BGP Peer Monitoring in VSX

[Mike_Goodwin]

We have a R80.10 VSX deployment where a couple of the VS's have a number of BGP peer relationships.

I'd like to add this into our network monitoring system to query the BGP peer status via SNMP using the context-switch to get the specific info per-VS.

TAC tell me this is not possible due to a code limitation - it seems the BGP MIB is not present in the VS specific SNMP deamon. Similarly, SNMP traps from this MIB won't work.

Anyone else doing this? What's the recommendation on how to monitor it? Via /var/log/routed_messages_vs messages? Or can we get the peer transition logs into Smart Event somehow?

I'm also thinking of scripting the monitoring via SSH from my management server to get the output of "show bgp peers", but it would be much nicer to have a native support tool rather than having to do a new SSH session every 30 seconds?

Thanks for any input!

Mike

[Fredrik_Holmber]

Interesting. SSH CLI scraping sounds a bit heavy, agreed, and since SNMP is out of the equation maybe a log shipper or some other way of forwarding /var/log/routed_messages_vs to an external log analyser (Elastic/Splunk etc.). Then you can generate alarms from there based on content and levels.

[PhoneBoy]

Another option would be to use the run-script API to make the relevant call to the gateway/vs.

At least that doesn't require an open SSH session.