important Points to Remember

You may stack up to eight S25/S50 systems in combination. You may stack up to 12 S60 systems You may stack up to 3 S4810 systems You may not stack different S-Series systems together (except the S25/S50) You may not connect 12G and 24G stack ports. All stack units must have the same version of FTOS.

S-Series Stacking Installation Tasks

Create an S-Series Stack Add a Unit to an S-Series Stack Remove a Unit from an S-Series Stack Merge Two S-Series Stacks Split an S-Series Stack

Create an S-Series Stack

Stacking modules are pluggable units in the back of the unit that switch traffic between units in a stack. Units are connected using bi-directional stacking cables; if you stacking modules have two ports, it does not matter if you connect port A to B, or A to A, or B to B. Install stacking modules before powering the unit. If you install a stacking module while the unit is online, FTOS does not register the new hardware; in this case, you must reload the unit.

742

Stacking S-Series Switches

You may stack any combination of S-Series models that have the same FTOS version. The S60 can be stacked in the following topologies: ring, cascade (also called daisy-chain) and braided ring. A braided ring topology can be used in a stack with four or more switches. It provides redundant paths while minimizing switch interconnect cable distances. In this topology, each switch has a directly attached data-path that is one to two peers away. This allows each switch member to have two active links within the stack. Figure 39-8 shows two common stacking topologies, ring and cascade (also called daisy-chain). A ring topology provides some performance gains and stack integrity.
Figure 39-8. Common S-Series Stacking Topologies

Ring Connection

Cascade Connection

A A A

B B B

A A A A

B B B B

Stacking 001

Facing the rear of an S-Series unit, stack-port are numbered from left to right, beginning with the highest Ethernet port number (n) plus 1. For example, for a 48-port unit with two 12-Gigabyte stacking modules, the stack-ports are 49, 50, 51, and 52, starting from the left. To add a unit to an existing stack:
Step 1 2 3 4 5 Task Verify that each unit has the same FTOS version prior to stacking them together. Pre-configure unit numbers for each unit so that the stacking is deterministic upon boot up. Configure the switch priority for each unit to make management unit selection deterministic. Connect the units using stacking cables. Power the stack one unit at a time. Start with the management unit, then the standby, followed by each of the members in order of their assigned stack number (or the position in the stack you want each unit to take). Allow each unit to completely boot, and verify that the unit is detected by the stack manager, and then power the next unit.
show system brief

Command Syntax
show version

Command Mode EXEC Privilege EXEC Privilege CONFIGURATION

stack-unit renumber

stack-unit priority

EXEC Privilege

Stacking S-Series Switches | 743

www.dell.com | support.dell.com

To display the status of the stacking ports, including the topology:

Task Display the stacking ports. Command Syntax
show system stack-ports

Command Mode EXEC Privilege

Figure 39-9 shows an example of a daisy-chain topology. Figure 39-10 shows the same stack converted to a ring by connecting stack-port 2/51 to 0/51; you may rearrange the stacking cables without triggering a unit reset, so long as the stack manager is never disconnected from the stack.
Figure 39-9. Displaying the S-Series Stacking Topology
Stack#show system stack-ports Topology: Daisy chain Interface Connection Link Speed Admin Link Trunk (Gb/s) Status Status Group -----------------------------------------------------------------0/51 12 up down 0/52 1/50 12 up up 1/49 2/52 12 up up 1/50 0/52 12 up up 2/51 12 up down 2/52 1/49 12 up up

Figure 39-10.

Displaying the S-Series Stacking Topology

FTOS#show system stack-ports Topology: Ring Interface Connection Link Speed Admin Link Trunk (Gb/s) Status Status Group -----------------------------------------------------------------0/51 2/51 12 up up 0/52 1/50 12 up up 1/49 2/52 12 up up 1/50 0/52 12 up up 2/51 0/51 12 up up 2/52 1/49 12 up up

Stacking Cable Redundancy

You can connect two units with two stacking cables as shown in, in case of a stacking port, module, or cable failure. Removal of only one of the cables does not trigger a reset.

744

Stacking S-Series Switches

Figure 39-11.

Stacking Cable Redundancy

A A A

B B B Stacking Cable Redundancy

Stacking 002

LED Status Indicators on an S-Series Stack

The stack unit is displayed in an LED panel on the front of each switch. Each panel displays the stack unit ID from 0 through 7 and: A for the management unit Bfor the standby management unit 0 for stack members

Add a Unit to an S-Series Stack

If you are adding units to a stack, you can either: allow FTOS to automatically assign the new unit a position in the stack, or manually determine each units position in the stack by configuring each unit to correspond with the stack before connecting it

Three configurable system variables affect how a new unit joins a stack: priority, stack number, and provision. Depending on which has the higher priority, either the standalone unit or the entire stack reloads (excluding the new unit). If the new unit has the higher priority, it becomes the new stack manager and the stack reloads, as shown in Figure 39-3, Figure 39-4, Figure 39-5, and Figure 39-6. If you add a unit that has a stack number that conflicts with the stack, the stack assigns the first available stack number, as shown in Figure 39-12 and Figure 39-13. If the stack has a provision for the stack-number that will be assigned to the new unit, the provision must match the unit type, or FTOS generates a type mismatch error, as show in Figure 39-14 and Figure 39-15.

After the new unit loads, it synchronizes its running and startup configurations with the stack.

Stacking S-Series Switches | 745

www.dell.com | support.dell.com

To manually assign a new unit a position in the stack:

Step 1 2 Task Command Syntax Command Mode

While the unit is unpowered, install stacking modules in the new unit. On the stack, determine the next available stack-unit number, and the management prioritity of the management unit. Create a virtual unit and assign it the next available stack-unit number. On the new unit, number it the next available stack-unit number. (OPTIONAL) On the new unit, assign a management priority based on whether you want the new unit to be the stack manager. Connect the new unit to the stack using stacking cables. Figure 39-12. Adding a Stack Unit with a Conflicting Stack NumberBefore
show system brief

show system stack-unit

stack-unit provision

EXEC Privilege

3 4 5

CONFIGURATION EXEC Privilege CONFIGURATION

stack-unit renumber stack-unit priority

------------------------STANDALONE BEFORE CONNECTION---------------------------------Standalone#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present S50V 1 Management online S50V S50V 7.8.1.0 52 2 Member not present 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] -----------------------------STACK BEFORE CONNECTION---------------------------------Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted]

746

Stacking S-Series Switches

Figure 39-13.

Adding a Stack Unit with a Conflicting Stack NumberAfter

------------------------STANDALONE AFTER CONNECTION---------------------------------00:08:45: %STKUNIT1-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit is present 00:08:45: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 00:08:47: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present Going for reboot. Reason is Stack merge [bootup messages omitted] Stack(stack-member-0)# -----------------------------STACK AFTER CONNECTION---------------------------------Stack#21:27:22: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 21:27:39: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 0 down - card removed 21:28:24: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 21:28:33: %STKUNIT1-M:CP %CHMGR-5-CHECKIN: Checkin from Stack unit 0 (type S50V, 52 ports) 21:28:33: %S50V:0 %CHMGR-0-PS_UP: Power supply 0 in unit 0 is up 21:28:34: %STKUNIT1-M:CP %CHMGR-5-STACKUNITUP: Stack unit 0 is up Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted]

Figure 39-14.

Adding a Stack Unit with a Conflicting Stack ProvisionBefore

------------------------STANDALONE BEFORE CONNECTION---------------------------------Standalone#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Management online S50V S50V 7.8.1.0 52 1 Member not present S50N 2 Member not present S50V 3 Member not present S50V 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] -----------------------------STACK BEFORE CONNECTION---------------------------------Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present S25N 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted]

Stacking S-Series Switches | 747

www.dell.com | support.dell.com

Figure 39-15.

Adding a Stack Unit with a Conflicting Stack NumberAfter

------------------------STANDALONE AFTER CONNECTION---------------------------------01:38:34: %STKUNIT0-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit is present 01:38:34: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 1 present 01:38:34: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 presentGoing for reboot. Reason is Stack merge Going for reboot. Reason is Stack merge [bootup messages omitted] Stack(stack-member-0)# -----------------------------STACK AFTER CONNECTION---------------------------------23:11:25: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 23:11:40: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 0 down - card removed 23:12:25: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 23:12:34: %STKUNIT1-M:CP %CHMGR-5-CHECKIN: Checkin from Stack unit 0 (type S50V, 52 ports) 23:12:34: %STKUNIT1-M:CP %CHMGR-3-STACKUNIT_MISMATCH: Mismatch: Stack unit 0 is type S50V type S25N required Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member type mismatch S25N S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted]

Remove a Unit from an S-Series Stack

The running-configuration and startup-configuration are synchronized on all stack units. A stack member that is disconnected from the stack maintain this configuration. To remove a stack member from the stack, disconnect the stacking cables from the unit. You may do this at any time, whether the unit is powered or unpowered, online or offline. Note that if you remove a unit in the middle of the stack, the stack will be split into multiple parts, and each will form a new stack according to the stacking algorithm described throughout this chapter.

748

Stacking S-Series Switches

Figure 39-16.

Removing a Stack MemberBefore

----------------------------STANDALONE BEFORE DISCONNECTION---------------------------------Standalone(stack-member-2)#? reset-self Reset this unit alone show Show running system information Standalone(stack-member-2)#show ? version Software version ---------------------------------STACK BEFORE DISCONNECTION---------------------------------Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

Stacking S-Series Switches | 749

www.dell.com | support.dell.com

Figure 39-17.

Removing a Stack MemberAfter

----------------------------STANDALONE AFTER DISCONNECTION---------------------------------Standalone(stack-member-2)# Going for reboot. Reason is Stack split [bootup messages omitted] Stack#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present S50V 1 Member not present S50N 2 Management online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present [output omitted] ---------------------------------STACK AFTER DISCONNECTION---------------------------------Stack#3w1d15h: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 2 down - card removed 3w1d15h: %STKUNIT1-M:CP %IFMGR-1-DEL_PORT: Removed port: Gi 2/1-48 3w1d15h: %STKUNIT0-S:CP %IFMGR-1-DEL_PORT: Removed port: Gi 2/1-48 Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Standby online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Member not present S50V 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

Replace a Stack Member

To replace a non-functional stack member:
Step 1 2 3 Task Verify that each unit has the same FTOS version prior to stacking them together. Command Syntax
show version

Command Mode EXEC Privilege

Insert the replacement unit only without connecting the stacking cables before powering the unit. Assign the stack-unit number of the unit that was replaced to the replacement unit.
stack-unit renumber

EXEC Privilege

750

Stacking S-Series Switches

Step 4

Task The management unit priority is 0 by default. If you configure the priority of the new unit to 1, the stack will reload. To avoid this scenario, configure the priority of the management unit to the highest value (14). Note: Do not configure the priority of the replacement unit, as this will be transferred from the management unit. Reload the switch to confirm the stack-unit number is correct.

Command Syntax
stack-unit priority

Command Mode CONFIGURATION

5 6

reset stack-unit 0-11

EXEC Privilege

Connect the new unit to the stack using stacking cables. Note: You must use Telnet to connect to the management unit and copy the running configuration (write-mem or copy run start) onto the replacement unit as the startup configuration, then reload the system for the new settings to take effect. If the running configuration is not copied to the replacement unit, the replacement unit will not have a startup configuration if it is removed from the stack and reloaded.

Merge Two S-Series Stacks

You may merge two stacks while they are powered and online. To merge two stacks, connect one stack to the other using stacking cables. You may not connect 12G and 24G stack ports. FTOS selects a primary stack manager from the two existing mangers. FTOS resets all the units in the losing stack, and they all become stack members. If there is no unit numbering conflict, the stack members retain their previous unit numbers. Otherwise, the stack manager assigns new unit numbers, based on the order that they come online. The stack manager overwrites the startup and running config on the losing stack members with its own.

Split an S-Series Stack

To split a stack, unplug the desired stacking cables.You may do this at any time, whether the stack is powered or unpowered, and the units are online or offline. Each portion of the split stack retains the startup and running configuration of the original stack. For a parent stack that is split into two child stacks, A and B, each with multiple units: If one of the new stacks receives the primary and the secondary management units, it is unaffected by the split. If one of the new stacks receives only the primary management unit, that units remains the stack manager, and FTOS elects a new secondary management unit. If one of the new stacks receives only the secondary management unit, it becomes the primary management, and FTOS elects a new secondary management unit. If one of the new stacks receives neither the primary nor the secondary management unit, the stack is reset so that a new election can take place.

Stacking S-Series Switches | 751

www.dell.com | support.dell.com

S-Series Stacking Configuration Tasks

Assign Unit Numbers to Units in an S-Series Stack Create a Virtual Stack Unit on an S-Series Stack Display Information about an S-Series Stack Influence Management Unit Selection on an S-Series Stack Manage Redundancy on an S-Series Stack Reset a Unit on an S-Series Stack Recover from Stack Link Flaps

Assign Unit Numbers to Units in an S-Series Stack

Each unit in the stack has a stack number that is either assigned by you or FTOS. Units are numbered from 0 to 7. Stack numbers are stored in NVRAM and are preserved upon reload.
Task Assign a stack-number to a unit. Command Syntax
stack-unit renumber

Command Mode EXEC Privilege

Note: Renumbering the stack manager triggers a failover, as shown in Message 1. Message 1 Renumbering the Stack Manager
Renumbering master unit will reload the stack. Proceed to renumber [confirm yes/no]: yes

Create a Virtual Stack Unit on an S-Series Stack

Use virtual stack units to configure ports on the stack before adding a new unit, or to prevent FTOS from assigning a particular stack-number.
Task Create a virtual stack unit. Command Syntax
stack-unit provision

Command Mode CONFIGURATION

Display Information about an S-Series Stack

Task Display for stack-identity, status, and hardware information on every unit in a stack (Figure 39-18). Command Syntax
show system

Command Mode EXEC Privilege

752

Stacking S-Series Switches

Task Display most of the information in show system, but in a more convenient tabular form (Figure 39-19). Display the same information in show system, but only for the specified unit (Figure 39-19). Display topology and stack link status for the entire stack. The available options separate the show system stack-port output into topology information from link status information (Figure 39-19). Figure 39-18.

Command Syntax
show system brief

Command Mode EXEC Privilege

show system stack-unit

EXEC Privilege

show system stack-ports [status | topology]

EXEC Privilege

Displaying Information about an S-Series Stackshow system

FTOS#show system Stack MAC : 00:01:e8:d5:f9:6f -- Unit 0 -Unit Type Status Next Boot Required Type Current Type Master priority Hardware Rev Num Ports Up Time FTOS Version Jumbo Capable POE Capable Burned In MAC No Of MACs

: : : : : : : : : : : : : :

Member Unit online online S50V - 48-port E/FE/GE with POE (SB) S50V - 48-port E/FE/GE with POE (SB) 0 2.0 52 30 min, 7 sec 7.8.1.0 yes yes 00:01:e8:d5:ef:81 3

-- Module 0 -Status : not present -- Module 1 -Status Module Type Num Ports Hot Pluggable

: : : :

online S50-01-12G-2S 2 no

- 2-port 12G Stacking (SB)

-- Power Supplies -Unit Bay Status Type --------------------------------------------------------------------------0 0 up AC 0 1 absent -- Fan Status -Unit TrayStatus Speed Fan0 Fan1 Fan2 Fan3 Fan4 Fan5 -------------------------------------------------------------------------------0 up low up up up up up up

Stacking S-Series Switches | 753

www.dell.com | support.dell.com

Figure 39-19.

Displaying Information about an S-Series Stackshow system brief

FTOS#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member online S50V S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present -- Module Info -Unit Module No Status Module Type Ports --------------------------------------------------------------------------0 0 not present No Module 0 0 1 online S50-01-12G-2S 2 1 0 online S50-01-12G-2S 2 1 1 not present No Module 0 2 0 not present No Module 0 2 1 online S50-01-12G-2S 2 -- Power Supplies -Unit Bay Status Type --------------------------------------------------------------------------0 0 up AC 0 1 absent 1 0 absent 1 1 up DC 2 0 up AC 2 1 absent -- Fan Status -Unit TrayStatus Speed Fan0 Fan1 Fan2 Fan3 Fan4 Fan5 -------------------------------------------------------------------------------0 up low up up up up up up 1 up low up up up up up up 2 up low up up up up up up

Figure 39-20.

Displaying Information about an S-Series Stackshow system stack-ports

FTOS#show system stack-ports Topology: Daisy chain Interface Connection Link Speed Admin Link Trunk (Gb/s) Status Status Group -----------------------------------------------------------------0/51 12 up down 0/52 2/51 12 up up 1/49 2/52 12 up up 1/50 12 up down 2/51 0/52 12 up up 2/52 1/49 12 up up

754

Stacking S-Series Switches

Influence Management Unit Selection on an S-Series Stack

Stack Priority is the system variable that FTOS uses to determine which units in the stack will be the primary and secondary management units. If multiple units tie for highest priority, then the unit with the highest MAC address prevails. If management was determined by priority only, a change in management occurs when:
Task Influence the selection of the stack management units. The unit with the numerically highest priority is elected the primary management unit, and the unit with the second highest priority is the secondary management unit. Default: 0 Range: 1-14

you powered down, or offline the management unit, or a failover occurs you disconnect the management unit from the stack
Command Syntax
stack-unit priority

Command Mode CONFIGURATION

Manage Redundancy on an S-Series Stack

Task Reset the current management unit, and make the secondary management unit the new primary. A new secondary is elected, and when the former stack manager comes back online, it becomes a member unit. Prevent the stack manager from rebooting after a failover. This command does not affect a forced failover, manual reset, or a stack-link disconnect. Display redundancy information. Command Syntax
redundancy force-failover stack-unit

Command Mode EXEC Privilege

redundancy disable-auto-reboot stack-unit

CONFIGURATION

show redundancy

EXEC Privilege

Reset a Unit on an S-Series Stack

You may reset any stack unit except for the master (Message 2).
Message 2 Master Reset Disallowed
% Error: Reset of master unit is not allowed.

Task Reload a stack-unit

Command Syntax
reset stack-unit 0-7

Command Mode EXEC Privilege

Stacking S-Series Switches | 755

www.dell.com | support.dell.com

Task Reload a member unit, from the unit itself Reset a stack-unit when the unit is in a problem state.

Command Syntax
reset-self reset stack-unit 0-7 hard

Command Mode EXEC Privilege EXEC Privilege

Monitor an S-Series Stack with SNMP

S-Series supports the following tables in f10-ss-chassis.mib for stack management through SNMP: chStackUnitTable chSysStackPortTable

Troubleshoot an S-Series Stack

Recover from Stack Link Flaps Recover from a Card Problem State on an S-Series Stack Recover from a Card Mismatch State on an S-Series Stack

Recover from Stack Link Flaps

S-Series Stack Link Integrity Monitoring enables units to monitor their own stack ports, and disable any stack port that flaps five times within 10 seconds. FTOS displays console messages the local and remote members of a flapping link, and on the primary and secondary management units as KERN-2-INT messages if the flapping port belongs to either of these units. In Figure 39-21, a stack-port on the manager flaps. The remote member, Member 2, displays a console message, and the manager and standby display KERN-2-INT messages. To re-enable the downed stack-port, power cycle the offending unit.

756

Stacking S-Series Switches

Figure 39-21.

Recovering from a Stack Link Flapping Error

--------------------------------------MANAGMENT UNIT----------------------------------------Error: Stack Port 50 has flapped 5 times within 10 seconds.Shutting down this st ack port now. Error: Please check the stack cable/module and power-cycle the stack. 10:55:20: %STKUNIT1-M:CP %KERN-2-INT: Error: Stack Port 50 has flapped 5 times w ithin 10 seconds.Shutting down this stack port now. 10:55:20: %STKUNIT1-M:CP %KERN-2-INT: Error: Please check the stack cable/module and power-cycle the stack. ---------------------------------------STANDBY UNIT-----------------------------------------10:55:18: %STKUNIT1-M:CP %KERN-2-INT: Error: Stack Port 50 has flapped 5 times within 10 seonds.Shutting down this stack port now. 10:55:18: %STKUNIT1-M:CP %KERN-2-INT: Error: Please check the stack cable/module and power-cycle the stack. ----------------------------------------MEMBER 2--------------------------------------------Error: Stack Port 51 has flapped 5 times within 10 seconds.Shutting down this stack port now. Error: Please check the stack cable/module and power-cycle the stack.

Recover from a Card Problem State on an S-Series Stack

If a unit added to a stack has a different FTOS version, the unit does not come online, and FTOS cites a card problem error, as shown in Figure 39-22. To recover, disconnect the new unit from the stack, change the FTOS version to match the stack, and then reconnect it to the stack.
Figure 39-22. Recovering from a Card Problem Error on an S-Series Stack

F Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member card problem S25N unknown 7.7.1.1 52 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

Recover from a Card Mismatch State on an S-Series Stack

A card mismatch occurs if the stack has a provision for the lowest available stack number which does not match the model of a newly added unit (Figure 39-23). To recover, disconnect the new unit. Then, either: remove the provision from the stack, then reconnect the standalone unit, or renumber the standalone unit with another available stack number on the stack.

Stacking S-Series Switches | 757

www.dell.com | support.dell.com

Figure 39-23.

Recovering from a Card Mismatch State on an S-Series Stack

-----------------------------------STANDALONE UNIT BEFORE-----------------------------------Standalone#show system brief Stack MAC : 00:01:e8:d5:ef:81 -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Management online S50V S50V 7.8.1.0 52 1 Member not present S50N 2 Member not present S50V 3 Member not present S50V 4 Member not present 5 Member not present 6 Member not present 7 Member not present ---------------------------------------STACK BEFORE-----------------------------------------Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member not present S25N 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present -----------------------------------STANDALONE UNIT AFTER-----------------------------------01:38:34: %STKUNIT0-M:CP %POLLMGR-2-ALT_STACK_UNIT_STATE: Alternate Stack-unit is present 01:38:34: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 1 present Going for reboot. Reason is Stack merge 01:38:34: %STKUNIT0-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 2 present ----------------------------------------STACK AFTER-----------------------------------------23:11:25: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 23:11:40: %STKUNIT1-M:CP %CHMGR-2-STACKUNIT_DOWN: Stack unit 0 down - card removed 23:12:25: %STKUNIT1-M:CP %CHMGR-5-STACKUNITDETECTED: Stack unit 0 present 23:12:34: %STKUNIT1-M:CP %CHMGR-5-CHECKIN: Checkin from Stack unit 0 (type S50V, 52 ports) 23:12:34: %STKUNIT1-M:CP %CHMGR-3-STACKUNIT_MISMATCH: Mismatch: Stack unit 0 is type S50V type S25N required Stack#show system brief Stack MAC : 00:01:e8:d5:f9:6f -- Stack Info -Unit UnitType Status ReqTyp CurTyp Version Ports --------------------------------------------------------------------------0 Member type mismatch S25N S50V 7.8.1.0 52 1 Management online S50N S50N 7.8.1.0 52 2 Standby online S50V S50V 7.8.1.0 52 3 Member not present 4 Member not present 5 Member not present 6 Member not present 7 Member not present

758

Stacking S-Series Switches

40
Storm Control
ces Storm Control for Multicast is supported on platforms: c s
Storm Control is supported on platforms: The storm control feature enables you to control unknown-unicast and broadcast traffic on Layer 2 and Layer 3 physical interfaces.
FTOS Behavior: On the E-Series, FTOS supports broadcast control for Layer 3 traffic only. To control Layer 2 broadcast traffic use the command storm-control unknown-unicast. On the C-Series and S-Series, FTOS supports broadcast control (command storm-control broadcast ) for Layer 2 and Layer 3 traffic.

FTOS Behavior: On E-Series, bi-directional traffic (unknown unicast and broadcast) along with egress storm control causes the configured traffic rates to be split between the involved ports. The percentage of traffic that each port receives after the split is not predictable. These ports can be in the same/ different port-pipes or on the same/different line cards.

FTOS Behavior: The minimum number of packets per second (PPS) that storm control can limit on the S60 is 2.

Configure Storm Control

Storm control is supported in INTERFACE mode and CONFIGURATION mode

Configure storm control from INTERFACE mode

Configure storm control from INTERFACE mode using the command storm control. From INTERFACE mode: You can only on configure storm control for ingress traffic. If you configure storm control from both INTERFACE and CONFIGURATION mode, the INTERFACE mode configurations override the CONFIGURATION mode configurations.

Storm Control | 759

www.dell.com | support.dell.com

The percentage of storm control is calculated based on the advertised rate of the line card, not by the speed setting.

Configure storm control from CONFIGURATION mode

Configure storm control from CONFIGURATION mode using the command storm control. From CONFIGURATION mode you can configure storm control for ingress and egress traffic. Do not apply per-VLAN QoS on an interface that has storm-control enabled (either on an interface or globally) On the E-Series, when broadcast storm-control is enabled on an interface or globally on the ingress and DSCP marking for a DSCP value 1 is configured for the data traffic, the traffic goes to queue 1 instead of queue 0. Similarly, if unicast storm-control is enabled on an interface or globally on the ingress, and DSCP marking for a DSCP value 2 is configured for the data traffic, the traffic goes to queue 2 instead of queue 0.

760

Storm Control

41
System Time and Date
Chapter 41, System Time and Date settings, and Network Time Protocol are supported on platforms:

es

System times and dates can be set and maintained through the Network Time Protocol (NTP). They are also set through FTOS CLIs and hardware settings. This chapter includes the following sections: Network Time Protocol Protocol Overview Implementation Information Configuring Network Time Protocol FTOS Time and Date Configuring time and date settings Set daylight savings time

Network Time Protocol

Network Time Protocol (NTP) synchronizes timekeeping among a set of distributed time servers and clients. The protocol also coordinates time distribution in a large, diverse network with a variety of interfaces. In NTP, servers maintain the time and NTP clients synchronize with a time-serving host. NTP clients choose from among several NTP servers to determine which offers the best available source of time and the most reliable transmission of information. NTP is a fault-tolerant protocol that will automatically select the best of several available time sources to synchronize to. Multiple candidates can be combined to minimize the accumulated error. Temporarily or permanently insane time sources will be detected and avoided. Dell Networking recommends configuring NTP for the most accurate time. In FTOS, other time sources can be configured (the hardware clock and the software clock). NTP is designed to produce three products: clock offset, roundtrip delay, and dispersion, all of which are relative to a selected reference clock.

System Time and Date | 761

www.dell.com | support.dell.com

Clock offset represents the amount to adjust the local clock to bring it into correspondence with the reference clock. Roundtrip delay provides the capability to launch a message to arrive at the reference clock at a specified time. Dispersion represents the maximum error of the local clock relative to the reference clock.

Since most host time servers will synchronize via another peer time server, there are two components in each of these three products, those determined by the peer relative to the primary reference source of standard time and those measured by the host relative to the peer. Each of these components are maintained separately in the protocol in order to facilitate error control and management of the subnet itself. They provide not only precision measurements of offset and delay, but also definitive maximum error bounds, so that the user interface can determine not only the time, but the quality of the time as well. In what may be the most common client/server model a client sends an NTP message to one or more servers and processes the replies as received. The server interchanges addresses and ports, overwrites certain fields in the message, recalculates the checksum and returns the message immediately. Information included in the NTP message allows the client to determine the server time with respect to local time and adjust the local clock accordingly. In addition, the message includes information to calculate the expected timekeeping accuracy and reliability, as well as select the best from possibly several servers. Following conventions established by the telephone industry [BEL86], the accuracy of each server is defined by a number called the stratum, with the topmost level (primary servers) assigned as one and each level downwards (secondary servers) in the hierarchy assigned as one greater than the preceding level. FTOS synchronizes with a time-serving host to get the correct time. You can set FTOS to poll specific NTP time-serving hosts for the current time. From those time-serving hosts, the system chooses one NTP host with which to synchronize and serve as a client to the NTP host. As soon as a host-client relationship is established, the networking device propagates the time information throughout its local network.

Protocol Overview
NTP message to one or more servers and processes the replies as received. The server interchanges addresses and ports, fills in or overwrites certain fields in the message, recalculates the checksum and returns it immediately. Information included in the NTP message allows each client/server peer to determine the timekeeping characteristics of its other peers, including the expected accuracies of their clocks. Using this information each peer is able to select the best time from possibly several other clocks, update the local clock and estimate its accuracy.

762

System Time and Date

Figure 41-1.

NTP Fields
Source Port (123) Destination Port (123) Length Checksum NTP Packet Payload

Range: +32 to -32

Leap Indicator

Status

Type

Precision

Est. Error

Est. Drift Rate

Reference Clock ID

Reference Timestamp

Originate Timestamp

Recieve Timestamp

Transmit Timestamp

Code: 00: No Warning 01: +1 second 10: -1 second 11: reserved

Range: 0-4 Code: 0: clock operating correctly 1: carrier loss 2: synch loss 3: format error 4: interface/link failure

Range: 0-4 Code: 0: unspecified 1: primary reference clock 2: secondary reference clock via NTP 3: secondary reference via some other host/protocol 4: eyeball-and-wristwatch

Implementation Information
Dell Networking systems can only be an NTP client.

Configuring Network Time Protocol

Configuring NTP is a one-step process: 1. Enable NTP. See page 764.

Related Configuration Tasks

Configure NTP broadcasts Set the Hardware Clock with the Time Derived from NTP Set the Hardware Clock with the Time Derived from NTP Disable NTP on an interface Configure a source IP address for NTP packets (optional)

System Time and Date | 763

www.dell.com | support.dell.com

Enable NTP
NTP is disabled by default. To enable it, specify an NTP server to which the Dell Networking system will synchronize. Enter the command multiple times to specify multiple servers. You may specify an unlimited number of servers at the expense of CPU resources.
Task Specify the NTP server to which the Dell Networking system will synchronize. Command
ntp server ip-address

Command Mode CONFIGURATION

Display the system clock state with respect to NTP using the command show ntp status from EXEC Privilege mode, as shown in Figure 41-2.
Figure 41-2. Displaying the System Clock State with respect to NTP
R6_E300(conf)#do show ntp status Clock is synchronized, stratum 2, reference is 192.168.1.1 frequency is -369.623 ppm, stability is 53.319 ppm, precision is 4294967279 reference time is CD63BCC2.0CBBD000 (16:54:26.049 UTC Thu Mar 12 2009) clock offset is 997.529984 msec, root delay is 0.00098 sec root dispersion is 10.04271 sec, peer dispersion is 10032.715 msec peer mode is client

Display the calculated NTP synchronization variables received from the server that the system will use to synchronize its clock using the command show ntp associations from EXEC Privilege mode, as shown in Figure 41-3.
Figure 41-3. Displaying the Calculated NTP Synchronization Variables
R6_E300(conf)#do show ntp associations remote ref clock st when poll reach delay offset disp ========================================================================== #192.168.1.1 .LOCL. 1 16 16 76 0.98 -2.470 879.23 * master (synced), # master (unsynced), + selected, - candidate

764

System Time and Date

Set the Hardware Clock with the Time Derived from NTP
Task Periodically update the system hardware clock with the time value derived from NTP. Command
ntp update-calendar

Command Mode CONFIGURATION

Figure 41-4. Displaying the Calculated NTP Synchronization Variables

R5/R8(conf)#do show calendar 06:31:02 UTC Mon Mar 13 1989 R5/R8(conf)#ntp update-calendar 1 R5/R8(conf)#do show calendar 06:31:26 UTC Mon Mar 13 1989 R5/R8(conf)#do show calendar 12:24:11 UTC Thu Mar 12 2009

Configure NTP broadcasts

With FTOS, you can receive broadcasts of time information. You can set interfaces within the system to receive NTP information through broadcast. To configure an interface to receive NTP broadcasts, use the following commands in the INTERFACE mode:
Task Set the interface to receive NTP packets. Command
ntp broadcast client

Command INTERFACE

Table 41-1.
2w1d11h : NTP: Maximum Slew:-0.000470, Remainder = -0.496884

Disable NTP on an interface

By default, NTP is enabled on all active interfaces. If you disable NTP on an interface, FTOS drops any NTP packets sent to that interface.

