02-12-2014 09:13 AM - edited 03-07-2019 06:09 PM
Hey All,
I'm trying to troubleshoot a problem with our Nexus 5020 switches. We've been having intermittent issues over the last three weeks with server reboots. Windows servers sometimes come up with their NICs missing their IP address. So far we know that this is only happening on servers that are connected to our Nexus 5020 switch pair.
I found this in the log that is close to the time when the problem started.
2014 Jan 26 00:41:42 dc1n5020-2 %VPC-2-PEER_KEEP_ALIVE_RECV_FAIL: In domain 1, VPC peer keep-alive receive has failed
2014 Jan 26 00:50:00 dc1n5020-2 %VPC-2-PEER_KEEP_ALIVE_RECV_FAIL: In domain 1, VPC peer keep-alive receive has failed
The VPC role has switched as well.
vPC role : primary, operational secondary
I'm thinking about restarting these switches. Is there something I can do that is less disruptive?
Thanks,
Patrick
02-12-2014 10:00 AM
you need to check the physical link that runs the vpc keep alive link . It's telling you it is failing at times which could be a physical issue such as cabling . Run extended ping tests across the vpc peer keepalive link to see if anything is dropping . Check for errors or drops on either side interfaces.
02-12-2014 10:24 AM
The switch that the peer keepalive traffic is on was restarted which caused this problem. The damage is done. How do I switch my VCP primary node back to primary?
02-12-2014 12:34 PM
I believe to get it to switch back you have to bounce the vpc peer link .
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide