Cisco Cisco Nexus 5010 Switch Guida Alla Progettazione
![Cisco](https://files.manualsbrain.com/attachments/7380d0050044647c30f5c24bbbf5d0c0b6d9bb84/common/fit/150/50/faa183d287233c52228cfea3dbc2a127fe780f60564fcb0955d9c3d1cd23/brand_logo.png)
© 2015 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information.
Page 45 of 119
●
Routing the peer keepalive from an SVI over a regular front-panel port provides the advantage that you can
connect Cisco Nexus 5000 Series Switches back to back if you want to do so; it also provides additional
verification of the health of the ASICs through which the peer keepalive flows. It provides the disadvantage
that the peer keepalive is routed according to the information contained in the default VRF instance (which
may cause the traffic to use the peer link instead of a different path); in addition, ISSU is not compatible with
this configuration.
When using a Layer 3 card for routing on the Cisco Nexus 5500 platform, you can allocate a VRF instance just for
this purpose and associate the SVI or a Layer 3 interface for the purpose of carrying the peer keepalive.
In most deployments, routing the peer keepalive over mgmt0 is the best option. With a Layer 3 card, routing the
peer keepalive over an SVI with a dedicated VRF instance is the next best option.
The configuration of the peer keepalive using mgmt0 is as follows:
interface mgmt0
ip address 10.51.35.18/27
vpc domain 2
peer-keepalive destination 10.51.35.17 source 10.51.35.18
The following is a sample configuration of the peer keepalive using an SVI:
! Port connecting Nexus5k01 to Nexus 5k02 directly
interface Ethernet1/31
description direct-peer-keepalive
switchport access vlan 3
spanning-tree port type edge
! Clearing the peer-link from the VLAN that is used for the peer-keepalive
! If using MST you need to configure BPDU filtering to prevent this port from
blocking
blocking
interface Po10
switchport trunk allowed vlan remove 3
! Defining the SVI that is used by the peer keepalive, in this case this SVI
! is defined within the VRF default
interface Vlan3
no ip redirects
ip address 10.50.3.10/31
no shutdown
! configuration of the peer-keepalive information with the associated vrf
vpc domain 2
peer-keepalive destination 10.50.3.11 source 10.50.3.10 vrf default
Note: When using a Layer 3 card on the Cisco Nexus 5500 platform, you can create a dedicated VRF instance
for the peer keepalive and hence specify this newly created VRF instance in the peer keepalive configuration.
for the peer keepalive and hence specify this newly created VRF instance in the peer keepalive configuration.
Most failures in the vPC topology are recovered within 1 or 2 seconds. Peer link failure and recovery is the most
disruptive type of failure, in part because the vPC secondary device needs to determine whether or not to shut
down vPC ports on the basis of whether or not keepalive messages are received from the vPC primary switch.