Corero SmartWall TDD Getting Started Guide (ESXi)
Corero SmartWall TDD Getting Started Guide (ESXi)
26 August 2022
If you are a United States government agency, this documentation and the software described herein are provided to you subject to the following:
This paragraph applies to all acquisitions of the software by or for the United States Government, or by any prime contractor or subcontractor (at any tier) under
any contract, grant, cooperative agreement or other activity with the United States Government (collectively, the “Government”). All technical data and
computer software are commercial in nature and developed solely at private expense. The software and documentation respectively are “commercial computer
software” and “commercial computer software documentation” as defined in DFARS 252.227-7014 (June 1995) and “commercial items” as defined in FAR 2.101
(a) and, to the maximum extent permitted by law, are provided with only such rights as are provided in Corero’s standard commercial license for the software
and documentation and this notice. Technical data is provided with limited rights only as provided in DFAR 252.227-7015 (November 1995) or FAR 52.227-14
(June 1987), whichever is applicable. Corero’s standard commercial license for the software and documentation and this notice shall govern the Government’s
use of the software, documentation, and technical data, and shall supersede any conflicting contractual terms or conditions. If these terms and conditions fail to
meet the Government’s needs or is inconsistent in any respect with Federal law, the Government must return the software and the documentation unused to
Corero. The following additional statement applies only to acquisitions governed by DFARS Subpart 227.4 (October 1988): “Restricted Rights – Use, duplication
and disclosure by the Government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at
DFARS 252.227-7013 (OCT. 1988).” The Contractor is Corero Network Security, Inc.
You agree not to remove or deface any portion of any legend provided on any licensed program or documentation contained in, or delivered to you in
conjunction with, this document.
No part of this documentation may be reproduced in any form or by any means or used to make any derivative work (such as translation, transformation, or
adaptation) without written permission from Corero.
The products described in this document are protected by US Patent No. 9,442,782, US Patent No. 10,341,364, and European Patent No. 1319296.
Any software on removable media described in this documentation, is furnished under a license agreement which is located on the Corero web site.
Corero®, First Line of Defense®, SecureWatch®, and SmartWall® are registered trademarks of Corero Network Security, Inc. All other trademarks and registered
trademarks are the property of their respective holders.
Contents 3
TDD Documentation 9
Core Concepts 14
Policy 14
Detection Profiles 14
CONTENTS
Clusters 14
Devices 14
Segments 14
Licenses 14
Analytics 15
Sampled Traffic 15
Telemetry 15
NETCONF 16
Deployment Checklist 17
Required information 25
System requirements 26
Host requirements 26
Additional Requirements 28
Setting the Inbound Sample Rate between the vNTD and the CMS 50
Prerequisites 50
Prerequisites 58
Prerequisites 68
To configure a Juniper Networks PTX Series router for use with the
SmartWall TDD system 69
Prerequisites 79
To adjust the Inbound Sample Rate between the vNTD and the CMS 80
Troubleshooting 84
Remote Device added to the CMS Devices table instead of the SWA 86
Remote Device Info table (System > Health) is showing warning against
new router 87
Traffic is entering the network, but the Defense device does not seem to
do anything with it 88
Snapshotting VMs 90
After restarting my server, the Corero applications haven't come back up90
Requesting Licenses 92
Isolate the sampled traffic NICs for PCI Passthrough and optimize host 93
Note: The SmartWall TDD User Guide available from inside the SWA and CMS User Guide
available from inside the CMS contain additional information compared to the versions of the
guides available on the Juniper Support Portal. This information is only available to customers
and is not publicly accessible.
l Remote Devices – The Juniper Networks MX or PTX Series router at the edge of the network being protected.
They send sampled traffic to the vNTD and are directed by vSWA to apply firewall filters to block DDoS attack
traffic.
l Defense Director – A bundle of three virtual applications:
l vSWA – The SmartWall SecureWatch Analytics Virtual Edition (vSWA) receives information from the
Detection Engine (via the vCMS) to identify the DDoS attacks currently active against your network. The
vSWA application then sends firewall filter commands to the router to filter the attack traffic as it
arrives at the router. The vSWA application also displays real-time and historical statistics that enable
you to analyze attacks on your network.
l vCMS – The SmartWall Central Management Server Virtual Edition (vCMS) controls the
Detection Engine and enables you to configure the attack mitigation policy used to distinguish attack
traffic from normal network traffic.
l Detection Engine (vNTD) – The SmartWall Network Threat Defense Virtual Edition (vNTD) is the Detec-
tion Engine for the SmartWall TDD. It detects DDoS attack traffic in mirrored samples sent from the
edge routers.
l Additional Detection Engines – The Defense Director bundle includes a single Detection Engine (vNTD). You
may need to purchase additional Detection Engines for your deployment.
Your Juniper Networks MX Series routers must meet the following criteria:
l It must support Sampled Mirror, Flexible Filtering, Ephemeral Configuration, and Remote Telemetry.
l Your router should be running one of the following Junos versions:
l junos:17.3R3-S8
l junos:18.2R3-S5
l junos:18.3R3-S2
l junos:18.4R2-S7
l junos:18.4R3-S2
l junos:19.1R3-S5
l junos:19.2R3
l junos:19.3R3
l junos:19.4R3
l junos:20.1R2
l junos:20.2R2
l junos:20.2R3
l junos:20.3R1
l junos:20.3X75-D10
l evo:20.1R2-EVO
l evo:20.2R2-EVO
Caution: Some JunOS versions not listed above will appear to work initially, but contain an
incompatibility which will degrade SmartWall TDD performance over time and then stop the
SmartWall TDD from mitigating attacks. Please use one of the versions above, or contact your
support representative to discuss the best version for your system.
Your Juniper Networks PTX Series routers must meet the following criteria:
l PTX10001-36MR
When you install a SmartWall device or application, you need to execute essential configuration tasks using the
Corero Provisioning Command Line Interface (pCLI). The pCLI is a set of commands you can use to define the initial
configuration of each SmartWall component. For initial configuration of any component, type setup in the pCLI to
launch a wizard which will guide you through the initial configuration options.
Policy
A Policy is a configuration of the attack mitigation features which tells the Defense devices how to handle incoming
traffic. Each policy is contained in a Detection Profile.
Detection Profiles
A Detection Profile is a container for a configuration of the attack mitigation features (Policy) in the CMS. When you
associate a Detection Profilewith a Cluster, it provides all the Defense devices in that Cluster with the same Policy for
handling incoming traffic. You can create one Detection Profile for your network or multiple Detection Profiles each
containing a different Policy.
Clusters
A Cluster is a set of identically configured Defense devices. When you create a new Cluster you must associate it with
a Detection Profile containing the Policy which controls how the devices in that Cluster respond to traffic.
Devices
l Defense devices – This is broader term for the vNTDs (SmartWall Network Threat Defense Virtual Edition
devices) which are used purely as Detection Engines in a SmartWall TDD deployment
l Remote Devices – This is a broader term for the Juniper Networks MX or PTX Series routers used to mitigate
DDoS attack traffic
Segments
A Segment is a set of 1 or 2 interfaces to which DDoS protection is applied. The first time you connect a Defense
device to the CMS, it identifies the available interfaces and records them as Segments. A vNTD has two available
interface ports which act as 2 single interface Detector Segments. If you don't require the second Segment, you can
disable it after deployment.
Licenses
Due to the different security measures protecting each license type, they must be requested and installed in
different ways.
Analytics
Analytics is the process of collecting and analyzing the event and system information generated by the Defense
devices. The Defense devices send analytics syslog messages to the CMS where that information is aggregated and
sent to SWA.
Sampled Traffic
This is a feed of a proportion of the traffic received by the Juniper Networks MX or PTX Series router ahead of any
mitigation. The vNTD uses this traffic to detect DDoS attacks, and enables the TDD system to generate the filter
instructions it sends to the Remote Device to block that attack traffic and permit non-attack traffic. For example, if
you have 1Tbps of traffic coming into a Remote Device, and a sample rate of 1:1000, the vNTD will see 1Gbps of
sampled traffic.
Telemetry
Telemetry is sent from the Juniper Networks MX or PTX Series router to the vSWA. It shows the network traffic
processed by the router including what was permitted or blocked by the TDD system.
The TDD requires a telemetry feed from every monitored router to the SWA application. There are two main
telemetry delivery methods:
l Native telemetry (UDP) – (Not supported by Juniper Networks PTX Series router) Telemetry is sent over
your traffic network between the router and SWA. This requires a non-management interface on the router
and on the SWA host unit.
l gRPC telemetry – Telemetry is sent over the 1GB Management network. With gRPC telemetry you have the
option to encrypt the telemetry traffic using SSL certificates.
You decide which telemetry type is used when you configure the Juniper Networks MX or PTX Series router. For a
Juniper Networks PTX Series router, or if you choose gRPC telemetry for a Juniper Networks MX Series router, you
must download 2 additional software files to the router during set up and then provide additional configuration
information when adding the router to the SWA as a remote device.
The TDD system uses NETCONF to configure the ephemeral firewall rules in the Juniper Networks MX or PTX Series
router to block or permit network traffic.
The SmartWall Service Portal enables you to offer Corero SmartWall DDoS Protection, as a managed service, to your
customers. The Service Portal is a customer-facing DDoS protection portal which uses traffic data from your
SmartWall TDD and displays the information in easy to read dashboards and reports. Your customers can log in to
the portal and view the attacks you have protected them against. For information on Service Portal versions which
are compatible with your SmartWall TDD, see the SmartWall TDD release notes.
Note: If you do not have a Service Portal and would like to add one to your existing TDD system,
contact your support representative for more information.
On your PC Save the TDD license file you received when you purchased the TDD.
On the host server Setup host server(s) and make any network changes.
Deploy vSWA and use the pCLI to setup management and secondary
On the host server
network interfaces.
On the host server Deploy vCMS and use the pCLI to setup management interface.
Deploy vNTDs (or high sampled traffic rate vNTDs) and use the pCLI to setup
On the host server
management interfaces.
Configure routers to send traffic samples to the vNTDs and receive filters
On router CLI
from SWA.
On your PC: CMS CLI Configure default defense policy for TDD system.
On your PC: CMS web UI Tune Policy for your network and, optionally, setup multiple Clusters and
and SWA web UI Protection Profiles.
Caution: When the TDD system is installed, it is recommended to set it to Monitor Mode (the
CMS Defense Mode which shows how the system will affect traffic when mitigating but does not
block any traffic). When you have evaluated and tuned the CMS Policy for your network traffic,
you can then switch the system into Mitigate Mode and begin blocking DDoS attack traffic.
l You may want to have the IP addresses ready for any other syslog clients you intend to use.
l You must allow outbound TCP port 443 traffic to enable the TDD to connect to the Corero License server. To
establish licensing, you must install a SecureWatch package on the vSWA.
l To use Corero's SecureWatch service, you will need to allow outbound TCP port 443 traffic to the DNS address
geo.securewatch.network and destination IP addresses: 3.13.10.107, 35.155.55.233, 35.158.5.95 and
13.127.221.235. This enables monitoring and remote management by Corero's Security Operations Center.
l Port-Mirroring Traffic samples only: When hosting a SmartWall TDD, truncated samples should not be used
on the Juniper Networks MX or PTX Series router. The JunOS command set forwarding-options port-
mirroring maximum-packet-length should have a value of 0 (disabled).
l Port-Mirroring Traffic samples only: Port-mirroring samples should not been received on the same physical
port as telemetry or management traffic.
l Port-Mirroring Traffic samples only: The MTU path from routers to the SmartWall TDD should account for
encapsulations (e.g GRE). An MTU of 2000 is usually adequate.
Tip: To set the MTU to 2000, you should consult the VMWare documentation for your release
to locate the advanced setting required to edit the MTU.
Note: The traffic sampling method is configured during router configuration and when you add
the vNTD to the CMS.
When you're using a detector segment, the vNTD can accept traffic samples in the following ways. Using the CMS,
you need to configure the Segment differently, depending how your traffic samples arrived at the vNTD:
l Mirrored traffic samples direct from the router – The vNTD must be directly connected to the router with no
truncated samples. No additional configuration required on the CMS.
l Mirrored traffic samples over a GRE tunnel – The vNTD only needs to be accessible from the router. Edit the
Segment to add an IPv4 Address, a Peer IPv4 Address, and enable GRE Ingestion.
Caution: If you have multiple routers sending traffic samples in different formats, you must
ensure all sample rates are the same. Otherwise, the mitigation thresholds and traffic charts in
the SWA will not work correctly.
Corero recommends following best practice industry standards to secure your SmartWall deployment:
l Connect management interfaces to secure networks, isolated from the Internet, to prevent unauthorized
access to the SWA, CMS, or NTD.
l If practical, run Corero CMS and SWA virtual machines on a dedicated server to physically isolate them from
other guest virtual machines.
l If running CMS and SWA on shared virtual infrastructure, ensure you apply latest security fixes as they
emerge. This may require:
l Patching the server BIOS for new CPU firmware.
l Patching other guest VMs and applications that may be running on the same server as Corero man-
agement components.
l Limit access to the CMS and SWA to authorized personnel. Ensure adequate access controls are in place.
l Ensure authorized personnel are associated with the correct access roles on the CMS and NTD.
l Use strong access credentials, regularly changed, through authentication service such as LDAP or RADIUS to
authenticate access to your users.
l Limit knowledge of privileged account credentials.
l After deployment, change default NTD credentials and manage using Authentication Groups in the CMS.
Note: The same NTD credentials are used for managing NTDs from the CMS and for accessing the NTD pCLI.
l Apply the latest software updates from Corero which contain the latest available security patches.
Firewall Considerations
Your SmartWall deployment should be protected by your network infrastructure’s firewall. You will need to modify
your firewall policy to allow access on certain ports.
The following diagram shows the default network ports required for a full SmartWall TDD deployment:
The following diagram shows a common deployment scenario for the SmartWall TDD. If you require further
assistance configuring the best deployment for your network, contact your support representative.
Separately hosted TDD applications managing one Juniper Networks MX or PTX Series router
In this scenario, the three SmartWall TDD applications are all installed separately, on the same management network
as the router they are receiving sampled traffic from. The diagram shows the three possible routes for telemetry to
be passed from the router to the SWA.
In this scenario, the three SmartWall TDD applications are all installed on a single server and are receiving sampled
traffic from a single router. The diagram shows the three possible routes for telemetry to be passed from the router
to the SWA.
Caution: (Native Telemetry deployments only) The management port of the vSWA should NOT
be used to receive telemetry from the MX router. Telemetry should always be received on the
Secondary port as they may require changes to the receiving interface MTU. The management
port is used to access the vSWA and for sending TDD mitigation commands to the router. It is
strongly recommended to always set the management port MTU to 1500 bytes to avoid
interoperability issues. MTU settings are available via the pCLI setup network command.
Your Juniper Networks MX Series routers must meet the following criteria:
l It must support Sampled Mirror, Flexible Filtering, Ephemeral Configuration, and Remote Telemetry.
l Your router should be running one of the following Junos versions:
l junos:17.3R3-S8
l junos:18.2R3-S5
l junos:18.3R3-S2
l junos:18.4R2-S7
l junos:18.4R3-S2
l junos:19.1R3-S5
l junos:19.2R3
l junos:19.3R3
l junos:19.4R3
l junos:20.1R2
l junos:20.2R2
l junos:20.2R3
l junos:20.3R1
l junos:20.3X75-D10
l evo:20.1R2-EVO
l evo:20.2R2-EVO
Caution: Some JunOS versions not listed above will appear to work initially, but contain an
incompatibility which will degrade SmartWall TDD performance over time and then stop the
SmartWall TDD from mitigating attacks. Please use one of the versions above, or contact your
support representative to discuss the best version for your system.
Your Juniper Networks PTX Series routers must meet the following criteria:
l PTX10001-36MR
You should have the following files saved locally (where xxxx represents the variable section of the file name):
Required information
You need to have the following network information available for each virtual edition before you begin to install:
l IP address and subnet mask – The IP address you will use to access the application from your core network.
l DNS IP address – The address of the DNS server you will use for the site. You may have more than one of
these.
l Default gateway IP address – The address of the default gateway for your site.
l NTP IP address – The address of the NTP server you wish to use for the site. You may have more than one of
these. Corero can provide you with an NTP server address if you do not have one of your own.
For vCMS and vSWA, you also need A SecureWatch® package file and unlock code. To connect your CMS and SWA
applications to the SecureWatch® Service, you will need to upload a SecureWatch® package file to both applications
after deployment.
To enable you to plan your deployment, you should decide on the type of applications you require. For vNTD
applications, this depends on your expected traffic rates.
There are multiple ways to deploy a vNTD on your own server. If you expect a high sampled traffic rate, you must
dedicate hardware resources on the host for the vNTD. If you expect a lower sampled traffic level, you can speed up
your installation by skipping those steps.
