Call Manager 5.1.2 server froze up, had to power cycle

Unanswered Question
Oct 6th, 2008

Has anyone seen this on version on a 7835H (ProLiant DL380 G4).

The server is a subscriber server, and was running fine for months, and over night froze up and we could ping it, but could not web into it, could not SSH into it, and the local console was not responding. The phones did not failover to the publisher server which they are configured to do either. We power cycled the subscriber with power button and everything is running fine again now.

I'm aware of the field notice with the HP servers freezing up, but that problem is supposed to be fixed in this version.

Any ideas?

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
will.alvord Mon, 10/06/2008 - 11:20

I had a similar issue happen late last week. 2 node cluster, but the 1st indication was a pub issue. The pub's web server stopped responding, but I could still ssh to it. All of the devices were reg'd to the sub, so a restart appeared to correct the issue.

Until I started checking RTMT side-by-side with an ssh session, running through some health checks. I notice that the callmgr service had stopped on the sub. I couldn't restart the service at all -- kept failing. The sub failed before the pub was fully up though, so there was about a 3-4min delay in device registrations which stunk. Anyways, I restarted the sub and it came up cleanly and all of the devices seamlessly registered back to it.

This cluster is on the same hardware as yours, but not the exact same software release. Also, this issue wasn't exactly the same as yours, but similar.

Maybe some others have had similar issues as well?



This Discussion