I have a server that is not responding when trying to access through GUI interface. We can ping the address and also access CLI from SSH session. I saw there was a bug found but I believe our version (3.0 (1c)) is passed the bug fix action. We did power off the server and when it came back up, were able to use the GUI interface once but it locked up and we are not able to access again. When compared to our other servers with the exact version, do not see any differences. Next I found a command to "reboot the CIMC" and was wondering if I can try this without re-setting the server because it is one of our subscribers and has many phones already registered to it. I want to avoid resetting if I can by using the "reboot CIMC" command. Any suggestions? or experience in this matter? Thanks in advance... I included the commands for reference.
Your statement not quite correct, a CIMC reboot will reboot ONLY the management interface for the server. It is not impacting to the OS or services running on it. It is a separate entity in the management plane for the server.
CIMC reboot will not cause the server to reboot.
When you say "locks up"....does the host OS itself become unresponsive....or the CIMC GUI?
Have you tried different web browsers, etc.
In such a situation, as you have encountered multiple similar issues. I would suggest reaching out to have the logging reviewed. The CIMC logs memory leaks in /var/log/messages and/or OBFL logging and thus it should be relatively easy (if you can provide a timestamp and logging that covers said timestamp) to determine what the cause of the CIMC becoming unresponsive was.
If you like you may open a case and private message me your case number and I can monitor it and/or provide feedback on the logging depending on my bandwidth.
Evan, thank you for your time. I did "reboot" the CIMC service from the CLI and it DID work! We are now able to log into the CIMC via the GUI like we had in the past. What I noticed is when the server went down after a power issue, it said "ac-cycle" as the reset reason. After reboot it says "graceful-reboot". I will assume this procedure (reboot cimc) allowed for the service to become available for use. Thank you guys for responding. Have a great rest of the week.
Great to hear! Thanks for your reply on this to let us know that it's back to normal. If it happens again, I would reiterate that it would be good to have TAC take a look at it to make sure nothing else is going on within the server.
There are multiple memory leaks that we've seen over time within the CIMC so it's possible that this is new, or simply an older one that wasn't fixed in the code in question here.
I have the same problem with C220M4 running firmware 3.0.3c
CIMC is accessible for a few days via https, then no response. I have to ssh to CIMC and reboot it, then it works again. This is a brand new system running APIC that just arrived this month, and all apic's in the cluster behave the same way. Very disappointing; old firmware had no issues. Hope Cisco can find a workaround that doesn't involve decomissioning all the equipment.
What exactly is the syntax you used to reboot the CIMC from CLI please. In v3.0 (4j) they seemed to have changed the commands.
esx01# scope cimc
esx01 /cimc # reboot
The above does not work, plus i cannot find any further info on the subject.