Sampled Traffic rates up to 250 Kpps or 1.2Gbps per core NICs: VMXNET3
Note: With a sample rate of 1:1000, this equates to up to 1.2 Tbps of protected 2 core vNTD
traffic (or up to 250 million pps). No PCI Passthrough
Caution: For the vNTD External Interface, VMXNET3 is recommended. E1000 NICs are supported,
but do not support multiple cores. Additionally, you cannot use E1000 NICs for both
Management and Data interfaces as they don't maintain a consistent port order. Contact your
support representative for more information on choosing a NIC.
System requirements
Host requirements
Before you begin, make sure the server you plan to deploy an application on meets the following requirements:
l vNTD configured for high sampled traffic rates only: VT-d must also be enabled.
l Datastore:
l Redundant RAID storage.
l Eternal Interfaces:
l For low sampled traffic rates – Up to 4 VMXNET3 or E1000NICs. Note: E1000 NICs do not
support multiple cores and you cannot use E1000 NICs for both external interfaces and
the management interface.
l For high sampled traffic rates – Up to 2 X710 (recommended to use firmware 6.8 or
higher) or equivalent NICs (X520, X540, and E810 supported) utilizing VT-d directed I/O
technology (PCI Passthrough).
Note: For SR-IOV support, the External Interface NICs must be X710, X520 or X540.
Note: When listing the number of cores, this refers to the physical cores not the hyper-thread
cores. These are set to enable optimum performance of the SmartWall system.
vCMS
l Memory:
l Minimum 16 GB of memory.
l CPU – Intel® Xeon® 64-Bit CPU (x86_64) Processor minimum 2GHz. vCMS must have:
l 6 physical cores.
l Datastore – Required 120 GB datastore.
l Networking – 1 management interface and 1 optional secondary management interface. (See Host Require-
ments for NIC types required for these interfaces)
vSWA
l Disk 2 – Minimum 240 GB datastore. Disk 2 can be expanded to your required size after deployment (
ESXi method). The space required by the SWA depends on your frequency of attack and how long you
need the data to be available. A good rule of thumb for most installations is 1.5 TB for 13 months of
data.
l Networking – 1 management interface and 1 optional secondary management interface. (See Host Require-
ments for NIC types required for these interfaces)
vNTD
l Memory – Minimum 12GB per socket (with seven 1GB -or equivalent- huge pages available for high sample
traffic rates).
l CPU – Intel® Xeon® 64-Bit CPU (x86_64) Processor from E3 family or later, supporting AVX instructions (Sandy
Bridge architecture or later), with minimum 2GHz. vNTD must have the correct core number for its external
NIC type:
l VirtIO NIC – 3 physical cores
l X710 (recommended to use firmware 6.8 or higher) or equivalent NIC (X520, X540, and E810 sup-
Note: If you're using SR-IOV it's best practice to use less cores for the vNTD. If you have X710
NICs use 2,3,4, or 5 physical cores. If you have X520 or X540 NICs use 2,3,or 5 physical cores
Additional Requirements
space when the storage needs expand, you could experience failures. Caution: If you use Thin pro-
visioning, you must make sure there is ample room for the disks to reach their full size in the future.
l Thick Provision Lazy Zeroed (Recommended) – Allocates all required disk space to the VM, but only
prepares space as it is required.
l Thick Provision Eager Zeroed – Allocates all required disk space to the VM, and immediately prepares
it for the VM. This can take some time to complete.
l You need at least one management network connection with connectivity to SmartWall devices, CMS and
SWA. The SWA requires a secondary network connection on the same subnet as the router's telemetry port,
to receive native telemetry from the routers.
l All TDD applications must use NTP servers for system time. Any time difference between applications can
cause unexpected behavior.
l NUMA Memory/CPU Affinity is recommended as it can improve performance, but is not required.
The following instructions cover how to deploy a virtual edition on an ESXi server which meets the necessary system
requirements. See the Appendix for deployment instructions for vNTDs expecting high inspection rates.
Note: The following commands include the --powerOn command to start the VM. If you don't
want to start the VM at this point, you can use VMWare client at a later time.
vSWA or vCMS:
ovftool --name="<vmName>" --datastore=<datastoreName> --diskMode=lazyZeroedThick
–-net:Management="<managementNetworkName>" --net:Secondary="<secondaryNetwork>"
--acceptAllEulas --powerOn <CoreroFileName> vi://root@<hostName>
vNTD:
Caution: Never deploy multiple vNTDs with the same network labels (<External_1_
NetworkName> and <External_2_NetworkName>) as that can cause a network loop. If you
do not specify network labels, the vNTD is given the default External and Internal labels.
For a single VM, ensure the <External_1_NetworkName> and <External_2_
NetworkName> are not the same
After install, the vNTD network adapters are not automatically connected, in case the selected ports will cause a
network loop once enabled. Check the allocated ports will not create a network loop, then enable ports using
vSphere WebClient.
1. In vSphere WebClient, right click on your new VM and select Edit Settings.
2. Scroll down to Network adapter 2. Check the network label is correct and then under Device Status, select
the Connect check box.
3. Repeat for Network adapter 3.
4. Click OK.
1. In vSphere WebClient, right-click the Host you want to create this Virtual Appliance on, and select Deploy OVF
template.
2. Select an OVF template:
l Select URL and type the full URL to the OVA file stored on an HTTP/HTTPS server.
l Select Local file, click Choose Files and select the OVA file for your required VM.
3. Click Next.
4. Type a Virtual Machine name.
6. Click Next.
7. Select the destination compute resource you want to use for this VM. This will be the host where you want
the VM to be deployed.
14. From Select virtual disk format, select Thick Provision Lazy Zeroed.
15. Select the storage disk you want to use.
l For the vNTD you need the Management network and either 1 or 2 networks for the number of inter-
faces on your vNTD (in the example: External_1 and External_2). Caution: The two External networks
cannot be the same.
20. Click Finish. The VM is created powered off. Do not power it on yet.
21. vNTD only: Check the allocated ports will not create a network loop when enabled, then enable ports:
a. In vSphere WebClient, right click on your new VM and select Edit Settings.
b. Scroll down to Network adapter 2. Check the network label is correct and then under Device Status,
select the Connect check box.
c. Repeat for Network adapter 3.
d. Click OK.
The amount of space required by the SWA depends on your frequency of attack and how long you need the data to
be available. A good rule of thumb for most installations is 1.5 TB for 13 months of data. The default disk image for a
vSWA deployment is 240GB. You can choose to resize disk 2 after deployment.
Note: If your SWA was installed prior to 9.5.0, you will need to resize disk 3 rather than disk 2.
Once you have installed the SWA, CMS, and vNTD applications on a KVM host, you can use the Corero Virtual Edition
Installation Checker to verify that the three applications are correctly installed. The script checks the following areas:
l Huge Pages – Checks 1GB Huge Pages is enabled and there are 12 allocated per socket
l CPU Pinning – Required for high traffic rates. Checks to make sure all vNTD CPUs are pinned, that all
pinned CPUs are isolated, and that no isolated CPUs on the host are unpinned.
You documentation package should include the Corero Virtual Edition Installation Checker script and accompanying
PDF guide which explains how to run the checker and how to resolve any issues shown.
After deployment, you need to configure the virtual appliance. To do this you need to access the pCLI and run the
setup wizard:
2. Log in to the virtual appliance using the default user account (admin) and password (smartwall).
Tip: Once you have an IP address associated with the appliance, you can use SSH on port 2222
using the default admin user account (e.g. ssh -p 2222 admin@<ipAddress>) and type the
default password smartwall when prompted.
This section describes how to use the provisioning CLI (pCLI) to set each component's basic configuration. To access
the pCLI for each SmartWall® component see the application specific instructions earlier in this guide.
The pCLI provides the initial interface for configuring SmartWall™ components. Once you use the setup wizard to
configure the application, you should perform all other tasks in the corresponding web interfaces. You can return to
the pCLI if you need to edit these basic configuration settings later.
Tip: You can type help at the pCLI console command prompt to see a list of all the other
available pCLI commands and Tab-completion is supported, enabling you to type a portion of the
command that identifies it uniquely and press Tab to finish it.
Note: This example shows the use of the pCLI to run the setup wizard on the vCMS. Other setup
wizard sessions may vary slightly, depending on the component.
2. Type setup to start the initial setup wizard, which will prompt you to change a number of basic configuration
settings. For each group of settings, type A to accept changes you’ve made, type C to go back and change a
setting, or type E to leave the current settings unchanged.
cms>setup
You will be asked a number of questions, along with the current values
in brackets. Please type in the desired configuration or you may press
Enter to keep the current value. After each section, you will be given
a chance to confirm and apply your configuration.
