Implementing Virtual Routing and Forwarding VRF On Cisco Nexus Data Center
Implementing Virtual Routing and Forwarding VRF On Cisco Nexus Data Center
Implementing
Virtual Routing and
Forwarding (VRF)
on Cisco Nexus
Data Center
Switches
1-800-COURSES www.globalknowledge.com
Implementing Virtual Routing and
Forwarding (VRF) on Cisco Nexus
Data Center Switches
Chris Olsen, Global Knowledge Instructor, Author, IT Consultant
Introduction
Cisco in the past six years has developed a wide array of leading-edge technologies for the data center with a
focus on the Nexus switches and UCS servers. Within the Nexus switches, Cisco has created many different
technical approaches to achieve the two high-level objectives of connectivity or isolation. Simply put, for users to
access email on servers, query a database, or open a web page, they must have connectivity into the data center.
However, the obvious benefit of connectivity also creates the same opportunity to spread viruses, Trojans, worms,
and denial of service (DOS) attacks as well as the full array of potential security breaches. Therefore, a secure
data center design must also implement isolation for all networking devices, hosts, and virtual machines that
don’t have a productive need for connectivity amongst each other.
Cisco provides a wide array of connectivity and isolation tools within the datacenter. This white paper addresses
Virtual Routing and Forwarding (VRF), which is a Layer 3 isolation mechanism for routing protocols.
A common use of VRFs is to isolate the management network from the production network. In the event that a
production network experiences any severe problems, to fix the network remotely, an engineer would need the
“broken network” to fix the network. Before VRFs, network designers often solved this dilemma by building a
network called a management network separate from the production network to aid troubleshooting with
remote connectivity. VRFs create a cost-effective management network fully isolated from the production
network while sharing the same hardware and cabling.
Another common use of VRFs is for isolation for multitenant cloud providers. Many public and private cloud
providers can create a single common data center infrastructure to provide services to several different customers
or tenants. VRFs can be implemented to ensure that the different tenants have a secure isolation from other
tenants while benefiting from the cost savings of a common infrastructure.
On any newly installed Nexus switch, there are two VRFs that exist by default, which are named default and
management. This first example is on a Nexus 5548 switch named Nexus5k.
Note that even basic connectivity from or through a Nexus device requires an understanding of VRFs. The
following two different ping tests to the same destination get different responses based solely on the VRF being
referenced, or not referenced.
VRFs require Switch Virtual Interfaces (SVI) for inter-VLAN routing which is enabled with the feature interface-
vlan command. The example shown here will be based on the routing protocol OSPF, which also needs to be
enabled. Then the VRF is applied to all applicable Layer 3 interfaces within the device.
Note the output of the OSPF routing protocol differs entirely based on the VRF. The following output includes
the VRF OSPFPrivate. There are no other OSPF routers in this simple example.
Also note how OSPF is enabled per interface, but also per interface per VRF.
Learn More
Learn more about how you can improve productivity, enhance efficiency, and sharpen your competitive edge
through training.
DCUFI - Implementing Cisco Data Center Unified Fabric v5.0
DCNX7K - Configuring Cisco Nexus 7000 Switches v3.0
DCNX5K - Implementing the Cisco Nexus 5000 and 2000 v2.0