we are using spectrum 9.2 and trying to monitor cisco BGP FSM State change alerts.Traps were already configured on the device.In the attributes list not able to see cbgpfsmstatechange attribute so we are not receiving any events for the device.
Can someone help me how to monitor Cisco BGPFSMStateChange.
In my 9.4 box, i see FSM state change event configured with no alarm. Event code:210d53. Is this available in your box?
I also suggest running wireshark with filter udp.port==162 to see if we are actually receiving the state change traps.
Thanks for your response. I'm able to find Event code : 210d53 in event configuration. I will run wireshark with filter udp.port==162 and trace the result and let you know the reply.
What is the clear event code for Event code:210d53 ??
how to get clear event codes ?
Out of the box there is no alarm asserted for 210d53 as we don't want alarms for state change. We do get the poll/trap based alarms when the BGP is down.
Incase you would like to have an alarm for 210d53, map an alarm and assert an event condition rule based on bgppeerstate event variable value.
Something like if bgppeerstate is established generate new event which will clear alarm for 210d53.
Thanks a lot for clear explanation..Much useful the comments!!
We are running CA Spectrum OneClick Version 10.2.0.0.245
There are two routers for which we are continuously receiving events/alarms which are getting created and within the next 1-3 seconds they are getting cleared.
Below is a sample of event received in spectrum for one of the routers.
A "cbgpFsmStateChange" event has occurred, from GnSNMPDev device, named uk-ld6-r91.ent.bhicorp.com.
The BGP cbgpFsmStateChange notification is generated for every BGP FSM state change. The bgpPeerRemoteAddr value is attached to the notification object ID.
bgpPeerLastError = bgpPeerLastError.bgpPeerRemoteAddr = X.X.X.XbgpPeerState = idlecbgpPeerLastErrorTxt = hold time expiredcbgpPeerPrevState = active
There are three BGP neighbors which are in idle state right now.
But there are no logs on the router for last two days. But the events are still continuously coming.
We have received approximately 4000 events on both the routers until now.
Could this be because of hold time expiration between bgp neighbors or could it be some setting on the CA Spectrum that we nee do to change to stop this ?
Please let me know if any additional information is required at this point of time.
Since these originate from the cbgpFsmStateChange trap from the device, there is nothing in Spectrum you can do other then configure the event not to assert an alarm.