3. Change the username and password from their defaults. This is strongly recommended. If you are configuring
SWA, you will be asked if you want to enable LDAP, as well.
Warning: Applying changes here will overwrite any users created in the
application
Secondary Interface:
State : Disabled
6. Configure the DNS settings: hostname, and the option for manual DNS configuration (DNS servers, DNS
domain, DNS search domains).
Configuration:
Hostname : cms
DNS Servers : None
DNS Domain : corero.com
DNS Search Domains :
Caution: You must use NTP servers for all TDD applications. Time differences between
applications can cause unexpected behavior.
Configuration:
Time source : Hypervisor
Time zone : America/New_York
cms>show
Management:
State : Enabled
Link State : Up
MAC Address : 00:50:56:88:62:82
MTU : 1500
DHCP Enabled : No
IPv4 Address : 10.20.27.100
Subnet Mask : 255.255.255.0
Default Gateway : 10.20.27.254
Secondary:
State : Disabled
Link State : Down
MAC Address : 00:50:56:88:20:ad
MTU : 1500
IPv4 Address : None
Subnet Mask : None
SecureWatch Status:
State : Disabled
SecureWatch ID : No package loaded
cms>
SSL Certificates
The pCLI can be used to load a signed SSL certificate for use with the web UI, to avoid the browser security warnings
which appear when you sue the default unsigned certificate. The certificate must be in PKCS#12 format, and include
the key, certificate, and CA certificate change to be used for SSL. The common name should match the hostname
assigned to the SWA appliance.
To load a certificate, type ssl-certificates https followed by the URI to the certificate file. The supported
protocols are FTP, SFTP, HTTP, and HTTPS. For example:
You must install a TDD license file in the SWA application as it enables the SWA to check you are licensed to use the
TDD system. The license file is provided in a SecureWatch Package format which is compatible with the SWA
application upload process.
Note: To use Corero's SecureWatch service, you will need to allow outbound TCP port 443 traffic
to the DNS address geo.securewatch.network and destination IP addresses: 3.13.10.107,
35.155.55.233, 35.158.5.95 and 13.127.221.235. This enables monitoring and remote
management by Corero's Security Operations Center.
Corero offers three forms of licensing for SmartWall TDD depending on your environment and service level choice:
l Corero SecureWatch Service customers – The license is delivered via the SecureWatch package file that is
delivered to connect to the SecureWatch Service via a secure VPN. Install the SecureWatch package file and
the licensing process does not require any further steps.
l Customers who are not connecting to Corero SecureWatch Service and have internet access available to the
SWA – With this option, the license is delivered through a SecureWatch package file that points to an internet
based license server. Install the SecureWatch package file and the SWA licensing process will contact the inter-
net based license server.
Note: The SecureWatch License File does not create a persistent connection to the license
server. It only requires a daily check. No data is ever sent over that connection.
l Customers who are not connecting to the SecureWatch Service and the SWA does not have access to the inter-
net – Please contact your Corero customer support representative for more information on disconnected
licensing.
1. You will receive the SecureWatch package and unlock code from Customer Support.
2. Save this file in a location that you can easily access from the computer you're using to access the SWA web
UI.
3. Open the SWA Web UI in a browser.
4. Navigate to System > SecureWatch Packages.
5. Click Choose File and select the saved package file.
6. If required, type in the Unlock Code.
7. Click Install Package. This will cause the SWA to restart and may take several minutes. When it's complete you
can log back in.
If your internet connectivity is through an HTTP Proxy then you will need to setup the SWA to enable the external
connection. This will differ depending on your chosen licensing method.
l SecureWatch Service VPN – The HTTP Proxy is setup and enabled in SecureWatch Packages:
1. Open the SWA Web UI in a browser.
2. Navigate to System > SecureWatch Packages
3. Check the box to Enable SecureWatch HTTP Proxy.
4. Type the IP Address of the HTTP Proxy Server.
5. Type the Port number for the HTTP Proxy port on your server. The default port on most servers is 3128.
6. (Optional) Configure authentication:
a. Select an Authentication Type : None, Basic or NTLM
b. Type the authentication Username and Password for the HTTP Server.
7. Click Save.
l SecureWatch License File – The HTTP Proxy is setup and enabled in General Settings:
1. Open the SWA Web UI in a browser.
2. Navigate to System > General Settings
3. Under HTTP Proxy, click the Enable slider. It should now be green, with the slider button to the right.
4. Type the IP Address of the HTTP Proxy Server.
5. Type the Port number for the HTTP Proxy port on your server. The default port on most servers is 3128.
6. Click Save. This may cause the SWA to restart.
For each vNTD you want to add to your CMS, you need a 10G license. After your vCMS is deployed and configured,
you can contact support for a license file unique to your system.
1. In the CMS Web UI, you can see your CMS's UUID at the top of the Home screen.
2. Contact your support representative to request a vNTD license file . You must quote your CMS UUID and your
Juniper SSRN (the SSRN will be on your sales agreement information).
Caution: License files are created to be specific to your CMS and cannot be transferred.
3. Save the license file somewhere you can easily access, from the computer you're using to access the CMS web
UI.
4. Open the CMS application in a browser and log in. (If you have not yet changed them, the default user-
name/password is admin/smartwall.
5. Navigate to System > Licensing.
6. Click Add.
7. You can either:
l Copy and paste the contents of the license file into the text box. You must include the license header
The Port-Mirror Sample Rate you will configure on the router (and the Ingress Sample rate you configure when
adding a vNTD to the CMS) controls how many samples are sent from the router to the vNTD. In addition to this, you
need to control the rate of samples sent from the vNTD to the CMS. The default setting in a CMS (1 in 1999) is not
suitable for a TDD deployment and must be adjusted.
l SmartWall TDD lab test system using smaller traffic volumes and attacks (from a traffic generator) –
Change the value to 1. This samples every packet received by the Defense device.
5. If you want to save the new configuration, and push your changes to any affected Defense devices, click
. Then, on the pop-up dialog, click Commit to push the changes (alternatively, you can click Discard
to discard any uncommitted changes).
Note: The default sample rates given above are correct for the majority of deployments and you
should make sure the Port-Mirror Sample Rate on your router is correct before adjusting the
Inbound Sample Rate.
You must add a SmartWall Defense device to the CMS before you can manage the attack mitigation Policy on that
device. The Defense devices are managed in a Cluster. You must edit this Cluster to expect devices which are using
sampled traffic.
Prerequisites
Note: The Port-Mirror Sample Rate (configured on the router) and Ingress Sample rate should be
identical. Both must be the rate factor which reduces the amount of traffic seen by the router to
a manageable size for the vNTD. A default value of 1000 would normally scale to approx
6. Click . Then, on the pop-up dialog, click Commit to push the changes.
11. Click . Then, on the pop-up dialog, click Commit to push the changes.
12. If you have already uploaded your vNTD license, your devices will have been automatically licensed when they
deployed. If you haven't uploaded the vNTD license, you should do so now following the method above and
then manually apply the license to each vNTD:
a. In the CMS, use the left-hand menu to navigate to Network > Devices.
b. On the Devices table, locate the vNTD you want to license.
c. In the Actions column, click and select License.
Note: Adding a device to the CMS does not cause any existing Policy to be pushed to it. You
must first add the device to a Cluster. To learn more about creating Clusters and managing
Policy, see the SmartWall Central Management Server User Guide.
After you've connected your vNTD to the vCMS, you can view information on the available interfaces on the device.
By default the vNTD comes with two available Segments. If you only require one Segment, you can disable the
additional segment.
a. Set the External IPv4 Address to the IP address of the external interface on the vNTD (for ter-
mination this is the tunnel endpoint)
b. Set the External Peer IPv4 Address to the IP address of the interface which is the last hop before
the traffic arrives at the vNTD (e.g the interface on the router which has received the sampled
traffic and is connected to the vNTD)
c. Set the GRE Ingestion drop-down to enabled.
8. Click Save.
9. If you want to save the new configuration, and push your changes to any affected Defense devices, click
. Then, on the pop-up dialog, click Commit to push the changes (alternatively, you can click Discard
to discard any uncommitted changes).
By default, SWA listens for syslog messages on port 9997. You will need to configure CMS to send syslog messages to
SWA on this port.
1. Open the CMS application in a browser and log in. (If you have not yet changed them, the default user-
name/password is admin/smartwall.
