Is ICMP polling possible for VSS links

[Khundmir4cisco]

Dear Expert, 

We have a number of distributor switches that has VSS configured. Many times the VSS stand-by becomes active and switch restarts  We would like to know the occurrence of this event by ICMP polling as we have not enabled traps for vsl link down. We have already scheduled a workflow in our NMS to generate information and status of vss and redundancy but we are trying to find out if this can be done by polling the vsl link. Or enabling vsl traps is the only solution. Is there any work around for achieving polling to vss. 

---

[balaji.bandi]

If i were you, i always investigate why this is happening as recurrance, most of the case due to link failure between device was fibre or some loose connections. woth checking IOS and any bugs related to this issue. ( since we do not have any information on this post what is the hardware and IOs running, too immature to comment on that side)

yes good to have NMS monitoring system and syslog generated to diagnosis the issue.

[Khundmir4cisco]

Thank you for the reply. 

Yes, we are already diagnosing and trying to fix that issue. In the meantime, we would like to have periodic ICMP polling to vss link in our NMS and the same I would like know if it is possible. We can set it up by enabling snmp traps for vsl but we do not want to enable those traps as we are using our NMS for polling only. I hope I am clear with my query..

[Moaz.Elzhrawey]

Hi,

Here is my take, if I were in your shoes, there are two solutions,

1- Straight forward, which is by trapping the VSL link and logs to the NMS system, along with debug activated for all vss events. 

2- You should have IP BFD activated and configured properly between both peers so you can detect Dual-Brain (Active/Active) scenarios.
By doing so, it's double edged sword, you will prevent dual active, and to track the issue causing the reboot, at minimum with timing and associated timestamp, trap the snmp and logs of BFD as well to the NMS.

*** Please rate this as "Answer" if it solved your problem.***

[Khundmir4cisco]

Thank you for your reply.

I am basically managing NMS. So, configurations and preventing the Dual-Active scenarios our Network Engineer will take care of it. I need to reflect these instances in NMS and that's my task.

For your 1st option,

I have already created and scheduled a workflow for vsl link status and to check vsl redundancy in NMS. I have scheduled it in 3 different times of the day. Is it what you are referring to? We also have dual-active fast hello configured to detect dual-active. Is there a way to poll these fast-hello interface. As far as debug is concerned, please guide me how to do it and how to get it in NMS.

For your 2nd option, I can ask my network team to do it and same I will monitor in NMS. Any reference you have to set up IP BPD.

[Moaz.Elzhrawey · Accepted solution]

here you are,

https://www.cisco.com/c/en/us/support/docs/switches/catalyst-6500-virtual-switching-system-1440/113627-dual-active-bfd-vss.html

## This is how to configure IP BFD, then you need to enable the logging/traps on the switches and polling BFD from your side (NMS)

Finally, if you you want to trap the dual active / fast hello, here is how to,

https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/snmp/configuration/15-sy/snmp-15-sy-book/nm-snmp-vsl-enhnce-dualactve-detctn.pdf