System Time and Date | 765

www.dell.com | support.dell.com

To disable NTP on an interface, use the following command in the INTERFACE mode:
Command Syntax
ntp disable

Command Mode INTERFACE

Purpose Disable NTP on the interface.

To view whether NTP is configured on the interface, use the show config command in the INTERFACE mode. If ntp disable is not listed in the show config command output, then NTP is enabled. (The show config command displays only non-default configuration information.)

Configure a source IP address for NTP packets

By default, the source address of NTP packets is the IP address of the interface used to reach the network. You can configure one interfaces IP address to be included in all NTP packets. To configure an IP address as the source address of NTP packets, use the following command in the CONFIGURATION mode:
Command Syntax
ntp source interface

Command Mode CONFIGURATION

Purpose Enter the following keywords and slot/port or number information: For a 1-Gigabit Ethernet interface, enter the keyword GigabitEthernet followed by the slot/port information. For a loopback interface, enter the keyword loopback followed by a number between 0 and 16383. For a port channel interface, enter the keyword lag followed by a number from 1 to 255 for TeraScale and ExaScale. For a SONET interface, enter the keyword sonet followed by the slot/port information. For a 10-Gigabit Ethernet interface, enter the keyword TenGigabitEthernet followed by the slot/port information. For a VLAN interface, enter the keyword vlan followed by a number from 1 to 4094. E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

To view the configuration, use the show running-config ntp command (Figure 38) in the EXEC privilege mode.

766

System Time and Date

Configure NTP authentication

NTP authentication and the corresponding trusted key provide a reliable means of exchanging NTP packets with trusted time sources. NTP authentication begins when the first NTP packet is created following the configuration of keys. NTP authentication in FTOS uses the MD5 algorithm and the key is embedded in the synchronization packet that is sent to an NTP time source.
FTOS Behavior: FTOS versions 8.2.1.0 and later use an encryption algorithm to store the authentication key that is different from previous FTOS versions; beginning in version 8.2.1.0, FTOS uses DES encryption to store the key in the startup-config when you enter the command ntp authentication-key. Therefore, if your system boots with a startup-configuration from an FTOS versions prior to 8.2.1.0 in which you have configured ntp authentication-key, the system cannot correctly decrypt the key, and cannot authenticate NTP packets. In this case you must re-enter this command and save the running-config to the startup-config.

To configure NTP authentication, use these commands in the following sequence in the CONFIGURATION mode:
Step 1 Command Syntax
ntp authenticate ntp authentication-key number md5 key

Command Mode CONFIGURATION CONFIGURATION

Purpose Enable NTP authentication. Set an authentication key. Configure the following parameters: number: Range 1 to 4294967295. This number must be the same as the number in the ntp trusted-key command. key: Enter a text string. This text string is encrypted. Define a trusted key. Configure a number from 1 to 4294967295. The number must be the same as the number used in the ntp authentication-key command.

ntp trusted-key number

CONFIGURATION

To view the NTP configuration, use the show running-config ntp command (Figure 40) in the EXEC privilege mode. Figure 41-5 shows an encrypted authentication key. All keys are encrypted.
Figure 41-5. show running-config ntp Command Example
FTOS#show running ntp ! ntp authenticate ntp authentication-key 345 md5 5A60910F3D211F02 ntp server 11.1.1.1 version 3 ntp trusted-key 345 FTOS#

encrypted key

System Time and Date | 767

www.dell.com | support.dell.com

Command Syntax
ntp server ip-address [key keyid] [prefer] [version number]

Command Mode CONFIGURATION

Purpose Configure an NTP server. Configure the IP address of a server and the following optional parameters: key keyid: Configure a text string as the key exchanged between the NTP server and client. prefer: Enter the keyword to set this NTP server as the preferred server. version number: Enter a number 1 to 3 as the NTP version.

R6_E300(conf)#1w6d23h : NTP: xmit packet to 192.168.1.1: leap 0, mode 3, version 3, stratum 2, ppoll 1024 rtdel 0219 (8.193970), rtdsp AF928 (10973.266602), refid C0A80101 (192.168.1.1) ref CD7F4F63.6BE8F000 (14:51:15.421 UTC Thu Apr 2 2009) org CD7F4F63.68000000 (14:51:15.406 UTC Thu Apr 2 2009) rec CD7F4F63.6BE8F000 (14:51:15.421 UTC Thu Apr 2 2009) xmt CD7F5368.D0535000 (15:8:24.813 UTC Thu Apr 2 2009) 1w6d23h : NTP: rcv packet from 192.168.1.1 leap 0, mode 4, version 3, stratum 1, ppoll 1024 rtdel 0000 (0.000000), rtdsp AF587 (10959.090820), refid 4C4F434C (76.79.67.76) ref CD7E14FD.43F7CED9 (16:29:49.265 UTC Wed Apr 1 2009) org CD7F5368.D0535000 (15:8:24.813 UTC Thu Apr 2 2009) rec CD7F5368.D0000000 (15:8:24.812 UTC Thu Apr 2 2009) xmt CD7F5368.D0000000 (15:8:24.812 UTC Thu Apr 2 2009) inp CD7F5368.D1974000 (15:8:24.818 UTC Thu Apr 2 2009) rtdel-root delay rtdsp - round trip dispersion refid - reference id org rec - (last?) receive timestamp xmt - transmit timestamp mode - 3 client, 4 server stratum - 1 primary reference clock, 2 secondary reference clock (via NTP) version - NTP version 3 leap -

Leap Indicator (sys.leap, peer.leap, pkt.leap): This is a two-bit code warning of an impending leap second to be inserted in the NTP time scale. The bits are set before 23:59 on the day of insertion and reset after 00:00 on the following day. This causes the number of seconds (rollover interval) in the day of insertion to be increased or decreased by one. In the case of primary servers the bits are set by operator intervention, while in the case of secondary servers the bits are set by the protocol. The two bits, bit 0 and bit 1, respectively, are coded as follows: Poll Interval: integer indicating the minimum interval between transmitted messages, in seconds as a power of two. For instance, a value of six indicates a minimum interval of 64 seconds. Precision: integer indicating the precision of the various clocks, in seconds to the nearest power of two. The value must be rounded to the next larger power of two; for instance, a 50-Hz (20 ms) or 60-Hz (16.67ms) power-frequency clock would be assigned the value -5 (31.25 ms), while a 1000-Hz (1 ms) crystal-controlled clock would be assigned the value -9 (1.95 ms).

768

System Time and Date

Root Delay (sys.rootdelay, peer.rootdelay, pkt.rootdelay): This is a signed fixed-point number indicating the total roundtrip delay to the primary reference source at the root of the synchronization subnet, in seconds. Note that this variable can take on both positive and negative values, depending on clock precision and skew. Root Dispersion (sys.rootdispersion, peer.rootdispersion, pkt.rootdispersion): This is a signed fixed-point number indicating the maximum error relative to the primary reference source at the root of the synchronization subnet, in seconds. Only positive values greater than zero are possible. Reference Clock Identifier (sys.refid, peer.refid, pkt.refid): This is a 32-bit code identifying the particular reference clock. In the case of stratum 0 (unspecified) or stratum 1 (primary reference source), this is a four-octet, left-justified, zero-padded ASCII string, for example (see Appendix A for comprehensive list): the case of stratum 2 and greater (secondary reference) this is the four-octet Internet address of the peer selected for synchronization. Reference Timestamp (sys.reftime, peer.reftime, pkt.reftime): This is the local time, in timestamp format, when the local clock was last updated. If the local clock has never been synchronized, the value is zero. Originate Timestamp: The departure time on the server of its last NTP message. If the server becomes unreachable, the value is set to zero. Receive Timestamp: The arrival time on the client of the last NTP message from the server. If the server becomes unreachable, the value is set to zero. Transmit Timestamp: The departure time on the server of the current NTP message from the sender. Filter dispersion is the error in calculating the minimum delay from a set of sample data from a peer.

FTOS Time and Date

The time and date can be set using the FTOS CLI.

Configuring time and date settings

The following list includes the configuration tasks for setting the system time: Set the time and date for the switch hardware clock Set the time and date for the switch software clock Set the timezone Set daylight savings time Set Daylight Saving Time Once Set Recurring Daylight Saving Time

System Time and Date | 769

www.dell.com | support.dell.com

Set the time and date for the switch hardware clock
Command Syntax
calendar set time month day year

Command Mode EXEC Privilege

Purpose Set the hardware clock to the current time and date. time: Enter the time in hours:minutes:seconds. For the hour variable, use the 24-hour format, for example, 17:15:00 is 5:15 pm. month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year. day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year year: Enter a four-digit number as the year. Range: 1993 to 2035.

FTOS#calendar set 08:55:00 september 18 2009 FTOS#

Set the time and date for the switch software clock
You can change the order of the month and day parameters to enter the time and date as time day month year. You cannot delete the software clock.

770

System Time and Date

The software clock runs only when the software is up. The clock restarts, based on the hardware clock, when the switch reboots.
Command Syntax
clock set time month day year

Command Mode EXEC Privilege

Purpose Set the system software clock to the current time and date. time: Enter the time in hours:minutes:seconds. For the hour variable, use the 24-hour format, for example, 17:15:00 is 5:15 pm. month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year. day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year year: Enter a four-digit number as the year. Range: 1993 to 2035.

FTOS#clock set 16:20:00 19 september 2009 FTOS#

Set the timezone

Coordinated Universal Time (UTC) is the time standard based on the International Atomic Time standard, commonly known as Greenwich Mean time. When determining system time, you must include the differentiator between UTC and your local timezone. For example, San Jose, CA is the Pacific Timezone with a UTC offset of -8.
Command Syntax
clock timezone timezone-name offset

Command Mode CONFIGURATION

Purpose Set the clock to the appropriate timezone. timezone-name: Enter the name of the timezone. Do not use spaces. offset: Enter one of the following: a number from 1 to 23 as the number of hours in addition to UTC for the timezone. a minus sign (-) followed by a number from 1 to 23 as the number of hours

System Time and Date | 771

www.dell.com | support.dell.com

Command Syntax

Command Mode

Purpose

FTOS#conf FTOS(conf)#clock timezone Pacific -8 FTOS(conf)#01:40:19: %RPM0-P:CP %CLOCK-6-TIME CHANGE: Timezone configuration changed from "UTC 0 hrs 0 mins" to "Pacific -8 hrs 0 mins" FTOS#

Set daylight savings time

FTOS supports setting the system to daylight savings time once or on a recurring basis every year.

772

System Time and Date

Set Daylight Saving Time Once

Set a date (and time zone) on which to convert the switch to daylight savings time on a one-time basis.
Command Syntax
clock summer-time time-zone date

Command Mode CONFIGURATION

Purpose Set the clock to the appropriate timezone and daylight savings time.
time-zone: Enter the three-letter name for the time zone. This name is displayed in the show clock output.

start-month start-day start-year start-time end-month end-day end-year end-time [offset]

start-month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year start-day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year. start-year: Enter a four-digit number as the year. Range: 1993 to 2035 start-time: Enter the time in hours:minutes. For the hour variable, use the 24-hour format, example, 17:15 is 5:15 pm. end-month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year. end-day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year. end-year: Enter a four-digit number as the year. Range: 1993 to 2035. end-time: Enter the time in hours:minutes. For the hour variable, use the 24-hour format, example, 17:15 is 5:15 pm. offset: (OPTIONAL) Enter the number of minutes to add during the summer-time period. Range: 1 to1440. Default: 60 minutes

System Time and Date | 773

www.dell.com | support.dell.com

Command Syntax

Command Mode

Purpose

FTOS(conf)#clock summer-time pacific date Mar 14 2009 00:00 Nov 7 2009 00:00 FTOS(conf)#02:02:13: %RPM0-P:CP %CLOCK-6-TIME CHANGE: Summertime configuration changed from "none" to "Summer time starts 00:00:00 Pacific Sat Mar 14 2009;Summer time ends 00:00:00 pacific Sat Nov 7 2009" FTOS(conf)#

Set Recurring Daylight Saving Time

Set a date (and time zone) on which to convert the switch to daylight savings time on a specific day every year. If you have already set daylight savings for a one-time setting, you can set that date and time as the recurring setting with the clock summer-time time-zone recurring command.

Command Syntax
clock summer-time time-zone recurring

Command Mode CONFIGURATION

Purpose Set the clock to the appropriate timezone and adjust to daylight savings time every year.
time-zone: Enter the three-letter name for the time zone. This name is displayed in the show clock output.

start-week start-day start-month start-time end-week end-day end-month end-time [offset]

start-week: (OPTIONAL) Enter one of the following as the week that daylight savings begins and then enter values for start-day through end-time: week-number: Enter a number from 1-4 as the number of the week in the month to start daylight savings time. first: Enter this keyword to start daylight savings time in the first week of the month. last: Enter this keyword to start daylight savings time in the last week of the month. start-month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year start-day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year.

774

System Time and Date

Command Syntax

Command Mode

Purpose start-year: Enter a four-digit number as the year. Range: 1993 to 2035 start-time: Enter the time in hours:minutes. For the hour variable, use the 24-hour format, example, 17:15 is 5:15 pm. end-week: If you entered a start-week, Enter the one of the following as the week that daylight savings ends: week-number: enter a number from 1-4 as the number of the week to end daylight savings time. first: enter the keyword first to end daylight savings time in the first week of the month. last: enter the keyword last to end daylight savings time in the last week of the month. end-month: Enter the name of one of the 12 months in English. You can enter the name of a day to change the order of the display to time day month year. end-day: Enter the number of the day. Range: 1 to 31. You can enter the name of a month to change the order of the display to time day month year. end-year: Enter a four-digit number as the year. Range: 1993 to 2035. end-time: Enter the time in hours:minutes. For the hour variable, use the 24-hour format, example, 17:15 is 5:15 pm. offset: (OPTIONAL) Enter the number of minutes to add during the summer-time period. Range: 1 to1440. Default: 60 minutes

FTOS(conf)#clock summer-time pacific recurring Mar 14 2009 00:00 Nov 7 2009 00:00 ? FTOS(conf)#02:02:13: %RPM0-P:CP %CLOCK-6-TIME CHANGE: Summertime configuration changed from "none" to "Summer time starts 00:00:00 Pacific Sat Mar 14 2009;Summer time ends 00:00:00 pacific Sat Nov 7 2009" FTOS(conf)#

Note: If you enter <CR> after entering the recurring command parameter, and you have already set a one-time daylight saving time/date, the system will use that time and date as the recurring setting.

System Time and Date | 775

www.dell.com | support.dell.com

Command Syntax

Command Mode

Purpose

FTOS(conf)#clock summer-time pacific recurring ? <1-4> Week number to start first Week number to start last Week number to start <cr> FTOS(conf)#clock summer-time pacific recurring FTOS(conf)#02:10:57: %RPM0-P:CP %CLOCK-6-TIME CHANGE: Summertime configuration changed from "Summer time starts 00:00:00 Pacific Sat Mar 14 2009 ; Summer time ends 00:00:00 pacific Sat Nov 7 2009" to "Summer time starts 02:00:00 Pacific Sun Mar 8 2009;Summer time ends 02:00:00 pacific Sun Nov 1 2009"

776

System Time and Date

42
Uplink Failure Detection (UFD)
Uplink Failure Detection (UFD) is supported on the following platforms:

s (S50, S60 only), MXL,

Feature Description
Uplink Failure Detection (UFD) provides detection of the loss of upstream connectivity and, if used with NIC teaming, automatic recovery from a failed link. A switch provides upstream connectivity for devices, such as servers. If a switch loses its upstream connectivity, downstream devices also lose their connectivity. However, the devices do not receive a direct indication that upstream connectivity is lost since connectivity to the switch is still operational. UFD allows a switch to associate downstream interfaces with upstream interfaces. When upstream connectivity fails, the switch disables the downstream links. Failures on the downstream links allow downstream devices to recognize the loss of upstream connectivity. For example, in Figure 42-2 Switches S1 and S2 both have upstream connectivity to Router R1 and downstream connectivity to the server. UFD operation is shown in Steps A through C: In Step A, the server configuration uses the connection to S1 as the primary path. Network traffic flows from the server to S1 and then upstream to R1. In Step B, the upstream link between S1 and R1 fails. The server continues to use the link to S1 for its network traffic, but the traffic is not successfully switched through S1 because the upstream link is down. In Step C, UFD on S1 disables the link to the server. The server then stops using the link to S1 and switches to using its link to S2 to send traffic upstream to R1.

