Руководство По Проектированию для Cisco Cisco Nexus 5010 Switch

 

 

Design Guide 

© 2010 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. 

Page 35 of 38 

 

The peer-link PortChannel connects vPC peers and carries all access VLANs (defined by the user). This link also 

carries additional traffic that the user does not need to define: more specifically, BPDUs and HSRP hellos and MAC 

address synchronization between the vPC peers. 

This link is by far the most important component of the vPC system. Although its failure does not disrupt existing vPC 

flows, its failure can impair the establishment of new flows and isolate orphan ports. Configuring the peer link in a 

redundant fashion helps ensure essentially uninterrupted connectivity between the vPC peers. The following script 

illustrates how to configure the peer link, which in this case is PortChannel 10: 

agg(config)# interface port-channel10 

agg(config-if)#  vpc peer-link 

agg(config-if)#  switchport trunk allowed vLAN <all access vLANs> 

The configuration of the peer link automatically installs Bridge Assurance on the peer link. This configuration is 

compatible with ISSU, so you can keep Bridge Assurance enabled on this link. 

The peer link carries a copy of the multicast traffic regardless of whether there are orphan ports that need to receive 

it. You should provision the bandwidth for the peer link accordingly. 

The peer-keepalive connectivity should never be carried as a VLAN on the peer link; otherwise, it will not provide any 

benefit. Instead, it should be carried over a routed infrastructure, and it does not need to be a direct point-to-point link. 

The following configuration illustrates the use of a dedicated Gigabit Ethernet interface for this purpose: 

vrf context vpc-keepalive 

 

interface Ethernet8/16 

  description tc-nexus7k02-vdc2 - vPC Heartbeat Link 

  vrf member vpc-keepalive 

  ip address 192.168.1.1/24 

  no shutdown 

 

vpc domain 1 

  peer-keepalive destination 192.168.1.2 source 192.168.1.1 vrf vpc-keepalive 

You should not use the mgmt0 interface for a direct back-to-back connection between Cisco Nexus 7000 Series 

systems because you cannot determine which supervisor is active at any given time. You can use it instead on the 

Cisco Nexus 5000 Series. 

The mgmt0 interface can be used both for management and for routing the peer keepalive through the out-of-band 

management network. In this case, each Cisco Nexus 7000 Series Switch is connected to the management network 

through mgmt0 of supervisor slots 5 and 6 and the Cisco Nexus 5000 Series through the single mgmt0 interface. 

By following this approach, regardless of which supervisor is active, the Cisco Nexus 7000 Series Switch has one of 

the mgmt0 interfaces connected to the management network, which can then be used for peer-keepalive purposes. 

PortChannels are configured by bundling Layer 2 ports (switch ports) on each Cisco Nexus switch through the 

command vpc, as shown in the following code. The system sends an error message if the PortChannel was not 

previously configured as a switch port. 

agg1(config)#interface ethernet2/9