2. Use the left-hand menu to navigate to System > Analytics & Syslog. Make sure the SERVERS tab is selected.
3. At the Analytics Servers table, click Add Server.
4. Type a Name for this server. You must only use alphanumerics, spaces, or .-&()_/@:= symbols.
5. Type the IP Address of the server (or its DNS name).
6. Enable or Disable Encryption for this server. The CMS and SWA come with self-signed SSL certificates. You can
choose to upload signed certificates to the CMS and SWA- see optional steps below.
7. Leave the default Port: 9997 for unencrypted or 9998 for encrypted connections.
8. Click Save.
9. Click . Then, on the pop-up dialog, click Commit to push the changes.
If you enable encryption, the connection between the CMS and SWA uses, by default, an in-built self-signed
certificate. If you want to use a signed certificate, you need to upload a PKC#S12 certificate to both sides of the
connection.
i. Click . Then, on the pop-up dialog, click Commit to push the changes.
2. Add a signed certificate to the part of the SWA that receives information from CMS:
a. Access the SWA pCLI:
l Open a console session. On an ESXi server, you can use VMware (select the VM and click Open
Console) or on a KVM server you can use virsh (command: virsh console <vmName>).
l SSH to the pCLI: ssh -p 2222 admin@<ipAddress>
b. Log in. If you haven't yet changed them, the default username and password is admin/smartwall.
c. To load a certificate, type ssl-certificates forwarder followed by the URI to the PKCS#12
format certificate file. The supported protocols are FTP, SFTP, HTTP, and HTTPS. For example: ssl-cer-
tificates forwarder sftp://[email protected]/certs/my_cert.p12
d. You will be prompted for a password to access the file location. If you password protected the PKCS#12
file, you will also be prompted for that password.
You must add a set of CMS admin credentials to enable the SWA to communicate mitigation changes back to the
CMS.
The CMS comes with a default self-signed Corero SSL certificate which your browser will list as "not secure". As soon
as possible, you should replace this with a signed certificate. This must be packaged in pkcs12 format and can
optionally be password protected.
1. Open the CMS application in a browser and log in. (If you have not yet changed them, the default user-
name/password is admin/smartwall.
2. Navigate to System > HTTPS.
3. Click Upload Certificate.
4. Select the pkcs12 certificate file on your computer, and click Open.
5. (Optional) Type in the Password for the certificate file.
6. Click OK.
7. Refresh the browser to view the new security rating.
Note: The certificate must be in PKCS#12 format, and include the private key, signed certificate,
and CA certificate change to be used for SSL. The common name should match the hostname
assigned to the SWA appliance.
Prerequisites
The configuration provided below assumes you have a new system with initial configuration already setup for your
network. At a minimum, the system must be licensed, have SSH enabled, be connected to your management
network and have a host name set on the router.
Caution: The hostname MUST be enabled on the forwarding plane of the Juniper Networks MX
Series router to correctly forward telemetry to the SmartWall TDD. If you need to set a
hostname (set system host-name <name>), you MUST restart the router afterward for the
change to take effect.
For gRPC telemetry only: You must download two additional software files. Contact your support representative for
assistance.
l network-agent-x86-32-17.4R1.16-C1.tgz
l junos-openconfig-x86-32-0.0.0.9.tgz
l <MX_ExternalInterface_Name> – The names of the external interfaces on your router which you want to
protect with the TDD system (e.g. xe-0/0/0).
l <MX_SampleTrafficInterface_Name> – The name of the interface on your router which you have alloc-
ated for sending sample traffic to the vNTD (e.g. xe-0/0/2).
l <MX_SampleTrafficInterface_IPv4_Subnet> – The IPv4 address of the interface on your router which
you have allocated for sample traffic, formatted as a CIDR (e.g. 192.168.66.201/24).
l <MX_SampleTrafficInterface_IPv6_Subnet> – The IPv6 address of the interface on your router which
you have allocated for sample traffic, formatted as a CIDR (e.g. 2000:2000:cccc:cccc::1/64).
l <MX_TelemetryInterface_Name> – (Native telemetry only) The name of the interface on your router
which you have allocated for sending telemetry to the vSWA (e.g. xe-0/0/3).
l <MX_TelemetryInterface_IP> – (Native telemetry only) The IPv4 address of the interface on your router
which you have allocated for telemetry (e.g. 192.168.99.201).
l <vNTD_ExInterface_MAC> – The MAC address of the external interface on your vNTD (e.g.
00:0c:29:36:94:5a). To find this, log into the vNTD pCLI and type the command show nic and look for
the MAC address under External.
l <vNTD_ExInterface_IPv4> and <vNTD_ExInterface_IPv6> – The IPv4 and IPv6 address you want to
allocate to the external interface on the vNTD (e.g. 192.168.66.105 and 2000:2000:cccc:cccc::2).
l For sampled traffic sent via port-mirroring and port mirror instances, this must be in the same subnet
as the interface on the router which you allocated for mirroring (<MX_SampleTrafficInterface_
Name>).
l For sampled traffic sent via a GRE tunnel, this must be the IP address you configured on the external
interface on the vNTD when you configured the Segment.
l <vSWA_SecondaryInterface_IP> – (Native telemetry only) The IP address of the secondary interface on
the vSWA (e.g. 192.168.99.113). You can configure this IP address by logging into the vSWA pCLI and using
the setup network wizard. The IP must be on the same network as the interface on the router which you alloc-
ated for telemetry (MX_TelemetryInterface).
To configure a Juniper Networks MX Series router for use with the SmartWall TDD system
Caution: If copying and pasting from the PDF, you can experience some loss of characters. If
possible, use the online help version of the documentation. Alternately, first copy this set of
commands into a plain text word processor (e.g. notepad) and check none of the hyphens or
spaces have been removed and that no additional returns have been added.
The following steps are required to configure a Juniper Networks MX Series router:
1. All methods below, unless stated, are performed in the configuration area of the router. To access the router
configuration area:
a. Open the Juniper Networks MX Series router CLI using an SSH client: ssh <username>@<ipad-
dress>
b. Enter your password to log in.
c. Enter configuration mode: configure
d. Use the example commands in each section to input your required configuration changes. Each line is
an individual command- paste a single line and press Enter before moving on to the next line.
2. Once you have performed all the commands in a section, you must commit the changes and exit configuration
mode:
a. To save your changes: commit
b. Exit configuration mode: exit
Caution: If you have multiple routers sending traffic samples in different formats, you must
ensure all sample rates are the same. Otherwise, the mitigation thresholds and traffic charts in
the SWA will not work correctly.
There are multiple methods for sending traffic samples to your vNTD. Select one of the following methods:
Note: The 1000 value you enter, in the first command in step 2, is the rate factor which reduces
the amount of traffic seen by the router to a manageable size for the vNTD (10Gbps or less). 1 in
every 1000 packets is sampled and sent to the vNTD. This enables you to sample from up to
10Tbit/sec of traffic for each vNTD. (Note: the sample rate assumes a run-length of 0). This is the
rate required by the vNTD.
The deployment instructions above use port-mirroring, set up as part of the CORERO-MITIGATE filter. If you need the
flexibility of using a port mirror instance outside of the CORERO-MITIGATE filter, you can use the following
alternative commands. The <MirrorFilter> and <MirrorFilter_IPv6> values must be replaced with the name
of the IPv4 and IPv6 filters you want to use to mirror traffic samples to the vNTD. The <instanceName> must be
replaced with the port mirror instance you want to use.
1. Identify an interface on the router which can be used for sending mirrored traffic to the vNTD. Set and label
this interface (using the description field):
set interfaces <MX_SampleTrafficInterface_Name> description Sample_Port_to_vNTD
2. Configure Port-Mirroring to forward a sample rate of traffic from the router to the vNTD. When you con-
figured the vNTD to accept Port-Mirroring samples, you did not have to set an IP address on the external inter-
face. In the commands below, the vNTD_ExInterface variables can be dummy IP addresses (in the same
subnet as the allocated sample forwarding interface on the router), as you are creating a static arp to the
vNTD MAC address.
set forwarding-options port-mirroring input rate 1000
set forwarding-options port-mirroring input run-length 0
set forwarding-options port-mirroring family inet output interface <MX_SampleTraf-
ficInterface_Name> next-hop <vNTD_ExInterface_IPv4>
set forwarding-options port-mirroring family inet6 output interface <MX_SampleTraf-
ficInterface_Name> next-hop <vNTD_ExInterface_IPv6>
set interfaces <MX_SampleTrafficInterface_Name> unit 0 family inet address <MX_
SampleTrafficInterface_IPv4_Subnet> arp <vNTD_ExInterface_IPv4> mac <vNTD_ExIn-
terface_MAC>
set interfaces <MX_SampleTrafficInterface_Name> unit 0 family inet6 address <MX_
SampleTrafficInterface_IPv6_Subnet> ndp <vNTD_ExInterface_IPv6> mac <vNTD_ExIn-
terface_MAC>
5. When you add this router to the SWA as a Remote Device, you must give it the device type ST_MX (rather
than the standard device type MX).
On the router a GRE tunnel needs to be configured from the router to the vNTD. Then you need to setup Port-
Mirroring to send the samples over this tunnel. The following placeholders are specific to this GRE tunnel method:
l <MX_GRETunnel_Name> – The GRE Tunnel (<MX_GRETunnel_Name>) must be named using the following
format where n is replaced with unique values for this GRE tunnel: gr-n/n/n (e.g. gr-0/0/0 if you have no
other tunnels by that name).
l <MX_IPv4> – One of the IP addresses associated with your router.
Note: If you require an alternative logical tunnel type, contact your support representative.
Note: The GRE tunnel interface requires an IP address in order for next-hop in step 2 to be
recursively valid. Dummy, non-routable, subnet examples (e.g. 10.99.100.0/31) are provided
in the commands below and can be left or modified as required.
1. Unless already enabled, you must enable the router to accept NetConf . This enables the TDD to push con-
figuration to the router:
set system services netconf ssh port 830
Note: If you want to use a custom NetConf port, you can replace 830 with your required port
number. You must also add the custom port number to the Remote Devices table entry for this
router.
2. Create an ephemeral instance of the configuration database named Corero and limit it to only store the last
500 commits in memory. This is where configuration from the TDD is sent.
set system configuration-database ephemeral instance Corero
set system configuration-database ephemeral purge-on-version 500
There are two options for configuring telemetry. Choose the method below which works for your network:
1. Identify an interface on the router (different from the sample traffic interface) which can be used for sending
telemetry to the vSWA. Set and label this interface (using the description field):
set interfaces <MX_TelemetryInterface_Name> description Interface_for_telemetry
2. Configure the telemetry sent to the SWA:
set services analytics streaming-server Corero remote-address <vSWA_Sec-
ondaryInterface_IP>
set services analytics streaming-server Corero remote-port 30000
Tip: If you need to add a new certificate to this router, you can use the following command:
set security certificates local <certificateName> load-key-file <URL>
Note: If your router is using Input Lists, you should replace the last command with the following:
set services analytics sensor Corero-FF-mitigate resource-filter "^
[^_].*"
Tip: If you need to add a new certificate to this router, you can use the following command:
set security certificates local <certificateName> load-key-file <URL>
1. Configure role-based access to routers by creating a new login class called TDD and a user within that class
called corero.
set system login user corero class TDD
set system login user corero authentication plain-text-password
set system login class TDD permissions configure
set system login class TDD permissions view
set system login class TDD permissions firewall-control
Note: When you add your routers to the Remote Devices table in SWA, you must enter the
username corero and the same password you configured here.
Prerequisites
The configuration provided below assumes you have a new system with initial configuration already setup for your
network. At a minimum, the system must be licensed, have SSH enabled, be connected to your management
network and have a host name set on the router.
Caution: The hostname MUST be enabled on the forwarding plane of the Juniper Networks PTX
Series router to correctly forward telemetry to the SmartWall TDD. If you need to set a
hostname (set system host-name <name>), you MUST restart the router afterward for the
change to take effect.
Before you begin, you also need to know the following information which you will use to replace the placeholders in
the commands below:
l <PTX_ExternalInterface_Name> – The names of the external interfaces on your router which you want
to protect with the TDD system (e.g. xe-0/0/0).
as the interface on the router which you allocated for mirroring (<PTX_SampleTrafficInterface_
Name>).
l For sampled traffic sent via a GRE tunnel, this must be the IP address you configured on the external
interface on the vNTD when you configured the Segment.
To configure a Juniper Networks PTX Series router for use with the SmartWall TDD system
Caution: If copying and pasting from the PDF, you can experience some loss of characters. If
possible, use the online help version of the documentation. Alternately, first copy this set of
commands into a plain text word processor (e.g. notepad) and check none of the hyphens or
spaces have been removed and that no additional returns have been added.
The following steps are required to configure a Juniper Networks PTX Series router:
1. Transport sampled packets from the router to the vNTD. Select one of the methods provided:
l Using port mirror sampling over a layer 2 direct connection
1. All methods below, unless stated, are performed in the configuration area of the router. To access the router
configuration area:
a. Open the Juniper Networks PTX Series router CLI using an SSH client: ssh <username>@<ipad-
dress>
b. Enter your password to log in.
c. Enter configuration mode: configure
d. Use the example commands in each section to input your required configuration changes. Each line is
an individual command- paste a single line and press Enter before moving on to the next line.
2. Once you have performed all the commands in a section, you must commit the changes and exit configuration
mode:
a. To save your changes: commit
b. Exit configuration mode: exit
Caution: If you have multiple routers sending traffic samples in different formats, you must
ensure all sample rates are the same. Otherwise, the mitigation thresholds and traffic charts in
the SWA will not work correctly.
There are multiple methods for sending traffic samples to your vNTD. Select one of the following methods:
Note: The 1000 value you enter, in the first command in step 2, is the rate factor which reduces
the amount of traffic seen by the router to a manageable size for the vNTD (10Gbps or less). 1 in
every 1000 packets is sampled and sent to the vNTD. This enables you to sample from up to
10Tbit/sec of traffic for each vNTD. (Note: the sample rate assumes a run-length of 0). This is the
rate required by the vNTD.
1. Identify an interface on the router which can be used for sending mirrored traffic to the vNTD. Set and label
this interface (using the description field):
set interfaces <PTX_SampleTrafficInterface_Name> description Sample_Port_to_vNTD
The deployment instructions above use port-mirroring, set up as part of the CORERO-MITIGATE filter. If you need the
flexibility of using a port mirror instance outside of the CORERO-MITIGATE filter, you can use the following
alternative commands. The <MirrorFilter> and <MirrorFilter_IPv6> values must be replaced with the name
of the IPv4 and IPv6 filters you want to use to mirror traffic samples to the vNTD. The <instanceName> must be
replaced with the port mirror instance you want to use.
1. Identify an interface on the router which can be used for sending mirrored traffic to the vNTD. Set and label
this interface (using the description field):
set interfaces <PTX_SampleTrafficInterface_Name> description Sample_Port_to_vNTD
2. Configure Port-Mirroring to forward a sample rate of traffic from the router to the vNTD. When you con-
figured the vNTD to accept Port-Mirroring samples, you did not have to set an IP address on the external inter-
face. In the commands below, the vNTD_ExInterface variables can be dummy IP addresses (in the same
subnet as the allocated sample forwarding interface on the router), as you are creating a static arp to the
vNTD MAC address.
set forwarding-options port-mirroring input rate 1000
set forwarding-options port-mirroring input run-length 0
set forwarding-options port-mirroring family inet output interface <PTX_SampleTraf-
ficInterface_Name> next-hop <vNTD_ExInterface_IPv4>
set forwarding-options port-mirroring family inet6 output interface <PTX_
SampleTrafficInterface_Name> next-hop <vNTD_ExInterface_IPv6>
set interfaces <PTX_SampleTrafficInterface_Name> unit 0 family inet address <PTX_
SampleTrafficInterface_IPv4_Subnet> arp <vNTD_ExInterface_IPv4> mac <vNTD_ExIn-
terface_MAC>
set interfaces <PTX_SampleTrafficInterface_Name> unit 0 family inet6 address <PTX_
SampleTrafficInterface_IPv6_Subnet> ndp <vNTD_ExInterface_IPv6> mac <vNTD_ExIn-
terface_MAC>
5. When you add this router to the SWA as a Remote Device, you must give it the device type ST_PTX (rather
than the standard device type PTX).
On the router a GRE tunnel needs to be configured from the router to the vNTD. Then you need to setup Port-
Mirroring to send the samples over this tunnel. The following placeholders are specific to this GRE tunnel method:
l <PTX_GRETunnel_Name> – The GRE Tunnel (<PTX_GRETunnel_Name>) must be named using the following
format where n is replaced with unique values for this GRE tunnel: gr-n/n/n (e.g. gr-0/0/0 if you have no
other tunnels by that name).
l <PTX_IPv4> – One of the IP addresses associated with your router.
Note: If you require an alternative logical tunnel type, contact your support representative.
Note: The GRE tunnel interface requires an IP address in order for next-hop in step 2 to be
recursively valid. Dummy, non-routable, subnet examples (e.g. 10.99.100.0/31) are provided
in the commands below and can be left or modified as required.
1. Unless already enabled, you must enable the router to accept NetConf . This enables the TDD to push con-
figuration to the router:
set system services netconf ssh port 830
Note: If you want to use a custom NetConf port, you can replace 830 with your required port
number. You must also add the custom port number to the Remote Devices table entry for this
router.
2. Create an ephemeral instance of the configuration database named Corero and limit it to only store the last
500 commits in memory. This is where configuration from the TDD is sent.
set system configuration-database ephemeral instance Corero
set system configuration-database ephemeral purge-on-version 500
The Juniper Networks PTX Series router must use gRPC telemetry:
Tip: If you need to add a new certificate to this router, you can use the following command:
set security certificates local <certificateName> load-key-file <URL>
1. Configure role-based access to routers by creating a new login class called TDD and a user within that class
called corero.
set system login user corero class TDD
set system login user corero authentication plain-text-password
set system login class TDD permissions configure
set system login class TDD permissions view
set system login class TDD permissions firewall-control
Note: When you add your routers to the Remote Devices table in SWA, you must enter the
username corero and the same password you configured here.
Caution: Juniper Networks PTX Series router do not support Native Telemetry. In the method
below, do not follow any steps identified as Native Telemetry only. You MUST follow the steps
identified as gRPC Telemetry only.
l ST_MX – For a Juniper Networks MX Series router using port mirror instances
l ST_PTX – For a Juniper Networks PTX Series router using port mirror instances
Note: If you configured role-based access when you set up the router, you must enter the
username corero and the same password you configured there.
Tip: On the table, you can use the following action options to Edit or Delete a remote device.
Note: The Port-Mirror Sample Rate and Ingress Sample rate should be identical. Both must be
the rate factor which reduces the amount of traffic seen by the router to a manageable size for
the vNTD. A default value of 1000 would normally scale to approx 1Tbit/sec. Values less than
1000 will give better fidelity on attack detection and traffic visualization but will add load to the
vNTD. A single vNTD has a peak capacity of 10Gbit/sec of sampled traffic when optimized. (Note:
the sample rate assumes a run-length of 0)
Prerequisites
l The vNTD device which monitors traffic for the TDD system must be in a Cluster.
To configure the Port-Mirroring Sample Rate on the Juniper Networks MX or PTX Series
router
6. If you want to save the new configuration, and push your changes to any affected Defense devices, click
. Then, on the pop-up dialog, click Commit to push the changes (alternatively, you can click Discard
to discard any uncommitted changes).
The Port-Mirror Sample Rate you configured on the router (and the Ingress Sample rate you configure on the
Cluster) controls how many samples are sent from the router to the vNTD. In addition to this, you may need to
adjust the rate of samples sent from the vNTD to the CMS. When you installed the TDD system, you modified the
Inbound Sample Rate to use one of the TDD default rates. If you find you're not seeing enough samples in the SWA,
you may need to adjust these rates.
l SmartWall TDD lab test system using smaller traffic volumes and attacks (from a traffic generator) –
Change the value to 1. This samples every packet received by the Defense device.
5. If you want to save the new configuration, and push your changes to any affected Defense devices, click
. Then, on the pop-up dialog, click Commit to push the changes (alternatively, you can click Discard
to discard any uncommitted changes).
Note: The default sample rates given above are correct for the majority of deployments and you
should make sure the Port-Mirror Sample Rate on your router is correct before adjusting the
Inbound Sample Rate.
vNTD:
vCMS:
vSWA:
You can check your system is now fully connected in the SWA web UI.
On the System > Health screen, the Remote Device Info table won't show green ticks against a router until a filter
has been sent and successfully received by the router. If you want to verify the connection before sending an active
filter, you can use the method below to send a dummy filter to your connected routers:
To achieve the maximum possible protection, however, you may then want to customize your Policy to optimize it
for the traffic that your organization's network encounters. The SmartWall TDD Detection Policy is configured in the
CMS (On the left menu: Policy > Detection).
For information on changing specific Policy settings, open the CMS and access the inbuilt CMS help by clicking
the help icon in the top right of each screen.
If you have certain routers which are receiving different traffic levels, when you begin tuning the default Policy, you
may find if would make sense to make certain changes for the Defense devices identifying attack traffic for some
routers and not others.
To do this, you need to create new Detection Profiles for each Policy version you need. And you need a Cluster for
each Detection Profile. Then you can group your Defense devices into those Clusters (which each provide a different
Detection Profile to a group of devices).
Note: This method provides an overview of the steps required. For more information on any of
the steps click the help icon in the top right of each screen: Policy>Detection Profiles and
Network>Clusters.
Access to the CMS web UI and the SWA web UI is provided via the management IP address configured for each at
setup time; make sure to use HTTPS for both, and specify port 8000 when accessing SWA:
To log in to CMS and SWA, make sure to use the administrator credentials that were specified at setup time. The
default username is admin and the default password is smartwall. If a different username/password combination
was specified during setup, you need to use those credentials instead.
CMS and SWA each provides a link to help documentation in the top menu bar.
In the CMS, click the help icon to access the CMS Knowledgebase. From the home page of the Knowledgebase
you can download additional help PDFs, browse for information using the expandable left-hand menu, or type a
search term in the search bar.
In SWA, clicking Help > User Guide displays a PDF file that describes the controls shown in each SWA screen.
Configuration changes in the CMS do not take effect until they are committed and any uncommitted changes can be
lost when you logout. Always remember to commit your changes (Commit > Commit), before you log out.
Adding a Defense device to the CMS doesn't automatically mean the device is reachable from the CMS, you are just
telling the CMS what device to look for. A variety of different problems could prevent the CMS from communicating
with the device.
In the CMS, click Network > Devices to display the Devices table. The Deployment State column shows connectivity
information for each managed device. The following states indicate the device is not connected:
If a device is "out of sync" it means that the Policy on the device does not match the corresponding Policy in the
CMS. You may occasionally see a device become out of sync after the device has been restarted, or after you
perform a software upgrade. In the CMS, click Network > Devices display the Devices table and look in the
Deployment State column to see which device is out of sync, and what action is required:
l Sync required – The device is connected but its Policy configuration does not match the current Policy com-
mitted to the CMS. The device could have become out of sync if it was unavailable when a change was com-
mitted in the CMS or if you have replaced a connected device with a new version (with the same IP address).
In the Devices table, click the action button and select Sync Device to push the Policy changes to the
device.
l Force sync required – The device is connected but there has been an unexpected error in the Policy con-
figuration. In the Devices table, click the action button and select Force Sync Device to wipe the old Policy
from the device and replace it with the current version stored in the CMS.
l Not in cluster – The device is not in a Cluster. Go to Network> Clusters, add the device to an existing Cluster
or create a new Cluster for it.
l Initial sync pending – The device is new and the CMS has not yet sent its Policy configuration. Wait a few
minutes and check again.
You must have at least 10Gbps available license capacity for the Defense device to automatically license and connect
to the CMS. If you don't, you will have to create some space by delicensing an old vNTD or buying additional license
1. In the CMS, use the left-hand menu to navigate to Network > Devices.
2. On the Devices table, locate the vNTD you want to license.
3. In the Actions column, click and select License.
Tip: If you need to delicense a vNTD, in the Actions column, click and select Delicense. The
delicense option is only available for currently licensed vNTDs. When you delicense a vNTD (or
add it to the CMS when there isn't enough license capacity available), it enters the not-licensed
state. In the not-licensed state, the devices do not send any information via syslog message
(except the device status), and cannot function as Detection Engines for the TDD.
Remote Device added to the CMS Devices table instead of the SWA
If you accidentally add a Remote Device (e.g. a router) to the Devices table in the CMS (Network > Devices), it will
appear with a status message of unexpected device type. Remote Devices cannot be stored in this table, it is
only for vNTDs. Delete the Remote Device from the table and instead add it to the Remote Devices screen in the
SWA Web UI (Mitigation >Remote Devices).
To use SmartWall Network Threat Defense virtual editions (vNTDs) you need to have a license for them. If you do not
have a vNTD license, or you have already allocated your full license capacity, when you add a new vNTD to the CMS
you will be unable to add it to a Cluster. In the CMS, click Network > Devices to display the Devices table and check
the Deployment State column to see if the vNTD is listed as not-licensed. To license this vNTD you need to contact
your Corero representative for additional license capacity and upload the new license file. Alternately, you can
choose to delicense another vNTD. Once you have the available license capacity, at the Devices table click the
action button next to the unlicensed vNTD, and select License.
SWA receives data from the managed Defense and Bypass devices via the CMS. If SWA is not receiving data, it will
not show any information on the Overview scren. Open the CMS in a browser and navigate to System > Analytics &
Syslog. On the Servers table, check the details of your SWA application to make sure you have the correct IP address
and the right port number (the default should be 9997). If you need to make a change click the edit button and
remember to commit your changes (Commit > Commit).
If the CMS is configured correctly to send data to SWA, but you still don't see all the expected data in SWA, it may be
that your devices are not reachable from the CMS. In the CMS, click Network > Devices to display the Devices table
Remote Device Info table (System > Health) is showing warning against new router
On the System > Health screen, the Remote Device Info table won't show green ticks against a router until a filter
has been sent and successfully received by the router. If you want to verify the connection before sending an active
filter, you can use the method below to send a dummy filter to your connected routers:
If you are expecting to see telemetry from a router but none is appearing in the SWA, you must check the router has
been successfully added to the SWA and CMS application.
First, check the SWA. On the Health screen (System>Health), if you cannot see the router in the Remote Devices Info
table, there has most likely been a mistake made when adding the routers.
Check the mitigation alerts have the correct hostnames for the routers: Alerts > Mitigation Application and
Mitigation Removal >Edit > Edit Alert > Corero Autonomic Response. Then under Device Names, make sure the
router hostnames are spelled identically to how they appear in the router and in the SWA Remote Devices table.
Also check that they are all separated by a single comma (no spaces).
Note: When configuring the Mitigation Application and Mitigation Removal alerts, the only sup-
ported configuration areas are Corero Autonomic Response and Trigger Action. Configuring
other areas may lead to unexpected results.
Check the SWA to make sure the correct IP addresses and access credentials are stored for each router. Open the
SWA Web UI > Mitigation > Remote Devices and click Edit next to each router to check the stored information. The
Finally, check the configuration on the forwarding plane of the Juniper Networks MX or PTX Series router. A good
place to start is to check the router is reachable and recieving traffic as expected. The following three
troubleshooting commands can help you begin to diagnose an issue, for more information see the Juniper
documentation for your router.
Caution: Be careful to get the order of words correct in monitor interface traffic.
Typing monitor traffic interface may start a TCP dump.
If you have some of your routers in a remote subnet from the SWA, the telemetry traffic may not be recognized on
the secondary interface. To fix this, you need to add a static route from the SWA to each router on the remote
subnet. You can do this in the SWA pCLI, for each router:
Traffic is entering the network, but the Defense device does not seem to do anything with it
If the tables and charts in SWA show inbound traffic entering the network, but there's no evidence that rules are
being triggered on the SWA, there may be no DDoS attacks occurring. However, if the traffic appears abnormal but
the system is not responding to it as expected, you should first check the system is healthy:
If your system is in good health, another possibility is that you may have accidentally changed the necessary defense
policy settings required to trigger mitigations. Check the CMS defense policy defaults are still in place. You can
contact your support representative for more information.
Note: If the TDD Flex-Rules show a revision number higher than 1, you may have accidentally
edited the filter definition . This can disrupt the TDD system's ability to mitigate attack traffic.
Contact your support representative for a copy of the original filter definition if you're
concerned.
In the CMS application, click on the Alarm icon in the Status bar and open the Alarm Center. Uncleared alarms
are listed, describing issues that require your attention.
If you have lost the admin user credentials for a vNTD, vCMS or vSWA virtual machine, you can reset the
username/password to their default values (admin/smartwall) without redeploying the VM.
Note: Resetting the administrative username and password will not affect any other user
credentials.
1. Create a password reset ISO file. You can use the following command to create a password reset ISO file in
Linux. These commands assume you already have the package containing the mkisofs command installed; if
you don't, you should download this first using the Linux package manager.
>reset_pw
mkisofs -input-charset utf-8 -quiet -o reset.iso reset_pw
2. Transfer the ISO file to a datastore which can be accessed by the virtual machine.
3. On the host where the virtual machine is deployed, set the CD-ROM drive for the virtual machine to the data-
store ISO file . Ensure that the device status is connected.
4. Restart the guest virtual machine. When the virtual machine reboots, it should display the following message:
Username/password reset to default
5. Login with the username: admin and the password: smartwall. After logging in, you can change the user-
name and/or password using the setup aaa command in the pCLI.
During troubleshooting, customer support may ask you for diagnostic packages from the affected systems. You can
download then through your browser in the following locations:
l For SWA – Open the SWA in a browser. System > Settings > Diagnostic Package > Download
l For CMS – Open the CMS in a browser. System > Diagnostics. Under Download file from CMS appliance,
select a source package type and click Download File.
l For a vNTD – Open the CMS in a browser. System > Diagnostics. Under Download file from a device, select a
source package type and the specific device. Click Download File.
Snapshotting VMs
For CMS and SWA VMs, creating snapshots using your host's snapshot features are supported (e.g. VMWare
snapshots on ESXi or virsh snapshot-create-as on KVM). vNTDs DO NOT support the use of snapshots in ESXi or KVM
hosts.
By default, Corero VMs are not configured to automatically start on ESXi server boot. Corero recommends you set
the VMs to automatically start by editing the ESXi servers host settings: .
l JTAC policies—For a complete understanding of our JTAC procedures and policies, review the JTAC User Guide
located at https://www.juniper.net/us/en/local/pdf/resource-guides/7100059-en.pdf.
l Product warranties—For product warranty information, visit https://www.juniper.net/support/warranty/.
l JTAC hours of operation—The JTAC centers have resources available 24 hours a day, 7 days a week, 365 days a
year.
For quick and easy problem resolution, Juniper Networks has designed an online self-service portal called the
Customer Support Center (CSC) that provides you with the following features:
To verify service entitlement by product serial number, use our Serial Number Entitlement (SNE) Tool:
https://entitlementsearch.juniper.net/entitlementsearch/
You can create a service request with JTAC on the Web or by telephone.
l Visit https://apex.juniper.net/
l Call 1-888-314-JTAC (1-888-314-5822 toll-free in the USA, Canada, and Mexico).
l Email: [email protected]
l Web: https://corero.force.com/support
l Telephone: Dial +1.978.212.1500 -> Select Option 2
To deploy a vNTD for high sampled traffic rates, you must complete the following steps:
1. Use an X710 or equivalent NIC(Note: VMXNET3 is not compatible with this method).
2. Deploy using vSphere WebClient
3. Isolate sampled traffic NICs to prepare for PCI Passthrough, and optimize the host
4. Deploy the vNTD
5. Allocate sampled traffic NICs to the vNTD to complete PCI Passthrough
Isolate the sampled traffic NICs for PCI Passthrough and optimize host
Caution: When you deploy a vNTD using vSphere Client, you cannot enable 1G HugePages to
improve performance. For line rate performance at small packet sizes, use a KVM deployment.
4. Isolate the NICs. In the Hardware tab, select the two NICs you want to use for the vNTD then click Toggle
passthrough.
1. In vSphere WebClient, right-click the Host you want to create this Virtual Appliance on, and select Deploy OVF
template.
l Select Local file, click Choose Files and select the OVA file for your required VM.
3. Click Next.
4. Type a Virtual Machine name.
6. Click Next.
7. Select the destination compute resource you want to use for this VM. This will be the host where you want
the VM to be deployed.
19. Click Finish. The VM is created powered off. Do not power it on yet.
20. Check the allocated ports will not create a network loop when enabled, then enable ports:
a. In vSphere WebClient, right click on your new VM and select Edit Settings.
b. Scroll down to Network adapter 2. Check the network label is correct and then under Device Status,
select the Connect check box.
c. Repeat for Network adapter 3.
d. Click OK.
Allocate sampled traffic NICs to complete PCI Passthrough and optimize the vNTD
Caution: The VM must be powered off before completing the following steps.
1. In vSphere WebClient, wait for the vNTD Virtual Appliance to finish building. Make sure it is powered off.
2. In the Navigator, click on Virtual Machines and click on the name of the new vNTD Virtual Appliance.
c. Add two PCI devices using Add other device > PCI device.
d. On each new PCI device, use the drop-down list to select the correct isolated NIC.
e. For the PCI device allocated to each NIC, click on the expand arrow next to New PCI device to display
more options and click Reserve all memory.
d. Click Save.
5. In the VM toolbar, click Power on.