Uplink Failure Detection (UFD) | 777

www.dell.com | support.dell.com

Figure 42-1. Uplink Failure Detection

How Uplink Failure Detection Works

UFD creates an association between upstream and downstream interfaces. The association of uplink and downlink interfaces is called an uplink-state group. An interface in an uplink-state group can be a physical interface or a port-channel (LAG) aggregation of physical interfaces. An enabled uplink-state group tracks the state of all assigned upstream interfaces. Failure on an upstream interface results in the automatic disabling of downstream interfaces in the uplink-state group. As a result, downstream devices can execute the protection or recovery procedures they have in place to establish alternate connectivity paths as shown in Figure 42-2.

778

Uplink Failure Detection (UFD)

Figure 42-2. Uplink Failure Detection Example

If only one of the upstream interfaces in an uplink-state group goes down, a specified number of downstream ports associated with the upstream interface are put into a link-down state. This number is user-configurable and is calculated by the ratio of upstream port bandwidth to downstream port bandwidth in the same uplink-state group. This calculation ensures that there are no traffic drops due to insufficient bandwidth on the upstream links to the routers/switches. By default, if all upstream interfaces in an uplink-state group go down, all downstream interfaces in the same uplink-state group are put into a link-down state. Using UFD, you can configure the automatic recovery of downstream ports in an uplink-state group when the link status of an upstream port changes. The tracking of upstream link status does not have a major impact on CPU usage.

UFD and NIC Teaming

Uplink Failure Detection on a switch can be used with network adapter teaming on a server (see NIC Teaming) to implement a rapid failover solution. For example, in Figure 42-2 the switch/router with UFD detects the uplink failure and automatically disables the associated downstream link port to the server. The server with NIC teaming detects the disabled link and automatically switches over to the backup link in order to continue to transmit traffic upstream.

Uplink Failure Detection (UFD) | 779

www.dell.com | support.dell.com

Important Points to Remember

When you configure Uplink Failure Detection, the following conditions apply: You can configure up to sixteen uplink-state groups. By default, no uplink-state groups are created. An uplink-state group is considered to be operationally up if it has at least one upstream interface in the link-up state. An uplink-state group is considered to be operationally down if it has no upstream interfaces in the link-up state. No uplink-state tracking is performed when a group is disabled or in an operationally down state. You can assign physical port or port-channel interfaces to an uplink-state group. You can assign an interface to only one uplink-state group. Each interface assigned to an uplink-state group must be configured as either an upstream or downstream interface, but not both. You can assign individual member ports of a port channel to the group. An uplink-state group can contain either the member ports of a port channel or the port channel itself, but not both. If you assign a port channel as an upstream interface, the port channel interface enters a link-down state when the number of port-channel member interfaces in a link-up state drops below the configured Minimum Number of Members parameter. If one of the upstream interfaces in an uplink-state group goes down, either a user-configurable set of downstream ports or all the downstream ports in the group are put in an operationally down state with an UFD Disabled error. The order in which downstream ports are disabled is from the lowest numbered port to the highest. If one of the upstream interfaces in an uplink-state group that was down comes up, the set of UFD-disabled downstream ports (which were previously disabled due to this upstream port going down) are brought up and the UFD Disabled error is cleared. If an uplink-state group is disabled, the downstream interfaces are not disabled regardless of the state of the upstream interfaces. If an uplink-state group has no upstream interfaces assigned, downstream interfaces cannot be disabled when an upstream link goes down. To enable the debug messages for events related to a specified uplink-state group or all groups, enter the debug uplink-state-group [group-id] command, where group-id is 1 to 16. To turn off debugging event messages, enter the no debug uplink-state-group [group-id] command. For an example of debug log messages, see Message 1.

780

Uplink Failure Detection (UFD)

Configuring Uplink Failure Detection

To configure Uplink Failure Detection, follow these steps:
Step 1 Command Syntax and Mode uplink-state-group group-id Command Mode: CONFIGURATION Description Creates an uplink-state group and enabling the tracking of upstream links on the switch/router. Valid group-id values are 1 to 16. To delete an uplink-state group, enter the no uplink-state-group group-id command. 2
{upstream | downstream} interface

Assigns a port or port-channel to the uplink-state group as an upstream or downstream interface. For interface, enter one of the following interface types: Fast Ethernet: fastethernet {slot/port | slot/port-range} 1-Gigabit Ethernet: gigabitethernet {slot/port |slot/port-range} 10-Gigabit Ethernet: tengigabitethernet {slot/port |slot/port-range} Port channel: port-channel {1-512 | port-channel-range} Where port-range and port-channel-range specify a range of ports separated by a dash (-) and/or individual ports/port channels in any order; for example:
upstream gigabitethernet 1/1-2,5,9,11-12 downstream port-channel 1-3,5

Command Mode: UPLINK-STATE-GROUP

A comma is required to separate each port and port-range entry. To delete an interface from the group, enter the no {upstream | downstream} interface command. 3
downstream disable links {number | all}

Configures the number of downstream links in the uplink-state group that will be disabled (Oper Down state) if one upstream link in the group goes down. number specifies the number of downstream links to be brought down. Range: 1 to 1024.
all brings down all downstream links in the group.

Command Mode: UPLINK-STATE-GROUP

Default: No downstream links are disabled when an upstream link goes down. Note: Downstream interfaces in an uplink-state group are put into a link-down state with an UFD-Disabled error message only when all upstream interfaces in the group go down. To revert to the default setting, enter the no downstream disable links command. 4
downstream auto-recover

Command Mode: UPLINK-STATE-GROUP

(Optional) Enables auto-recovery so that UFD-disabled downstream ports in the uplink-state group come up when a disabled upstream port in the group comes back up. Default: Auto-recovery of UFD-disabled downstream ports is enabled. To disable auto-recovery, enter the no downstream auto-recover command.

Uplink Failure Detection (UFD) | 781

www.dell.com | support.dell.com

Step 5

Command Syntax and Mode

description text

Description (Optional) Enters a text description of the uplink-state group. Maximum length: 80 alphanumeric characters.

Command Mode: UPLINK-STATE-GROUP 6

no enable

(Optional) Disables upstream-link tracking without deleting the uplink-state group. Default: Upstream-link tracking is automatically enabled in an uplink-state group. To re-enable upstream-link tracking, enter the enable command.

Command Mode: UPLINK-STATE-GROUP

Clearing a UFD-Disabled Interface

You can manually bring up a downstream interface in an uplink-state group that has been disabled by UFD and is in a UFD-disabled error state. To re-enable one or more disabled downstream interfaces and clear the UFD-disabled error state, enter the following command:
Command Syntax
clear ufd-disable {interface interface | uplink-state-group group-id }

Description Re-enables a downstream interface on the switch/router that is in a UFD-disabled error state so that it can send and receive traffic. For interface, enter one of the following interface types: Fast Ethernet: fastethernet {slot/port | slot/port-range} 1-Gigabit Ethernet: gigabitethernet {slot/port | slot/port-range} 10-Gigabit Ethernet: tengigabitethernet {slot/port | slot/port-range} Port channel: port-channel {1-512 | port-channel-range} Where port-range and port-channel-rangee specify a range of ports separated by a dash (-) and/or individual ports/port channels in any order; for example:
gigabitethernet 1/1-2,5,9,11-12 port-channel 1-3,5

Command Mode: EXEC

A comma is required to separate each port and port-range entry.

clear ufd-disable {interface interface | uplink-state-group group-id}

re-enables all UFD-disabled downstream interfaces in the group. Range: 1 to 16.

782

Uplink Failure Detection (UFD)

Message 1 shows the Syslog messages displayed when you clear the UFD-disabled state from all disabled downstream interfaces in an uplink-state group by entering the clear ufd-disable uplink-state-group group-id command. All downstream interfaces return to an operationally up state.
Message 1 Syslog Messages before and after entering clear ufd-disable uplink-state-group Command (S50)
02:36:43: %RPM0-P:CP %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Te 0/ 46 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Downstream interface set 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Downstream interface set 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Downstream interface set 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Downstream interface set 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface 02:36:43: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to UFD to UFD to UFD to UFD state state state state to down: Te 0/46 error-disabled: Fo error-disabled: Fo error-disabled: Fo error-disabled: Fo to down: Fo 13/0 to down: Fo 13/1 to down: Fo 13/3 to down: Fo 13/5 13/0 13/1 13/3 13/5

02:37:29: %RPM0-P:CP %IFMGR-5-ASTATE_DN: Changed interface Admin state to down: Te 0/ 47 02:37:29: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Te 0/47 02:37:29 : UFD: Group:3, UplinkState: DOWN 02:37:29: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed uplink state group state to down: Group 3 02:37:29: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Downstream interface set to UFD error-disabled: Fo 13/6 02:37:29: %RPM0-P:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Fo 13/6

02:38:31 : UFD: Group:3, UplinkState: UP 02:38:31: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed uplink state group state to up: Group 3 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Downstream interface cleared from UFD error-disabled: Fo 13/0 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Downstream interface cleared from UFD error-disabled: Fo 13/1 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Downstream interface cleared from UFD error-disabled: Fo 13/3 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Downstream interface cleared from UFD error-disabled: Fo 13/5 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Downstream interface cleared from UFD error-disabled: Fo 13/6 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Fo 13/0 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Fo 13/1 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Fo 13/3 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Fo 13/5 02:38:53: %RPM0-P:CP %IFMGR-5-OSTATE_UP: Changed interface state to up: Fo 13/6

Uplink Failure Detection (UFD) | 783

www.dell.com | support.dell.com

Displaying Uplink Failure Detection

To display information on the Uplink Failure Detection feature, enter any of the following show commands:
Show Command Syntax show uplink-state-group [group-id] [detail] Command Mode: EXEC Description Displays status information on a specified uplink-state group or all groups. Valid group-id values are 1 to 16. detail displays additional status information on the upstream and downstream interfaces in each group (see Figure 42-3). show interfaces interface Command Mode: EXEC Displays the current status of a port or port-channel interface assigned to an uplink-state group. interface specifies one of the following interface types: Fast Ethernet: Enter fastethernet slot/port. 1-Gigabit Ethernet: Enter gigabitethernet slot/port. 10-Gigabit Ethernet: Enter tengigabitethernet slot/port. Port channel: Enter port-channel {1-512}. If a downstream interface in an uplink-state group has been disabled (Oper Down state) by uplink-state tracking because an upstream port went down, the message error-disabled[UFD] is displayed in the output (see Figure 42-4). show running-config uplink-state-group [group-id] Command Mode: EXEC Or show configuration Command Mode: UPLINK-STATE-GROUP Displays the current configuration of all uplink-state groups (Figure 42-5) or a specified group (Figure 42-6). Valid group-id values are 1 to 16.

784

Uplink Failure Detection (UFD)

Figure 42-3.

show uplink-state-group Command Output (S50)

FTOS# show uplink-state-group Uplink Uplink Uplink Uplink Uplink Uplink State State State State State State Group: Group: Group: Group: Group: Group: 1 3 5 6 7 16 Status: Status: Status: Status: Status: Status: Enabled, Up Enabled, Up Enabled, Down Enabled, Up Enabled, Up Disabled, Up

FTOS# show uplink-state-group 16 Uplink State Group: 16 Status: Disabled, Up

FTOS#show uplink-state-group detail (Up): Interface up (Dwn): Interface down Uplink State Group : 1 Upstream Interfaces : Downstream Interfaces :

(Dis): Interface disabled

Status: Enabled, Up

Uplink State Group : 3 Status: Enabled, Up Upstream Interfaces : Gi 0/46(Up) Gi 0/47(Up) Downstream Interfaces : Te 13/0(Up) Te 13/1(Up) Te 13/3(Up) Te 13/5(Up) Te 13/6(Up) Uplink State Group : 5 Status: Enabled, Down Upstream Interfaces : Gi 0/0(Dwn) Gi 0/3(Dwn) Gi 0/5(Dwn) Downstream Interfaces : Te 13/2(Dis) Te 13/4(Dis) Te 13/11(Dis) Te 13/12(Dis) Te 13/13(Dis) Te 13/14(Dis) Te 13/15(Dis) Uplink State Group : 6 Upstream Interfaces : Downstream Interfaces : Uplink State Group : 7 Upstream Interfaces : Downstream Interfaces : Status: Enabled, Up

Status: Enabled, Up

Uplink State Group : 16 Status: Disabled, Up Upstream Interfaces : Gi 0/41(Dwn) Po 8(Dwn) Downstream Interfaces : Gi 0/40(Dwn)

Uplink Failure Detection (UFD) | 785

www.dell.com | support.dell.com

Figure 42-4.

show interfaces Command: UFD Output (S50)

FTOS#show interfaces gigabitethernet 7/45 GigabitEthernet 7/45 is up, line protocol is down (error-disabled[UFD]) Hardware is Force10Eth, address is 00:01:e8:32:7a:47 Current address is 00:01:e8:32:7a:47 Interface index is 280544512 Internet address is not set MTU 1554 bytes, IP MTU 1500 bytes LineSpeed 1000 Mbit, Mode auto Flowcontrol rx off tx off ARP type: ARPA, ARP Timeout 04:00:00 Last clearing of "show interface" counters 00:25:46 Queueing strategy: fifo Input Statistics: 0 packets, 0 bytes 0 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 0 Multicasts, 0 Broadcasts 0 runts, 0 giants, 0 throttles 0 CRC, 0 overrun, 0 discarded Output Statistics: 0 packets, 0 bytes, 0 underruns 0 64-byte pkts, 0 over 64-byte pkts, 0 over 127-byte pkts 0 over 255-byte pkts, 0 over 511-byte pkts, 0 over 1023-byte pkts 0 Multicasts, 0 Broadcasts, 0 Unicasts 0 throttles, 0 discarded, 0 collisions Rate info (interval 299 seconds): Input 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Output 00.00 Mbits/sec, 0 packets/sec, 0.00% of line-rate Time since last interface status change: 00:01:23

Figure 42-5.

show running-config uplink-state-group Command: UFD Output (S50)

FTOS#show running-config uplink-state-group ! no enable uplink state track 1 downstream GigabitEthernet 0/2, 4, 6, 11-19 upstream TengigabitEthernet 0/48, 52 upstream PortChannel 1 ! uplink state track 2 downstream GigabitEthernet 0/1, 3, 5, 7-10 upstream TengigabitEthernet 0/56, 60

Figure 42-6.

show configuration Command: UFD Output (S50)

FTOS(conf-uplink-state-group-16)# show configuration ! uplink-state-group 16 no enable description test downstream disable links all downstream GigabitEthernet 0/40 upstream GigabitEthernet 0/41 upstream Port-channel 8

786

Uplink Failure Detection (UFD)

Sample Configuration: Uplink Failure Detection

Figure 42-7 shows a sample configuration of Uplink Failure Detection on a switch/router in which you: Configure uplink-state group 3. Add downstream links Gigabitethernet 0/1, 0/2, 0/5, 0/9, 0/11, and 0/12. Configure two downstream links to be disabled if an upstream link fails. Add upstream links Gigabitethernet 0/3 and 0/4. Add a text description for the group. Verify the configuration with various show commands.
FTOS(conf)# uplink-state-group 3 00:08:11: %STKUNIT0-M:CP %IFMGR-5-ASTATE_UP: Changed uplink state group Admin state to up: Group 3 FTOS(conf-uplink-state-group-3)# downstream gigabitethernet 0/1-2,5,9,11-12 FTOS(conf-uplink-state-group-3)# downstream disable links 2 FTOS(conf-uplink-state-group-3)# upstream gigabitethernet 0/3-4 00:10:00: %STKUNIT0-M:CP %IFMGR-5-OSTATE_DN: Downstream interface set to UFD error-disabled: Gi 0/1 FTOS# 00:10:00: %STKUNIT0-M:CP %IFMGR-5-OSTATE_DN: Changed interface state to down: Gi 0/1 FTOS(conf-uplink-state-group-3)# description Testing UFD feature

Figure 42-7. Configuring Uplink Failure Detection (S50)

FTOS(conf-uplink-state-group-3)# show config ! uplink-state-group 3 description Testing UFD feature downstream disable links 2 downstream GigabitEthernet 0/1-2,5,9,11-12 upstream GigabitEthernet 0/3-4 FTOS(conf-uplink-state-group-3)# FTOS(conf-uplink-state-group-3)#exit FTOS(conf)#exit FTOS# 00:13:06: %STKUNIT0-M:CP %SYS-5-CONFIG_I: Configured from console by

console

FTOS# show running-config uplink-state-group ! uplink-state-group 3 description Testing UFD feature downstream disable links 2 downstream GigabitEthernet 0/1-2,5,9,11-12 upstream GigabitEthernet 0/3-4

FTOS# show uplink-state-group 3 Uplink State Group: 3 Status: Enabled, Up

FTOS# show uplink-state-group detail (Up): Interface up (Dwn): Interface down (Dis): Interface disabled Uplink State Group : 3 Status: Enabled, Up Upstream Interfaces : Gi 0/3(Up) Gi 0/4(Dwn) Downstream Interfaces : Gi 0/1(Dis) Gi 0/2(Dwn) Gi 0/5(Dwn) Gi 0/9(Dwn) Gi 0/11(Dwn) Gi 0/12(Dwn)

Uplink Failure Detection (UFD) | 787

www.dell.com | support.dell.com |

788

Uplink Failure Detection (UFD)

43
Upgrade Procedures
Find the upgrade procedures
Go to the FTOS Release Notes for your system type to see all the requirements to upgrade to the desired FTOS version. Follow the procedures in the FTOS Release Notes for the software version you wish to upgrade to.

