1

The system started resetting itself exactly every 2 hours. These are the messages in the iLO2 log:

Informational iLO 2 03/07/2010 20:40 03/07/2010 20:40 1 Server power restored. Caution iLO 2 03/07/2010 20:40 03/07/2010 20:40 1 Server reset.

It's not an ASR reset (that would show in the log) Redundant power supplies, swapped but no change. Turned off all virtual machines (i.e. now only running hypervisor) but not OK Boot HP smartstart diagnostics disk, ALL OK Diagnostic disk reports no errors Went back to booting Hypervisor and the problem is back.

Seems the hyper-V system disk has got a time based program (virus) causing the reset. I thought the hypervisor had a small attack surface and should be OK. All virtual machines (SBS2008, Win7 and Win XP) and network computers are protected with TrendMicro WFBS.

I am about to rebuild the disk (I have backups) but wondered if there were any suggestions to try first???

5
  • I'd have thought that those messages indicated that the ILO was telling you that the physical power supplies were losing supply power, not a controlled shutdown from within the OS. Are you 100% sure that there isn't something going wrong with your AC\UPS supply? I'm not familiar with ILO logs though - do you see the same log entry if you forcibly reset a server?
    – Helvick
    Mar 7, 2010 at 11:41
  • 1
    I can not believe it's a virus in the hypervisor. Could it be the UPS software is flawed and telling it power is about to fail? I'd disable all UPS and power sensors & drivers.
    – Knox
    Mar 7, 2010 at 11:49
  • A forced reset say: Caution iLO 2 03/07/2010 18:37 03/07/2010 18:37 1 Host server reset by: setup. The power supplies are OK ... I can remove either and there is no restart.
    – GT
    Mar 7, 2010 at 11:54
  • One power supply is direct from the AC, the other via the UPS. I am not using management software for the UPS.
    – GT
    Mar 7, 2010 at 11:55
  • 1
    Seems similar to this forums11.itrc.hp.com/service/forums/bizsupport/… . Either a hardware fault or a problem with the OS Agent's handling of hardware alerts. Have you escalated this to HP yet, seems very likely to be a known issue. I know early versions of the HP G6's had some power\power management deficiencies so it definitely worth calling them.
    – Helvick
    Mar 7, 2010 at 12:10

2 Answers 2

0

This may be unrelated, but if you happen to have installed the RC of Hyper-V server 2008 R2, the expiration was on March 1st I believe. After the RC version expires, it begins to reboot every two hours.

1
  • Yes I WAS running the RC version, even though I had the final version ready to load (never got a round to it). Silly me thought there would be some kind of Microsoft message to tell me if the RC had expired, not just re-boots! Just finished loading the new version and it seems OK now. Thanks all for the troubleshooting ideas.
    – GT
    Mar 8, 2010 at 11:04
0

I had something similar happen on a ML350G5. It turned out that my power supply backplane needed to be replaced. I suggest calling support and going through it with them. It took a very short time to determine that this was the problem.

You must log in to answer this question.