cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
17591
Views
5
Helpful
5
Replies

vPC Peer-Link Failure

Marko Pribanic
Level 1
Level 1

Hello,

In the case I have two N5k acting as a vPC peers and I lose the vPC peer-link between two of them, but I do not lose the vPC peer-keepalive link, what would happen when the vPC peer-link comes back again?

As I understand in the case of vPC peer-link failure all vPC member ports on the secondary N5k will be shut down. When the vPC peer-link comes back again what would happen?

I have read that in that case the vPC member ports will not come back automatically, but they will remain disabled until you do manual recovery. Is that really so?

Is there some way that we can automate the process upon recovery?

Thanks

1 Accepted Solution

Accepted Solutions

With just peer-link going down and coming up, I have not seen an issue in 4.x as. Probably your instructor was talking about an issue where with one vPC peer shutdown and reload of the other vPC peer would suspend all vPCs until the peer which was shutdown was brought back up. This issue is addressed in the 5.0 code with "reload restore" knob

http://www.cisco.com/en/US/docs/switches/datacenter/nexus5000/sw/layer2/502_n1_1/Cisco_n5k_layer2_config_gd_rel_502_N1_1_chapter8.html#concept_AE79781DC1554F0780F6C7BF7FC77624

View solution in original post

5 Replies 5

Prashanth Krishnappa
Cisco Employee
Cisco Employee

With vPC peer-link down, but keep-alive link up, the vPC secondary will suspend all its vPC port-channels. However when the vPC peer-link comes back up, the expected behavior is for the vPC secondary switch to bring back up all its vPC channels. No manual intervention is needed for this. Are you seeing any other behavior?

I am not experiencing any strange behavior (yet).

Last week I attended the UCS boot-camp and the provided materials stated that behavior.

The instructor was not been able to clarify on this so I am just curious does this really work as it was written.

I must say that the materials provided were rather old, so maybe with new 5.0 NX-OS code this issue is fixed (if issue even existed at all).

With just peer-link going down and coming up, I have not seen an issue in 4.x as. Probably your instructor was talking about an issue where with one vPC peer shutdown and reload of the other vPC peer would suspend all vPCs until the peer which was shutdown was brought back up. This issue is addressed in the 5.0 code with "reload restore" knob

http://www.cisco.com/en/US/docs/switches/datacenter/nexus5000/sw/layer2/502_n1_1/Cisco_n5k_layer2_config_gd_rel_502_N1_1_chapter8.html#concept_AE79781DC1554F0780F6C7BF7FC77624

I need to understand the difference on below commands-

Peer-config-check-bypass

Reload restore or Auto recovery 

Also i have below doubts-

•-          I have 2 Nexus boxes(connected using peer links and keep alive link) running on production. If I put ‘auto-recovery’ command in primary box, what will be the overall impact as the same command is not put on second box.

•-          To function ‘auto-recobvery’ command is it necessary to have ‘peer-config-check-bypass’ command as well?

I have one more scenario-

•-          Lets say I have 2 nexus connected via peer-link and keepalive link and I have configured ‘auto-recovery’ command on both boxes under vPC domain. Peer link goes down first but keepalive link is still active, In this case secondary nexus will put its vPC and interfaces down. After sometime keep-alive link alos goes down. After failing to receive 3 continous keepalive packets, seconday box makes all its vPCs up due to ‘auto recovery’ command. What will happen the moments second box starts receiving keepalive packets and peer link is also up.

The reload restore command has been removed/replaced and the new feature is
now called auto recovery. Auto recovery covers the use case that reload
restore addressed, plus more.

If both switches reload, and only one switch boots up, auto-recovery allows
that switch to assume the role of the primary switch. The vPC links come up
after a configurable period of time if the vPC peer-link and the
peer-keepalive fail to become operational within that time. If the peer-link
comes up but the peer-keepalive does not come up, both peer switches keep
the vPC links down. This feature is similar to the reload restore feature in
Cisco NX-OS Release 5.0(2)N1(1) and earlier releases. The reload delay
period can range from 240 to 3600 seconds.

When you disable vPCs on a secondary vPC switch because of a peer-link
failure and then the primary vPC switch fails, the secondary switch
reenables the vPCs. In this scenario, the vPC waits for three consecutive
keepalive failures before recovering the vPC links.

The vPC consistency check cannot be performed when the peer link is lost.
When the vPC peer link is lost, the operational secondary switch suspends
all of its vPC member ports while the vPC member ports remain on the
operational primary switch. If the vPC member ports on the primary switch
flaps afterwards (for example, when the switch or server that connects to
the vPC primary switch is reloaded), the ports remain down due to the vPC
consistency check and you cannot add or bring up more vPCs.

For more information, please refer to the Operations Guide: As a best practice,
auto-recovery should be enabled in vPC.

HTH,

Alex

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: