Error message: PIM-4-DUP_VPC_NBR - why is a 2nd VPC peer detected

[twgraham]

I have two data centers running N7K/N5K/N2K.  The N7K are VPC to N5K using VPC domain id 1 at both sites.  The N5K to N2K use VPC domain id 2. I am getting the message:

%PIM-4-DUP_VPC_NBR:  pim [5418]  Unexpected 2nd vPC peer xx.xx.xx.xx detected

Some VLANS are extended between the two data centers so it seems like some kind of peer detection is running over the extended VLANS. The peers should be discovered at the endpoints of the peer-link. What discovery traffic is causing the error message and if the VPC is using PIM for discovery what multicast address is used so I can block it on the VLAN extension?  Thanks.

[InayathUlla Sharieff]

The VPC domain identifier must be unique across the two VPC domains as it will create this
kind of issue if it is found there are multiple devices in the same L2 domain with the
same VPC domain ID configured.

This is supported in the vPC Best Practices Guide for Building a vPC Domain: Guidelines
and
Restrictions<http://www.cisco.com/c/dam/en/us/td/docs/switches/datacenter/sw/design/vpc
_design/vpc_best_practices_design_guide.pdf#page=24&zoom=auto,105,229 >

HTH

REgards

Inayath

[twgraham]

Thanks. I realize the best practice.  In my case what is done is done. The issue seems to be cosmetic as the two datacenters seem to work fine except the thousands of SYSLOG messages.

I would like to block the VPC  traffic that is traversing the L2 link between the data centers. Based on the message identifier I have to assume that it is some kind of PIM multicast traffic. Also out of curiosity I would like to know what this traffic is. Peer discovery should happen over the peer link and not across every VPC VLAN.

Perhaps I could solve this by blocking all PIM on the offending VLANs. Unfortunately one of the VLANs needs to transport PIM multicast. I would think that there is a known multicast address that is causing the messages and if I block that address with a  VACL the messages would cease.

[danailpetrov]

Hi mate,

it might not be the best solution, but you can get rid of these messages by lowering the logging level of pim service: 

logging level ip pim 2

Hope that makes sense.

[mspence17]

I had this issue and it turned out I had a misconfiguration on one of my OTV VLANS which caused it.  I had copied an existing OTV VLAN configuration that contained ip pim sparse-mode.  When that got copied over to all 4 SVI interfaces at our main and remote data center then we started getting these messages.  Once I had removed it from the SVI configuration the messages stopped.  Just adding this to see if it helps anyone else