Get Help with upgrades

Direct any questions or concerns about FTOS Upgrade Procedures to the Dell Networking Technical Support Center. You can reach Technical Support: On the Web: http://support.dell.com/force10 By email: [email protected] By phone: US and Canada: 866.965.5800, International: 408.965.5800

Upgrade Procedures | 789

790 www.dell.com | support.dell.com

Upgrade Procedures

44
Virtual LANs (VLAN)
VLANs are supported on platforms

ces

This section contains the following subsections: Default VLAN Port-Based VLANs VLANs and Port Tagging Configuration Task List for VLANs Enable Null VLAN as the Default VLAN

Virtual LANs, or VLANs, are a logical broadcast domain, or logical grouping of interfaces in a LAN, in which all data received is kept locally and broadcast to all members of the group. When in Layer 2 mode, VLANs move traffic at wire speed and can span multiple devices. FTOS supports up to 4093 port-based VLANs and 1 Default VLAN, as specified in IEEE 802.1Q. VLANs provide the following benefits: Improved security because you can isolate groups of users into different VLANs Ability to create one VLAN across multiple devices

For more information on VLANs, refer to IEEE Standard 802.1Q Virtual Bridged Local Area Networks. In this guide, see also: Bulk Configuration in Chapter 15, Interfaces VLAN Stacking

For a complete listing of all commands related to FTOS VLANs, see these FTOS Command Reference chapters: Interfaces chapter Port Authentication (802.1x) section in the Security chapter Chapter 13, GARP VLAN Registration Protocol. Chapter 35, Service Provider Bridging Chapter 29, Per-VLAN Spanning Tree Plus. For E-Series, see also the ACL VLAN Group and Force10 Resilient Ring Protocol chapters.

Virtual LANs (VLAN) | 791

www.dell.com | support.dell.com

Table 44-1 displays the defaults for VLANs in FTOS.

Table 44-1. Feature Spanning Tree group ID Mode Default VLAN ID VLAN Defaults on FTOS Default All VLANs are part of Spanning Tree group 0 Layer 2 (no IP address is assigned) VLAN 1

Default VLAN
When interfaces are configured for Layer 2 mode, they are automatically placed in the Default VLAN as untagged interfaces. Only untagged interfaces can belong to the Default VLAN.
Note: In FTOS software version 8.3.3.9, the default startup configuration is all 1G ports included in VLAN1.

Figure 44-1 displays the outcome of placing an interface in Layer 2 mode. To configure an interface for Layer 2 mode, use the switchport command. In Step 1, the switchport command places the interface in Layer 2 mode. In Step 2, the show vlan command in EXEC privilege mode indicates that the interface is now part of the Default VLAN (VLAN 1).
Figure 44-1. Interfaces and the Default VLAN Example
FTOS(conf)#int gi 3/2 FTOS(conf-if)#no shut FTOS(conf-if)#switchport FTOS(conf-if)#show config ! interface GigabitEthernet 3/2 no ip address switchport no shutdown FTOS(conf-if)#end FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 Status Active Active Q U T T Ports Gi 3/2 Po1(So 0/0-1) Gi 3/0

Step 1the switchport command places the interface in Layer 2 mode

Step 2the show vlan command indicates that the interface is now assigned to VLAN 1 (the * indicates the Default VLAN)

FTOS#

792

Virtual LANs (VLAN)

By default, VLAN 1 is the Default VLAN. To change that designation, use the default vlan-id command in the CONFIGURATION mode. You cannot delete the Default VLAN.
Note: An IP address cannot be assigned to the Default VLAN. To assign an IP address to a VLAN that is currently the Default VLAN, create another VLAN and assign it to be the Default VLAN. For details on assigning IP addresses, see Assign an IP address to a VLAN.

Untagged interfaces must be part of a VLAN. To remove an untagged interface from the Default VLAN, you must create another VLAN and place the interface into that VLAN. Alternatively, enter the no switchport command, and FTOS removes the interface from the Default VLAN. A tagged interface requires an additional step to remove it from Layer 2 mode. Since tagged interfaces can belong to multiple VLANs, you must remove the tagged interface from all VLANs, using the no tagged interface command. Only after the interface is untagged and a member of the Default VLAN can you use the no switchport command to remove the interface from Layer 2 mode. For more information, see VLANs and Port Tagging.

Port-Based VLANs
Port-based VLANs are a broadcast domain defined by different ports or interfaces. In FTOS, a port-based VLAN can contain interfaces from different line cards within the chassis. FTOS supports 4094 port-based VLANs.
Note: E-Series ExaScale platforms support 4094 VLANs with FTOS version 8.2.1.0 and later. Earlier ExaScale supports 2094 VLANS.

Port-based VLANs offer increased security for traffic, conserve bandwidth, and allow switch segmentation. Interfaces in different VLANs do not communicate with each other, adding some security to the traffic on those interfaces. Different VLANs can communicate between each other by means of IP routing. Because traffic is only broadcast or flooded to the interfaces within a VLAN, the VLAN conserves bandwidth. Finally, you can have multiple VLANs configured on one switch, thus segmenting the device. Interfaces within a port-based VLAN must be in Layer 2 mode and can be tagged or untagged in the VLAN ID.

VLANs and Port Tagging

To add an interface to a VLAN, it must be in Layer 2 mode. After you place an interface in Layer 2 mode, it is automatically placed in the Default VLAN. FTOS supports IEEE 802.1Q tagging at the interface level to filter traffic. When tagging is enabled, a tag header is added to the frame after the destination and source MAC addresses. That information is preserved as the frame moves through the network. Figure 44-2 illustrates the structure of a frame with a tag header. The VLAN ID is inserted in the tag header.

Virtual LANs (VLAN) | 793

www.dell.com | support.dell.com

Figure 44-2. Tagged Frame Format

Ethernet Preamble Destination Address 6 octets Source Address 6 octets Tag Header 4 octets Protocol Type 2 octets Data Frame Check Sequence 4 octets

FN00001B

45 - 1500 octets

The tag header contains some key information used by FTOS: The VLAN protocol identifier identifies the frame as tagged according to the IEEE 802.1Q specifications (2 bytes). Tag Control Information (TCI) includes the VLAN ID (2 bytes total). The VLAN ID can have 4,096 values, but 2 are reserved.

Note: The insertion of the tag header into the Ethernet frame increases the size of the frame to more than the 1518 bytes specified in the IEEE 802.3 standard. Some devices that are not compliant with IEEE 802.3 may not support the larger frame size.

Information contained in the tag header allows the system to prioritize traffic and to forward information to ports associated with a specific VLAN ID. Tagged interfaces can belong to multiple VLANs, while untagged interfaces can belong only to one VLAN.

Configuration Task List for VLANs

This section contains the following VLAN configuration tasks: Create a port-based VLAN (mandatory) Assign interfaces to a VLAN (optional) Assign an IP address to a VLAN (optional) Enable Null VLAN as the Default VLAN

Create a port-based VLAN

The Default VLAN as VLAN 1 is part of the system startup configuration and does not require configuration. To configure a port-based VLAN, you must create the VLAN and then add physical interfaces or port channel (LAG) interfaces to the VLAN. To create a port-based VLAN, use the following command in the CONFIGURATION mode:
Command Syntax
interface vlan vlan-id

Command Mode CONFIGURATION

Purpose Configure a port-based VLAN (if the vlan-id is different from the Default VLAN ID) and enter INTERFACE VLAN mode. After you create a VLAN, you must assign interfaces in Layer 2 mode to the VLAN to activate the VLAN.

794

Virtual LANs (VLAN)

Use the show vlan command (Figure 44-3) in the EXEC privilege mode to view the configured VLANs.
Figure 44-3. show vlan Command Example
FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 4 5 6 Status Inactive Active Active Active Active Active Q U U U T U U U Ports So 9/4-11 Gi 0/1,18 Gi 0/2,19 Gi 0/3,20 Po 1 Gi 0/12 So 9/0

FTOS#

A VLAN is active only if the VLAN contains interfaces and those interfaces are operationally up. In Figure 44-3, VLAN 1 is inactive because it contains the interfaces that are not active. The other VLANs listed in the Figure 44-3 contain enabled interfaces and are active.
Note: In a VLAN, the shutdown command stops Layer 3 (routed) traffic only. Layer 2 traffic continues to pass through the VLAN. If the VLAN is not a routed VLAN (that is, configured with an IP address), the shutdown command has no affect on VLAN traffic.

When you delete a VLAN (using the no interface vlan vlan-id command), any interfaces assigned to that VLAN are assigned to the Default VLAN as untagged interfaces.

Assign interfaces to a VLAN

Only interfaces in Layer 2 mode can be assigned to a VLAN using the tagged and untagged commands. Use the switchport command to place an interface in Layer 2 mode. These Layer 2 interfaces can further be designated as tagged or untagged. For more information, refer to the Interfaces chapter and Configure Layer 2 (Data Link) Mode. When an interface is placed in Layer 2 mode by the switchport command, the interface is automatically designated untagged and placed in the Default VLAN. To view which interfaces are tagged or untagged and to which VLAN they belong, use the show vlan command. For example, Figure 44-3 shows that six VLANs are configured, and two interfaces are assigned to VLAN 2. The Q column in the show vlan command example notes whether the interface is tagged (T) or untagged (U). For more information on this command, see the command statement in the Layer 2 chapter of the FTOS Command Reference. To view just the interfaces that are in Layer 2 mode, enter the show interfaces switchport command in the EXEC privilege mode or EXEC mode.

Virtual LANs (VLAN) | 795

www.dell.com | support.dell.com

To tag frames leaving an interface in Layer 2 mode, you must assign that interface to a port-based VLAN to tag it with that VLAN ID. To tag interfaces, use these commands in the following sequence:
Step 1 Command Syntax
interface vlan vlan-id tagged interface

Command Mode CONFIGURATION INTERFACE

Purpose Access the INTERFACE VLAN mode of the VLAN to which you want to assign the interface. Enable an interface to include the IEEE 802.1Q tag header.

Figure 44-4 shows the steps to add a tagged interface (in this case, port channel 1) to VLAN 4.
Figure 44-4. Example of Adding an Interface to Another VLAN
FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 Status Inactive Active Active Q Ports T T T T Po1(So 0/0-1) Gi 3/0 Po1(So 0/0-1) Gi 3/1

FTOS#config FTOS(conf)#int vlan 4 FTOS(conf-if-vlan)#tagged po 1 FTOS(conf-if-vlan)#show conf ! interface Vlan 4 no ip address tagged Port-channel 1 FTOS(conf-if-vlan)#end FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 4 FTOS# Status Inactive Active Active Active Q Ports T T T T T Po1(So 0/0-1) Gi 3/0 Po1(So 0/0-1) Gi 3/1 Po1(So 0/0-1)

Use the show vlan command to view the interfaces status. Interface (po 1) is tagged and in VLAN 2 and 3

In a port-based VLAN, use the tagged command to add the interface to another VLAN.

The show vlan command output displays the interfaces (po 1) changed status.

Except for hybrid ports, only a tagged interface can be a member of multiple VLANs. Hybrid ports can be assigned to two VLANs if the port is untagged in one VLAN and tagged in all others. When you remove a tagged interface from a VLAN (using the no tagged interface command), it will remain tagged only if it is a tagged interface in another VLAN. If the tagged interface is removed from the only VLAN to which it belongs, the interface is placed in the Default VLAN as an untagged interface.

796

Virtual LANs (VLAN)

Use the untagged command to move untagged interfaces from the Default VLAN to another VLAN:
Step 1 Command Syntax
interface vlan vlan-id untagged interface

Command Mode CONFIGURATION INTERFACE

Purpose Access the INTERFACE VLAN mode of the VLAN to which you want to assign the interface. Configure an interface as untagged. This command is available only in VLAN interfaces.

The no untagged interface command removes the untagged interface from a port-based VLAN and places the interface in the Default VLAN. You cannot use the no untagged interface command in the Default VLAN. Figure 44-5 illustrates the steps and commands to move an untagged interface from the Default VLAN to another VLAN.
Figure 44-5. Example of Moving an Untagged Interface to Another VLAN
FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 Status Active Active Active Q U T T T T Ports Gi 3/2 Po1(So 0/0-1) Gi 3/0 Po1(So 0/0-1) Gi 3/1

Use the show vlan command to determine interface status. Interface (gi 3/2) is untagged and in the Default VLAN (vlan 1).

4 Inactive FTOS#conf FTOS(conf)#int vlan 4 FTOS(conf-if-vlan)#untagged gi 3/2 FTOS(conf-if-vlan)#show config ! interface Vlan 4 no ip address untagged GigabitEthernet 3/2 FTOS(conf-if-vlan)#end FTOS#show vlan Codes: * - Default VLAN, G - GVRP VLANs NUM 1 2 3 4 FTOS# Status Inactive Active Active Active Q Ports T T T T U Po1(So 0/0-1) Gi 3/0 Po1(So 0/0-1) Gi 3/1 Gi 3/2

In a port-based VLAN (vlan 4), use the untagged command to add the interface to that VLAN.

The show vlan command output displays the interfaces changed status (gi 3/2). Since the Default VLAN no longer contains any interfaces, it is listed as inactive.

The only way to remove an interface from the Default VLAN is to place the interface in Default mode by entering the no switchport command in the INTERFACE mode.

Virtual LANs (VLAN) | 797

www.dell.com | support.dell.com

Assign an IP address to a VLAN

VLANs are a Layer 2 feature. For two physical interfaces on different VLANs to communicate, you must assign an IP address to the VLANs to route traffic between the two interfaces. The shutdown command in INTERFACE mode does not affect Layer 2 traffic on the interface; the shutdown command only prevents Layer 3 traffic from traversing over the interface.
Note: An IP address cannot be assigned to the Default VLAN, which, by default, is VLAN 1. To assign another VLAN ID to the Default VLAN, use the default vlan-id vlan-id command.

To assign an IP address, use the following command in INTERFACE mode:

Command Syntax
ip address ip-address mask [secondary]

Command Mode INTERFACE

Purpose Configure an IP address and mask on the interface. ip-address mask Enter an address in dotted-decimal format (A.B.C.D) and the mask must be in slash format (/24). secondary This is the interfaces backup IP address. You can configure up to eight secondary IP addresses.

In FTOS, VLANs and other logical interfaces can be placed in Layer 3 mode to receive and send routed traffic. For details, see Bulk Configuration.

VLAN Interface Counters

VLAN counters can be enabled for either Ingress packets, egress packets, or both. VLAN counters are disabled by default, and are supported on E-Series ExaScale exonly.
Command Syntax
enable vlan-counter [ingress | egress | all]

Command Mode CONFIGURATION

Purpose Configure ingress, egress or both counters for VLAN interfaces.

To return to the default without any VLAN counters, enter no enable vlan-counter.
Note: VLAN output counters may show higher than expected values because source-suppression drops are counted.

Native VLANs
Traditionally, ports can be either untagged for membership to one VLAN or tagged for membership to multiple VLANs. An untagged port must be connected to a VLAN-unaware station (one that does not understand VLAN tags), and a tagged port must be connected to a VLAN-aware station (one that generates and understands VLAN tags).

798

Virtual LANs (VLAN)

Native VLAN support breaks this barrier so that a port can be connected to both VLAN-aware and VLAN-unaware stations. Such ports are referred to as hybrid ports. Physical and port-channel interfaces may be hybrid ports. Native VLAN is useful in deployments where a Layer 2 port can receive both tagged and untagged traffic on the same physical port. The classic example is connecting a VOIP phone and a PC to the same port of the switch. The VOIP phone is configured to generate tagged packets (with VLAN = VOICE VLAN), and the attached PC generates untagged packets. To configure a port so that it can be a member of an untagged and tagged VLANs:
Step 1 2 3 4 Task Command Command Mode INTERFACE INTERFACE INTERFACE VLAN INTERFACE

Remove any Layer 2 or Layer 3 configurations from the interface. Configure the interface for hybrid mode. Configure the interface for switchport mode. Add the interface to a tagged or untagged VLAN.
portmode hybrid switchport

[tagged | untagged]

Note: An existing switchport or port channel interface cannot be configured for Native VLAN. Interfaces must have no other Layer 2 or Layer 3 configurations when entering the command portmode hybrid or a message like Message 1 is displayed. Message 1 Native VLAN Error
% Error: Port is in Layer-2 mode Gi 5/6.

Enable Null VLAN as the Default VLAN

In a Carrier Ethernet for Metro Service environment, service providers who perform frequent reconfigurations for customers with changing requirements occasionally enable multiple interfaces, each connected to a different customer, before the interfaces are fully configured. This presents a vulnerability because both interfaces are initially placed in the native VLAN, VLAN 1, and for that period customers are able to access each other's networks. FTOS has a Null VLAN to eliminate this vulnerability. When you enable the Null VLAN, all ports are placed into by it default, so that even if you activate the physical ports of multiple customers, no traffic is allowed to traverse the links until each port is place in another VLAN.
Task Disable the default VLAN, so that all ports belong to the Null VLAN until configured as a member of another VLAN. Command Syntax
default-vlan disable

Command Mode CONFIGURATION

Default: the default VLAN is enabled (no default-vlan disable).

Virtual LANs (VLAN) | 799

800 www.dell.com | support.dell.com

Virtual LANs (VLAN)

45
Virtual Router Redundancy Protocol (VRRP)
Virtual Router Redundancy Protocol (VRRP) is supported on platforms This chapter covers the following information: VRRP Overview VRRP Benefits VRRP Implementation VRRP Configuration Sample Configurations

ces

Virtual Router Redundancy Protocol (VRRP) is designed to eliminate a single point of failure in a statically routed network. This protocol is defined in RFC 2338 and RFC 3768.

VRRP Overview
VRRP specifies a MASTER router that owns the next hop IP and MAC address for end stations on a LAN. The MASTER router is chosen from the virtual routers by an election process and forwards packets sent to the next hop IP address. If the MASTER router fails, VRRP begins the election process to choose a new MASTER router and that new MASTER continues routing traffic. VRRP uses the Virtual Router Identifier (VRID) to identify each virtual router configured The IP address of the MASTER router is used as the next hop address for all end stations on the LAN. The other routers represented by IP addresses are BACKUP routers. VRRP packets are transmitted with the virtual router MAC address as the source MAC address. The MAC address is in the following format: 00-00-5E-00-01-{VRID}. The first three octets are unchangeable. The next two octets (00-01) indicate the address block assigned to the VRRP protocol, and are unchangeable. The final octet changes depending on the VRRP Virtual Router Identifier and allows for up to 255 VRRP routers on a network. Figure 45-1 shows a typical network configuration using VRRP. Instead of configuring the hosts on the network 10.10.10.0 with the IP address of either Router A or Router B as their default router; their default router is the IP Address configured on the virtual router. When any host on the LAN segment wants to access the Internet, it sends packets to the IP address of the virtual router.

Virtual Router Redundancy Protocol (VRRP) | 801

www.dell.com | support.dell.com

In Figure 45-1 below, Router A is configured as the MASTER router. It is configured with the IP address of the virtual router and sends any packets addressed to the virtual router through interface GigabitEthernet 1/1 to the Internet. As the BACKUP router, Router B is also configured with the IP address of the virtual router. If for any reason Router A becomes unavailable, VRRP elects a new MASTER Router. Router B assumes the duties of Router A and becomes the MASTER router. At that time, Router B responds to the packets sent to the virtual IP address. All workstations continue to use the IP address of the virtual router to address packets destined to the Internet. Router B receives and forwards them on interface GigabitEthernet 10/1. Until Router A resumes operation, VRRP allows Router B to provide uninterrupted service to the users on the LAN segment accessing the Internet.
Figure 45-1. Basic VRRP Configuration

INTERNET

Router A Master Router Virtual IP 10.10.10.3 Priority 255 Interface gi 1/0 10.10.10.1

Interface gi 1/1 63.62.154.23

Interface gi 10/1 204.1.78.37

Router B Backup Router Virtual IP 10.10.10.3 Priority 100 Interface gi 10/0 10.10.10.2

Virtual Router

Virtual IP Address 10.10.10.3

10.10.10.0/24 LAN Segment

IP Addresses Default Gateway

10.10.10.4 10.10.10.3

10.10.10.5 10.10.10.3

10.10.10.6 10.10.10.3

For more detailed information on VRRP, refer to RFC 2338, Virtual Router Redundancy Protocol.

802

Virtual Router Redundancy Protocol (VRRP)

FN0001_lp

VRRP Benefits
With VRRP configured on a network, end-station connectivity to the network is not subject to a single point-of-failure. End-station connections to the network are redundant and they are not dependent on IGP protocols to converge or update routing tables.

VRRP Implementation
E-Series supports an unlimited total number of VRRP groups on the router while supporting up to 255 VRRP groups on a single interface (Table 45-1). C-Series supports a total of 128 VRRP groups on the switch with varying number of maximum VRRP groups per interface (Table 45-1). S-Series supports a total of 120 VRRP groups on a switch with FTOS or a total of 20 VRRP groups when using SFTOS. The S-Series supports varying number of maximum VRRP groups per interface (Table 45-1). Within a single VRRP group, up to 12 virtual IP addresses are supported. Virtual IP addresses can belong to the primary or secondary IP address subnet configured on the interface. You can ping all the virtual IP addresses configured on the Master VRRP router from anywhere in the local subnet. Though FTOS on E-Series supports unlimited VRRP groups, default VRRP settings may affect the maximum number of groups that can be configured and work efficiently, as a result of hardware throttling VRRP advertisement packets reaching the RP2 processor on the E-Series, the CP on the C-Series, or the FP on the S-Series. To avoid throttling VRRP advertisement packets, Dell Networking recommends you to increase the VRRP advertisement interval to a value higher than the default value of 1 second. The recommendations are as follows:
Table 45-1. Recommended VRRP Advertise Intervals
Recommended Advertise Interval E-Series ExaScale Groups/Interface E-Series TeraScale

Total VRRP Groups

E-Series

C-Series

S-Series

C-Series

S-Series

Less than 250 Between 250 and 450 Between 450 and 600 Between 600 and 800 Between 800 and 1000 Between 1000 and 1200 Between 1200 and 1500

1 second 2 seconds 3 seconds 4 seconds 5 seconds 7 seconds 8 seconds

1 second 2 - 3 seconds 4 seconds 5 seconds 5 seconds 7 seconds 8 seconds

1 second 2 - 3 seconds 3 - 4 seconds 4 seconds 5 seconds 7 seconds 8 seconds

512 512 512 512 512 512 512

255 255 255 255 255 255 255

12 24 36 48 84 100 120

12 24 36 48 84 100 120

Virtual Router Redundancy Protocol (VRRP) | 803

www.dell.com | support.dell.com

Note: The 1500 VRRP groups are supported in FTOS Release 6.3.1.0 and later.

The recommendations in Table 45-1 may vary depending on various factors like ARP broadcasts, IP broadcasts, or STP before changing the advertisement interval. When the number of packets processed by RP2/CP/FP processor increases or decreases based on the dynamics of the network, the advertisement intervals in may increase or decrease accordingly.
CAUTION: Increasing the advertisement interval increases the VRRP Master dead interval, resulting in an increased failover time for Master/Backup election. Take extra caution when increasing the advertisement interval, as the increased dead interval may cause packets to be dropped during that switch-over time.

VRRP Configuration
By default, VRRP is not configured.

Configuration Task List for VRRP

The following list specifies the configuration tasks for VRRP: Create a Virtual Router (mandatory) Assign Virtual IP addresses (mandatory) Set VRRP Group (Virtual Router) Priority (optional) Configure VRRP Authentication (optional) Disable Preempt (optional) Change the Advertisement interval (optional) Track an Interface (optional)

For a complete listing of all commands related to VRRP, refer to FTOS Command Line Interface.

Create a Virtual Router

To enable VRRP, you must create a Virtual Router. In FTOS, a VRRP Group is identified by the Virtual Router Identifier (VRID). To enable a Virtual Router, use the following command in the INTERFACE mode. To delete a VRRP group, use the no vrrp-group vrid command in the INTERFACE mode.
Task Create a virtual router for that interface with a VRID. Command Syntax
vrrp-group vrid

Command Mode INTERFACE

VRID Range: 1-255

804

Virtual Router Redundancy Protocol (VRRP)

Task

Command Syntax

Command Mode

Note: The interface must already have a Primary IP Address defined, and be enabled. Figure 45-2. Command Example: vrrp-group
FTOS(conf)#int gi 1/1 FTOS(conf-if-gi-1/1)#vrrp-group 111 FTOS(conf-if-gi-1/1-vrid-111)# Virtual Router ID and VRRP Group identifier

Figure 45-3. Command Example Display: show config for the Interface
FTOS(conf-if-gi-1/1)#show conf ! interface GigabitEthernet 1/1 ip address 10.10.10.1/24 ! vrrp-group 111 no shutdown FTOS(conf-if-gi-1/1)#

Note that the interface has an IP Address and is enabled

Assign Virtual IP addresses

Virtual routers contain virtual IP addresses configured for that VRRP Group (VRID). A VRRP group does not transmit VRRP packets until you assign the Virtual IP address to the VRRP group. E-Series supports an unlimited total number of VRRP Groups on the router while supporting up to 255 VRRP groups on a single interface (Table 45-1). C-Series supports a total of 128 VRRP groups on the switch with varying number of maximum VRRP groups per interface (Table 45-1). S-Series supports a total of 120 VRRP groups on a switch with FTOS or a total of 20 VRRP groups when using SFTOS. The S-Series supports varying number of maximum VRRP groups per interface (Table 45-1). To activate a VRRP Group on an interface (so that VRRP group starts transmitting VRRP packets), configure at least one Virtual IP address in a VRRP group. The Virtual IP address is the IP address of the Virtual Router and does not require the IP address mask. You can configure up to 12 Virtual IP addresses on a single VRRP Group (VRID). The following rules apply to virtual IP addresses: The virtual IP addresses must be in the same subnet as the primary or secondary IP addresses configured on the interface. Though a single VRRP group can contain virtual IP addresses belonging to multiple IP subnets configured on the interface, Dell Networking recommends you configure virtual IP addresses belonging to the same IP subnet for any one VRRP group.

Virtual Router Redundancy Protocol (VRRP) | 805

www.dell.com | support.dell.com

For example, an interface (on which VRRP is to be enabled) contains a primary IP address of 50.1.1.1/24 and a secondary IP address of 60.1.1.1/24. The VRRP Group (VRID 1) must contain virtual addresses belonging to either subnet 50.1.1.0/24 or subnet 60.1.1.0/24, but not from both subnets (though FTOS allows the same). If the virtual IP address and the interfaces primary/secondary IP address are the same, the priority on that VRRP group MUST be set to 255. The interface then becomes the OWNER router of the VRRP group and the interfaces physical MAC address is changed to that of the owner VRRP groups MAC address. If multiple VRRP groups are configured on an interface, only one of the VRRP Groups can contain the interface primary or secondary IP address.

Configure a Virtual IP address with these commands in the following sequence in the INTERFACE mode.
Step 1 2 Task Configure a VRRP group. Configure virtual IP addresses for this VRID. Command Syntax
vrrp-group vrrp-id VRID Range: 1-255 virtual-address ip-address1 [...ip-address12] Range: up to 12 addresses

Command Mode INTERFACE INTERFACE -VRID

Figure 45-4. Command Example: virtual-address

FTOS(conf-if-gi-1/1-vrid-111)#virtual-address 10.10.10.1 FTOS(conf-if-gi-1/1-vrid-111)#virtual-address 10.10.10.2 FTOS(conf-if-gi-1/1-vrid-111)#virtual-address 10.10.10.3 FTOS(conf-if-gi-1/1-vrid-111)#

Figure 45-5. Command Example Display: show config for the Interface
FTOS(conf-if-gi-1/1)#show conf ! interface GigabitEthernet 1/1 ip address 10.10.10.1/24 ! vrrp-group 111 priority 255 virtual-address 10.10.10.1 virtual-address 10.10.10.2 virtual-address 10.10.10.3 ! vrrp-group 222 no shutdown FTOS(conf-if-gi-1/1)#

Note that the Primary IP address and the Virtual IP addresses are on the same subnet

Figure 45-6 shows the same VRRP group configured on multiple interfaces on different subnets.

806

Virtual Router Redundancy Protocol (VRRP)

Figure 45-6. Command Example Display: show vrrp

Same VRRP Group (VRID) FTOS#do show vrrp -----------------GigabitEthernet 1/1, VRID: 111, Net: 10.10.10.1 State: Master, Priority: 255, Master: 10.10.10.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 1768, Gratuitous ARP sent: 5 Virtual MAC address: 00:00:5e:00:01:6f Virtual IP address: 10.10.10.1 10.10.10.2 10.10.10.3 10.10.10.10 Authentication: (none) -----------------GigabitEthernet 1/2, VRID: 111, Net: 10.10.2.1 State: Master, Priority: 100, Master: 10.10.2.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 27, Gratuitous ARP sent: 2 Virtual MAC address: 00:00:5e:00:01:6f Virtual IP address: 10.10.2.2 10.10.2.3 Authentication: (none) FTOS#

Different Virtual IP addresses

When the VRRP process completes its initialization, the State field contains either Master or Backup.

Set VRRP Group (Virtual Router) Priority

Setting a Virtual Router priority to 255 ensures that router is the owner virtual router for the VRRP group. VRRP elects the MASTER router by choosing the router with the highest priority. THe default priority for a Virtual Router is 100. The higher the number, the higher the priority. If the MASTER router fails, VRRP begins the election process to choose a new MASTER router based on the next-highest priority. If two routers in a VRRP group come up at the same time and have the same priority value, the interfaces physical IP addresses are used as tie-breakers to decide which is MASTER. The router with the higher IP address will become MASTER.

