07-09-2015 12:27 AM - edited 03-01-2019 12:16 PM
Hey folks,
some strange things happen yesterday evening. I got round about 120 messages about "threshold crossed: TCA :etherTxStats..." and "IO Module Warning".
I looked in the events of UCSM. Every blade server in our UCS started a FSM. Is this a routine from UCS? We have no idea what happen. All 32 blade server running ESXi 5.5. We didn't change anything.
FSM Events see attached file. I filtered all "Skipped" events. For me, it looks like CIMC has started new inventory for all servers? Do you know something about that or do you have same happenings? Is there a internal scheduled task in UCS?
UCS version 2.2(3f) (all systems = same firmware bundle; UCSM, FIs, Blades..)
thanks a lot
/Hugo
Solved! Go to Solution.
07-09-2015 05:34 AM
Would be interesting to have a look at the audit logs around that time and the DME (database Management Enginer) to see what really happens... You better open a TAC case and then post the outcome here for future users facing the same cause this will require deep investigation of the logs before we an tell you what/why it happeed.
-Kenny
07-09-2015 03:05 AM
Hello Danny,
Are these B200 M4s? if yes then check the bug:-
https://tools.cisco.com/bugsearch/bug/CSCut61527/?referring_site=bugquickviewclick
there's no scheduled activity in UCS that will do this!
thanks,
Saurabh
07-09-2015 04:14 AM
Thanks for fast reply.
We are using B200 M3. The servers didn't reboot. There was no impact in our VMware environment.
thanks,
Hugo
07-09-2015 05:34 AM
Would be interesting to have a look at the audit logs around that time and the DME (database Management Enginer) to see what really happens... You better open a TAC case and then post the outcome here for future users facing the same cause this will require deep investigation of the logs before we an tell you what/why it happeed.
-Kenny
10-08-2015 02:18 AM
I opened a TAC case, but the result is null. Between the monitoring windows of TAC, the problems doesn't occured again. So they cannot say much about that.
We will see, wether the problem occurs again or it is fixed by an update.
07-09-2015 11:55 AM
Danny,
I believe Kenny is right here. This requires logs analysis to find the possible cause.
you can collect the Tech-support files and upload them to the case (once you get it created) directly.
Download help for tech-support files:- http://tools.cisco.com/squish/8cf53
Upload the files using the new HTML upload tool: https://cway.cisco.com/csc
thanks,
Saurabh
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