NMI received for unkown reason

rjo98 · 06-03-2012, 11:33 AM

I had a server that was hung so had to power cycle it. Looking in the messages file, these are the last three lines it has prior to the reboot. Anyone know what this means? Dell server with an external array, RHEL.

Jun 3 05:18:38 ServerName kernel: Uhhuh. NMI received for unknown reason 21 on CPU 0.
Jun 3 05:18:38 ServerName kernel: Dazed and confused, but trying to continue
Jun 3 05:18:38 ServerName kernel: Do you have a strange power saving mode enabled?

robertjinx · 06-04-2012, 07:38 AM

This is some kind of hardware related issue, you can boot maybe with unknown_nmi_panic=1 and see what happens.

rjo98 · 06-04-2012, 07:46 AM

Thanks for the reply. I restarted the system without that and it appears to be working ok now. not sure how to tell what caused it though.

grim76 · 06-04-2012, 09:28 AM

Make sure to have your firmware and drivers up to date. Another thing that I have noticed is some erratic behavior with cstates turned on in the bios on Dell servers.

robertjinx · 06-04-2012, 01:29 PM

There are a lot of issues with NMI due to hardware, so listen to grim76 and keep your hardware up to date.

rjo98 · 06-04-2012, 02:12 PM

Firmware and drivers are up to date as far as I know. Not sure about the cstate settings though. Any way to check them when the system is already powered up?

grim76 · 06-05-2012, 09:33 AM

I don't think you can check them while the system is online as they are BIOS settings. I would just confirm that all your firmware is where it should be and on your next reboot check your cstates.

rjo98 · 06-05-2012, 11:37 AM

Thanks, i'll give it a shot.