Virtual Router Redundancy Protocol (VRRP) | 807

www.dell.com | support.dell.com

Configure the VRRP Groups priority with the following command in the VRRP mode:
Task Configure the priority for the VRRP group. Command Syntax INTERFACE -VRID Command Mode
priority priority

Range: 1-255 Default: 100 Figure 45-7. Command Example: priority in Interface VRRP mode
FTOS(conf-if-gi-1/2)#vrrp-group 111 FTOS(conf-if-gi-1/2-vrid-111)#priority 125

Figure 45-8. Command Example Display: show vrrp

FTOS#show vrrp -----------------GigabitEthernet 1/1, VRID: 111, Net: 10.10.10.1 State: Master, Priority: 255, Master: 10.10.10.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 2343, Gratuitous ARP sent: 5 Virtual MAC address: 00:00:5e:00:01:6f Virtual IP address: 10.10.10.1 10.10.10.2 10.10.10.3 10.10.10.10 Authentication: (none) -----------------GigabitEthernet 1/2, VRID: 111, Net: 10.10.2.1 State: Master, Priority: 125, Master: 10.10.2.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 601, Gratuitous ARP sent: 2 Virtual MAC address: 00:00:5e:00:01:6f Virtual IP address: 10.10.2.2 10.10.2.3 Authentication: (none) FTOS(conf)#

Configure VRRP Authentication

Simple authentication of VRRP packets ensures that only trusted routers participate in VRRP processes. When authentication is enabled, FTOS includes the password in its VRRP transmission, and the receiving router uses that password to verify the transmission.
Note: All virtual routers in the VRRP group must be configured the same: authentication must be enabled with the same password or authentication is disabled.

808

Virtual Router Redundancy Protocol (VRRP)

Configure simple authentication with the following command in the VRRP mode:
Task Configure a simple text password. Command Syntax
authentication-type simple [encryption-type] password

Command Mode INTERFACE-VRID

Parameters:
encryption-type: 0 indicates unencrypted; 7 indicates encrypted password: plain text

Figure 45-9. Command Example: authentication-type

FTOS(conf-if-gi-1/1-vrid-111)#authentication-type ? FTOS(conf-if-gi-1/1-vrid-111)#authentication-type simple 7 FTOS

Encryption type (encrypted)

Password

Figure 45-10.

Command Example: show config in VRID mode with a Simple Password Configured

FTOS(conf-if-gi-1/1-vrid-111)#show conf ! vrrp-group 111 authentication-type simple 7 387a7f2df5969da4 priority 255 virtual-address 10.10.10.1 virtual-address 10.10.10.2 virtual-address 10.10.10.3 virtual-address 10.10.10.10 FTOS(conf-if-gi-1/1-vrid-111)#

Encrypted password

Disable Preempt
The preempt command is enabled by default, and it forces the system to change the MASTER router if another router with a higher priority comes online. Prevent the BACKUP router with the higher priority from becoming the MASTER router by disabling preempt.
Note: All virtual routers in the VRRP group must be configured the same: all configured with preempt enabled or configured with preempt disabled.

Virtual Router Redundancy Protocol (VRRP) | 809

www.dell.com | support.dell.com

Since preempt is enabled by default, disable the preempt function with the following command in the VRRP mode. Re-enable preempt by entering the preempt command. When preempt is enabled, it does not display in the show commands, because it is a default setting.,
Task Prevent any BACKUP router with a higher priority from becoming the MASTER router. Figure 45-11. Command Syntax
no preempt

Command Mode INTERFACE-VRID

Command Example: no preempt

FTOS(conf-if-gi-1/1)#vrrp-group 111 FTOS(conf-if-gi-1/1-vrid-111)#no preempt FTOS(conf-if-gi-1/1-vrid-111)#show conf

Figure 45-12.

Command Example Display: show config in VRID mode

FTOS(conf-if-gi-1/1-vrid-111)#show conf ! vrrp-group 111 authentication-type simple 7 387a7f2df5969da4 no preempt priority 255 virtual-address 10.10.10.1 virtual-address 10.10.10.2 virtual-address 10.10.10.3 virtual-address 10.10.10.10 FTOS(conf-if-gi-1/1-vrid-111)#

Change the Advertisement interval

By default, the MASTER router transmits a VRRP advertisement to all members of the VRRP group every 1 second, indicating it is operational and is the MASTER router. If the VRRP group misses 3 consecutive advertisements, then the election process begins and the BACKUP virtual router with the highest priority transitions to MASTER.
Note: Dell Networking recommends you to increase the VRRP advertisement interval to a value higher than the default value of 1 second to avoid throttling VRRP advertisement packets. If you do change the time interval between VRRP advertisements on one router, you must change it on all participating routers.

Change that advertisement interval with the following command in the VRRP mode:
Task Change the advertisement interval setting. Command Syntax
advertise-interval seconds Range: 1-255 seconds Default: 1 second

Command Mode INTERFACE-VRID

810

Virtual Router Redundancy Protocol (VRRP)

Figure 45-13.

Command Example: advertise-interval

FTOS(conf-if-gi-1/1)#vrrp-group 111 FTOS(conf-if-gi-1/1-vrid-111)#advertise-interval 10 FTOS(conf-if-gi-1/1-vrid-111)#

Figure 45-14.

Command Example Display: advertise-interval in VRID mode

FTOS(conf-if-gi-1/1-vrid-111)#show conf ! vrrp-group 111 advertise-interval 10 authentication-type simple 7 387a7f2df5969da4 no preempt priority 255 virtual-address 10.10.10.1 virtual-address 10.10.10.2 virtual-address 10.10.10.3 virtual-address 10.10.10.10 FTOS(conf-if-gi-1/1-vrid-111)#

Track an Interface
Set FTOS to monitor the state of any interface according to the Virtual group. Each VRRP group can track up to 12 interfaces, which may affect the priority of the VRRP group. If the tracked interface goes down, the VRRP groups priority is decreased by a default value of 10 (also known as cost). If the tracked interfaces state goes up, the VRRP groups priority is increased by 10. The lowered priority of the VRRP group may trigger an election. As the Master/Backup VRRP routers are selected based on the VRRP groups priority, tracking features ensure that the best VRRP router is the Master for that group. The sum of all the costs of all the tracked interfaces should not exceed the configured priority on the VRRP group. If the VRRP group is configured as Owner router (priority 255), tracking for that group is disabled, irrespective of the state of the tracked interfaces. The priority of the owner group always remains at 255. To track an interface, use the following command in the VRRP mode:
Task Monitor an interface and, optionally, set a value to be subtracted from the interfaces VRRP group priority. Command Syntax
track interface [priority-cost cost] Cost Range: 1-254 Default: 10

Command Mode INTERFACE-VRID

The sum of all the costs for all tracked interfaces must be less than or equal to the configured priority of the VRRP group.

Virtual Router Redundancy Protocol (VRRP) | 811

www.dell.com | support.dell.com

Figure 45-15.

Command Example: track

FTOS(conf-if-gi-1/1)#vrrp-group 111 FTOS(conf-if-gi-1/1-vrid-111)#track gigabitethernet 1/2 FTOS(conf-if-gi-1/1-vrid-111)#

Figure 45-16.

Command Example Display: track in VRID mode

FTOS(conf-if-gi-1/1-vrid-111)#show conf ! vrrp-group 111 advertise-interval 10 authentication-type simple 7 387a7f2df5969da4 no preempt priority 255 track GigabitEthernet 1/2 virtual-address 10.10.10.1 virtual-address 10.10.10.2 virtual-address 10.10.10.3 virtual-address 10.10.10.10 FTOS(conf-if-gi-1/1-vrid-111)#

Sample Configurations
The following configurations are examples for enabling VRRP. These are not comprehensive directions. They are intended to give you a some guidance with typical configurations. You can copy and paste from these examples to your CLI. Be sure you make the necessary changes to support your own IP Addresses, Interfaces, Names, etc. Figure 45-17 is a sample configuration for enabling VRRP. Figure 45-18 illustrates the topology created with that CLI configuration.

812

Virtual Router Redundancy Protocol (VRRP)

Figure 45-17.

Configure VRRP

Router 2
R2(conf)#int gi 2/31 R2(conf-if-gi-2/31)#ip address 10.1.1.1/24 R2(conf-if-gi-2/31)#no shut R2(conf-if-gi-2/31)#vrrp-group 99 R2(conf-if-gi-2/31-vrid-99)#virtual 10.1.1.2 R2(conf-if-gi-2/31-vrid-99)#no shut R2(conf-if-gi-2/31)#show conf ! interface GigabitEthernet 2/31 ip address 10.1.1.1/24 ! vrrp-group 99 virtual-address 10.1.1.3 no shutdown R2(conf-if-gi-2/31)#end R2#show vrrp -----------------GigabitEthernet 2/31, VRID: 99, Net: 10.1.1.1 State: Master, Priority: 100, Master: 10.1.1.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 817, Gratuitous ARP sent: 1 Virtual MAC address: 00:00:5e:00:01:63 Virtual IP address: 10.1.1.3 Authentication: (none) R2#

Router 3
R3(conf)#int gi 3/21 R3(conf-if-gi-3/21)#ip add 10.1.1.1/24 R3(conf-if-gi-3/21)#no shut R3(conf-if-gi-3/21)#vrrp-group 99 R3(conf-if-gi-3/21-vrid-99)#no shut R3(conf-if-gi-3/21-vrid-99)#virtual 10.1.1.3 R3(conf-if-gi-3/21)#show conf ! interface GigabitEthernet 3/21 ip address 10.1.1.1/24 no shutdown ! vrrp-group 99 virtual-address 10.1.1.3 no shutdown R3(conf-if-gi-3/21)#end R3#show vrrp -----------------GigabitEthernet 3/21, VRID: 99, Net: 10.1.1.1 State: Backup, Priority: 100, Master: 10.1.1.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 698, Bad pkts rcvd: 0, Adv sent: 0, Gratuitous ARP sent: 0 Virtual MAC address: 00:00:5e:00:01:63 Virtual IP address: 10.1.1.3 Authentication: (none) R3#

Virtual Router Redundancy Protocol (VRRP) | 813

www.dell.com | support.dell.com

Figure 45-18.

VRRP Topography Illustration

R2#show vrrp -----------------GigabitEthernet 2/31, VRID: 99, Net: 10.1.1.1 State: Master, Priority: 100, Master: 10.1.1.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 0, Bad pkts rcvd: 0, Adv sent: 661, Gratuitous ARP sent: 1 Virtual MAC address: 00:00:5e:00:01:63 Virtual IP address: 10.1.1.3 Authentication: (none) R2#

State Master: R2 was the first interface configured with VRRP

Virtual MAC is automatically assigned and is the same on both Routers

State Backup: R3 was the second interface configured with VRRP

R3#show vrrp -----------------GigabitEthernet 3/21, VRID: 99, Net: 10.1.1.1 State: Backup, Priority: 100, Master: 10.1.1.1 (local) Hold Down: 0 sec, Preempt: TRUE, AdvInt: 1 sec Adv rcvd: 331, Bad pkts rcvd: 0, Adv sent: 0, Gratuitous ARP sent: 0 Virtual MAC address: 00:00:5e:00:01:63 Virtual IP address: 10.1.1.3 Authentication: (none) R3#

10.1.1.2
GigE 2/31

10.1.1.1
GigE 3/21

R2

VRID 99 10.1.1.3

R3

Internet

814

Virtual Router Redundancy Protocol (VRRP)

46
Standards Compliance
This appendix contains the following sections: IEEE Compliance RFC and I-D Compliance MIB Location
Note: Unless noted, when a standard cited here is listed as supported by FTOS, FTOS also supports predecessor standards. One way to search for predecessor standards is to use the http://tools.ietf.org/ website. Click on Browse and search IETF documents, enter an RFC number, and inspect the top of the resulting document for obsolescence citations to related RFCs.

IEEE Compliance
802.1AB LLDP 802.1D Bridging, STP 802.1p L2 Prioritization 802.1Q VLAN Tagging, Double VLAN Tagging, GVRP 802.1s MSTP 802.1w RSTP 802.1X Network Access Control (Port Authentication) 802.3ab Gigabit Ethernet (1000BASE-T) 802.3ac Frame Extensions for VLAN Tagging 802.3ad Link Aggregation with LACP 802.3ae 10 Gigabit Ethernet (10GBASE-W, 10GBASE-X) 802.3af Power over Ethernet 802.3ak 10 Gigabit Ethernet (10GBASE-CX4) 802.3i Ethernet (10BASE-T) 802.3u Fast Ethernet (100BASE-FX, 100BASE-TX) 802.3x Flow Control 802.3z Gigabit Ethernet (1000BASE-X) ANSI/TIA-1057 LLDP-MED Force10 FRRP (Force10 Redundant Ring Protocol)

Standards Compliance | 815

www.dell.com | support.dell.com

Force10 PVST+ SFF-8431 SFP+ Direct Attach Cable (10GSFP+Cu) MTU 9,252 bytes

RFC and I-D Compliance

The following standards are supported by FTOS, and are grouped by related protocol. The columns showing support by platform indicate which version of FTOS first supports the standard.
Note: Checkmarks () in the E-Series column indicate that FTOS support was added before FTOS version 7.5.1.

General Internet Protocols

FTOS support, per platform RFC# 768 793 854 959 1321 1350 1661 1989 1990 1994 2474 2615 2615 2698 3164 Full Name User Datagram Protocol Transmission Control Protocol Telnet Protocol Specification File Transfer Protocol (FTP) The MD5 Message-Digest Algorithm The TFTP Protocol (Revision 2) The Point-to-Point Protocol (PPP) PPP Link Quality Monitoring The PPP Multilink Protocol (MP) PPP Challenge Handshake Authentication Protocol (CHAP) Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers PPP over SONET/SDH PPP over SONET/SDH A Two Rate Three Color Marker The BSD syslog Protocol 7.6.1 7.5.1 7.6.1 7.7.1 7.5.1 E-Series E-Series S-Series C-Series TeraScale ExaScale 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

draft-ietf-bfd Bidirectional Forwarding Detection -base-03

816

Standards Compliance

General IPv4 Protocols

FTOS support, per platform RFC# 791 792 826 1027 1035 1042 1191 1305 1519 1542 1812 2131 2338 3021 3046 3069 3128 Full Name Internet Protocol Internet Control Message Protocol An Ethernet Address Resolution Protocol Using ARP to Implement Transparent Subnet Gateways DOMAIN NAMES - IMPLEMENTATION AND SPECIFICATION (client) A Standard for the Transmission of IP Datagrams over IEEE 802 Networks Path MTU Discovery Network Time Protocol (Version 3) Specification, Implementation and Analysis Classless Inter-Domain Routing (CIDR): an Address Assignment and Aggregation Strategy Clarifications and Extensions for the Bootstrap Protocol Requirements for IP Version 4 Routers Dynamic Host Configuration Protocol Virtual Router Redundancy Protocol (VRRP) Using 31-Bit Prefixes on IPv4 Point-to-Point Links DHCP Relay Agent Information Option VLAN Aggregation for Efficient IP Address Allocation Protection Against a Variant of the Tiny Fragment Attack E-Series E-Series S-Series C-Series TeraScale ExaScale 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.7.1 7.8.1 7.8.1 7.6.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.7.1 7.8.1 7.8.1 7.5.1 8.1.1 7.7.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

Standards Compliance | 817

www.dell.com | support.dell.com

General IPv6 Protocols

FTOS support, per platform RFC# 1886 1981 (Partial) 2460 2461 (Partial) 2462 (Partial) 2463 2464 2675 3587 4291 Full Name DNS Extensions to support IP version 6 Path MTU Discovery for IP version 6 Internet Protocol, Version 6 (IPv6) Specification Neighbor Discovery for IP Version 6 (IPv6) IPv6 Stateless Address Autoconfiguration Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification Transmission of IPv6 Packets over Ethernet Networks IPv6 Jumbograms IPv6 Global Unicast Address Format Internet Protocol Version 6 (IPv6) Addressing Architecture S-Series 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 C-Series 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 E-Series E-Series TeraScale ExaScale 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1 8.2.1

818

Standards Compliance

Border Gateway Protocol (BGP)

FTOS support, per platform RFC# 1997 2385 2439 2545 2796 2842 2858 2918 3065 4360 4893 5396 Full Name BGP Communities Attribute Protection of BGP Sessions via the TCP MD5 Signature Option BGP Route Flap Damping Use of BGP-4 Multiprotocol Extensions for IPv6 Inter-Domain Routing BGP Route Reflection: An Alternative to Full Mesh Internal BGP (IBGP) Capabilities Advertisement with BGP-4 Multiprotocol Extensions for BGP-4 Route Refresh Capability for BGP-4 Autonomous System Confederations for BGP BGP Extended Communities Attribute BGP Support for Four-octet AS Number Space Textual Representation of Autonomous System (AS) Numbers 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 7.8.1 8.1.2 7.8.1 7.8.1 S-Series 7.8.1 7.8.1 7.8.1 C-Series 7.7.1 7.7.1 7.7.1 7.8.1 7.7.1 7.7.1 7.7.1 7.7.1 7.7.1 7.7.1 7.7.1 8.1.2 7.7.1 7.7.1 E-Series TeraScale 7.6.1 7.7.1 8.1.2 E-Series ExaScale 8.1.1 8.1.1 8.1.1 8.2.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.2.1 8.1.1 8.1.1

draft-ietf-id A Border Gateway Protocol 4 (BGP-4) r-bgp4-20 draft-ietf-id Graceful Restart Mechanism for BGP r-restart-06

Standards Compliance | 819

www.dell.com | support.dell.com

Open Shortest Path First (OSPF)

FTOS support, per platform RFC# 1587 2154 2328 2370 2740 3623 4222 Full Name The OSPF Not-So-Stubby Area (NSSA) Option OSPF with Digital Signatures OSPF Version 2 The OSPF Opaque LSA Option OSPF for IPv6 Graceful OSPF Restart Prioritized Treatment of Specific OSPF Version 2 Packets and Congestion Avoidance 7.8.1 7.6.1 S-Series 7.6.1 7.6.1 7.6.1 7.6.1 E-Series E-Series C-Series TeraScale ExaScale 7.5.1 7.5.1 7.5.1 7.5.1 7.8.1 7.5.1 7.5.1 8.1.1 8.1.1 8.1.1 8.1.1 8.2.1 8.1.1 8.1.1

820

Standards Compliance

Intermediate System to Intermediate System (IS-IS)

FTOS support, per platform RFC# 1142 1195 2763 2966 3373 Full Name OSI IS-IS Intra-Domain Routing Protocol (ISO DP 10589) Use of OSI IS-IS for Routing in TCP/IP and Dual Environments Dynamic Hostname Exchange Mechanism for IS-IS Domain-wide Prefix Distribution with Two-Level IS-IS Three-Way Handshake for Intermediate System to Intermediate System (IS-IS) Point-to-Point Adjacencies IS-IS Cryptographic Authentication Intermediate System to Intermediate System (IS-IS) Extensions in Support of Generalized Multi-Protocol Label Switching (GMPLS) M-ISIS: Multi Topology (MT) Routing in Intermediate System to Intermediate Systems (IS-ISs) Restart Signaling for IS-IS Point-to-point operation over LAN in link-state routing protocols Routing IPv6 with IS-IS S-Series C-Series E-Series TeraScale E-Series ExaScale 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

3567 3784

8.1.1 8.1.1

5120

7.8.1

8.2.1

5306 draft-ietf-isis-ig p-p2p-over-lan06 draft-ietf-isis -ipv6-06

8.3.1

8.3.1 8.1.1

7.5.1

8.2.1 8.1.1

draft-kaplan-isis Extended Ethernet Frame Size Support -ext-eth-02

Standards Compliance | 821

www.dell.com | support.dell.com

Routing Information Protocol (RIP)

FTOS support, per platform RFC# 1058 2453 Full Name Routing Information Protocol RIP Version 2 S-Series 7.8.1 7.8.1 C-Series 7.6.1 7.6.1 E-Series TeraScale E-Series ExaScale 8.1.1 8.1.1

Multiprotocol Label Switching (MPLS)

FTOS support, per platform RFC# 2702 3031 3032 3209 3630 3784 Full Name Requirements for Traffic Engineering Over MPLS Multiprotocol Label Switching Architecture MPLS Label Stack Encoding RSVP-TE: Extensions to RSVP for LSP Tunnels Traffic Engineering (TE) Extensions to OSPF Version 2 Intermediate System to Intermediate System (IS-IS) Extensions for Traffic Engineering (TE) Multiprotocol Label Switching (MPLS) Traffic Engineering (TE) Management Information Base (MIB) Multiprotocol Label Switching (MPLS) Label Switching Router (LSR) Management Information Base (MIB) Fast Reroute Extensions to RSVP-TE for LSP Tunnels Detecting Multi-Protocol Label Switched Data Plane Failures (MPLS TE/LDP Ping & Traceroute LDP Specification Extensions to GMPLS Resource Reservation Protocol (RSVP) Graceful Restart S-Series C-Series E-Series TeraScale E-Series ExaScale 8.3.1 8.3.1 8.3.1 8.3.1 8.3.1 8.3.1

3812

8.3.1

3813

8.3.1

4090 4379

8.3.1 8.3.1

5036 5063

8.3.1 8.3.1

822

Standards Compliance

Multicast
FTOS support, per platform RFC# 1112 2236 2710 3376 3569 Full Name Host Extensions for IP Multicasting Internet Group Management Protocol, Version 2 Multicast Listener Discovery (MLD) for IPv6 Internet Group Management Protocol, Version 3 An Overview of Source-Specific Multicast (SSM) Multicast Source Discovery Protocol (MSDP) Multicast Listener Discovery Version 2 (MLDv2) for IPv6 Protocol Independent Multicast - Dense Mode (PIM-DM): Protocol Specification (Revised) 7.6.1 Considerations for Internet Group Management Protocol (IGMP) and Multicast (IGMPv1/v2) Listener Discovery (MLD) Snooping Switches 7.6.1 (IGMPv1/v2) 7.8.1 7.7.1 S-Series 7.8.1 7.8.1 C-Series 7.7.1 7.7.1 E-Series TeraScale 7.5.1 SSM for IPv4/ IPv6 IGMPv1/v2/ v3, MLDv1 Snooping IPv4/ IPv6 8.2.1 IGMPv1/ v2/v3, MLDv1 Snooping 8.2.1 PIM-SM for IPv4/ IPv6 E-Series ExaScale 8.1.1 8.1.1 8.2.1 8.1.1 8.2.1 SSM for IPv4 8.1.1 8.2.1

7.8.1 7.7.1 SSM for IPv4 SSM for IPv4

3618 3810 3973 4541

draft-ietf-pi Protocol Independent Multicast - Sparse Mode m-sm-v2-ne (PIM-SM): Protocol Specification (Revised) w-05

7.8.1 PIM-SM for IPv4

7.7.1

Standards Compliance | 823

www.dell.com | support.dell.com

Network Management
FTOS support, per platform RFC# 1155 1156 1157 1212 1215 1493 Full Name Structure and Identification of Management Information for TCP/IP-based Internets Management Information Base for Network Management of TCP/IP-based internets A Simple Network Management Protocol (SNMP) Concise MIB Definitions A Convention for Defining Traps for use with the SNMP Definitions of Managed Objects for Bridges [except for the dot1dTpLearnedEntryDiscards object] RIP Version 2 MIB Extension OSPF Version 2 Management Information Base Introduction to Community-based SNMPv2 SNMPv2 Management Information Base for the Internet Protocol using SMIv2 SNMPv2 Management Information Base for the Transmission Control Protocol using SMIv2 SNMPv2 Management Information Base for the User Datagram Protocol using SMIv2 Definitions of Managed Objects for Data Link Switching using SMIv2 IP Forwarding Table MIB Definitions of Managed Objects for the Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH) Interface Type Introduction and Applicability Statements for Internet Standard Management Framework An Architecture for Describing Simple Network Management Protocol (SNMP) Management Frameworks 7.6.1 7.6.1 7.5.1 7.5.1 7.6.1 7.6.1 7.6.1 7.6.1 S-Series 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 7.6.1 C-Series 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 7.5.1 E-Series TeraScale E-Series ExaScale 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

1724 1850 1901 2011 2012

7.5.1 7.5.1 7.5.1 7.5.1 7.5.1

8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

2013 2024 2096 2558

7.6.1 7.6.1 7.6.1

7.5.1 7.5.1 7.5.1

8.1.1 8.1.1 8.1.1

2570 2571

8.1.1 8.1.1

824

Standards Compliance

Network Management (continued)

FTOS support, per platform RFC# 2572 Full Name Message Processing and Dispatching for the Simple Network Management Protocol (SNMP) User-based Security Model (USM) for version 3 of the Simple Network Management Protocol (SNMPv3) View-based Access Control Model (VACM) for the Simple Network Management Protocol (SNMP) Coexistence Between Version 1, Version 2, and Version 3 of the Internet-standard Network Management Framework Structure of Management Information Version 2 (SMIv2) Textual Conventions for SMIv2 Conformance Statements for SMIv2 RADIUS Authentication Client MIB, except the following four counters: radiusAuthClientInvalidServerAddresses radiusAuthClientMalformedAccessResponses radiusAuthClientUnknownTypes radiusAuthClientPacketsDropped Definitions of Managed Objects for the Ethernet-like Interface Types Definitions of Managed Objects for Bridges with Traffic Classes, Multicast Filtering and Virtual LAN Extensions Definitions of Managed Objects for the Virtual Router Redundancy Protocol Remote Network Monitoring Management Information Base: Ethernet Statistics Table, Ethernet History Control Table, Ethernet History Table, Alarm Table, Event Table, Log Table The Interfaces Group MIB Remote Authentication Dial In User Service (RADIUS) S-Series 7.6.1 C-Series 7.5.1 E-Series TeraScale E-Series ExaScale 8.1.1

2574

7.6.1

7.5.1

8.1.1

2575

7.6.1

7.5.1

8.1.1

2576

7.6.1

7.5.1

8.1.1

2578 2579 2580 2618

7.6.1 7.6.1 7.6.1 7.6.1

7.5.1 7.5.1 7.5.1 7.5.1

8.1.1 8.1.1 8.1.1 8.1.1

2665 2674

7.6.1 7.6.1

7.5.1 7.5.1

8.1.1 8.1.1

2787 2819

7.6.1 7.6.1

7.5.1 7.5.1

8.1.1 8.1.1

2863 2865

7.6.1 7.6.1

7.5.1 7.5.1

8.1.1 8.1.1

Standards Compliance | 825

www.dell.com | support.dell.com

Network Management (continued)

FTOS support, per platform RFC# 3273 Full Name Remote Network Monitoring Management Information Base for High Capacity Networks (64 bits): Ethernet Statistics High-Capacity Table, Ethernet History High-Capacity Table Version 2 of the Protocol Operations for the Simple Network Management Protocol (SNMP) Management Information Base (MIB) for the Simple Network Management Protocol (SNMP) Remote Monitoring MIB Extensions for High Capacity Alarms, High-Capacity Alarm Table (64 bits) IEEE 802.1X Remote Authentication Dial In User Service (RADIUS) Usage Guidelines Definitions of Managed Objects for the Multiprotocol Label Switching (MPLS), Label Distribution Protocol (LDP) Protocol Independent Multicast MIB 7.8.1 7.7.1 7.8.1 7.6.1 7.7.1 7.6.1 S-Series 7.6.1 C-Series 7.5.1 E-Series TeraScale E-Series ExaScale 8.1.1

3416

7.6.1

7.5.1

8.1.1

3418

7.6.1

7.5.1

8.1.1

3434

7.6.1

7.5.1

8.1.1

3580 3815

7.6.1

7.5.1

8.1.1 8.3.1

5060

8.1.1 8.1.1

ANSI/TIA-1057 The LLDP Management Information Base extension module for TIA-TR41.4 Media Endpoint Discovery information draft-grant-taca cs-02 draft-ietf-idr-bg p4-mib-06 draft-ietf-isis-w g-mib-16 The TACACS+ Protocol Definitions of Managed Objects for the Fourth Version of the Border Gateway Protocol (BGP-4) using SMIv2 Management Information Base for Intermediate System to Intermediate System (IS-IS): isisSysObject (top level scalar objects) isisISAdjTable isisISAdjAreaAddrTable isisISAdjIPAddrTable isisISAdjProtSuppTable Management Information Base module for LLDP configuration, statistics, local system data and remote systems data components.

7.6.1 7.8.1

7.5.1 7.7.1

8.1.1 8.1.1

8.1.1

IEEE 802.1AB

7.7.1

7.6.1

7.6.1

8.1.1

826

Standards Compliance

Network Management (continued)

FTOS support, per platform RFC# IEEE 802.1AB Full Name The LLDP Management Information Base extension module for IEEE 802.1 organizationally defined discovery information. (LLDP DOT1 MIB and LLDP DOT3 MIB) The LLDP Management Information Base extension module for IEEE 802.3 organizationally defined discovery information. (LLDP DOT1 MIB and LLDP DOT3 MIB) S-Series 7.7.1 C-Series 7.6.1 E-Series TeraScale 7.6.1 E-Series ExaScale 8.1.1

IEEE 802.1AB

7.7.1

7.6.1

7.6.1

8.1.1

ruzin-mstp-mib- Definitions of Managed Objects for Bridges 02 (Traps) with Multiple Spanning Tree Protocol sFlow.org sFlow.org sFlow Version 5 sFlow Version 5 MIB

7.6.1 7.7.1 7.7.1 7.8.1

7.6.1 7.6.1 7.6.1 7.7.1

7.6.1 7.6.1

8.1.1 8.1.1 8.1.1 8.1.1 8.1.1

FORCE10-BGP Force10 BGP MIB 4-V2-MIB (draft-ietf-idr-bgp4-mibv2-05) FORCE10-FIB- Force10 CIDR Multipath Routes MIB (The IP MIB Forwarding Table provides information that you can use to determine the egress port of an IP packet and troubleshoot an IP reachability issue. It reports the autonomous system of the next hop, multiple next hop support, and policy routing support) FORCE10-CSCHASSIS-MIB Force10 C-Series Enterprise Chassis MIB

7.5.1 7.6.1 7.6.1 7.6.1 8.1.1

FORCE10-IF-E Force10 Enterprise IF Extension MIB (extends XTENSION-MI the Interfaces portion of the MIB-2 (RFC B 1213) by providing proprietary SNMP OIDs for other counters displayed in the show interfaces output) FORCE10-LIN KAGG-MIB FORCE10-CH ASSIS-MIB Force10 Enterprise Link Aggregation MIB Force10 E-Series Enterprise Chassis MIB

7.6.1

7.5.1

8.1.1 8.1.1 8.1.1

FORCE10-COP Force10 File Copy MIB (supporting SNMP Y-CONFIG-MI SET operation) B FORCE10-MO N-MIB Force10 Monitoring MIB

7.7.1

7.7.1

7.6.1

7.5.1

8.1.1

Standards Compliance | 827

www.dell.com | support.dell.com

Network Management (continued)

FTOS support, per platform RFC# Full Name S-Series 7.6.1 7.6.1 7.6.1 7.6.1 7.5.1 7.5.1 8.1.1 8.1.1 C-Series 7.5.1 E-Series TeraScale E-Series ExaScale 8.1.1

FORCE10-PRO Force10 Product Object Identifier MIB DUCTS-MIB FORCE10-SSCHASSIS-MIB Force10 S-Series Enterprise Chassis MIB

FORCE10-SMI Force10 Structure of Management Information FORCE10-SYS Force10 System Component MIB (enables the TEM-COMPO user to view CAM usage information) NENT-MIB FORCE10-TCMIB Force10 Textual Convention

7.6.1 7.6.1

7.5.1 7.5.1

8.1.1 8.1.1

FORCE10-TRA Force10 Trap Alarm MIB P-ALARM-MI B

828

Standards Compliance

MIB Location
Force10 MIBs are under the Force10 MIBs subhead on the Documentation page of iSupport: https://www.force10networks.com/csportal20/KnowledgeBase/Documentation.aspx You also can obtain a list of selected MIBs and their OIDs at the following URL: https://www.force10networks.com/csportal20/MIBs/MIB_OIDs.aspx Some pages of iSupport require a login. To request an iSupport account, go to: https://www.force10networks.com/CSPortal20/Support/AccountRequest.aspx If you have forgotten or lost your account information, contact Dell Force10 TAC for assistance.

Standards Compliance | 829

830 www.dell.com | support.dell.com

Standards Compliance