Unexplained Reboots - DL385 RHEL4 AS x86_64

[kenny chia]

Hi all
I am facing the same problem with a DL385G1 server with RH AS 4 update 3 64 bit. PSP is 7.51.

The strange thing is that I have an additional temperature log after the ASR log (Not before). Can I conclude that the ASR caused a temperature violation?

0003 Critical 19:19 09/09/2006 19:19 09/09/2006 0001
LOG: ASR Lockup Detected: (casm device driver alerted)

0004 Caution 19:21 09/09/2006 19:21 09/09/2006 0001
LOG: POST Error: 1610-Temperature violation detected Waiting 5 minutes for system to cool Press Esc key to resume booting without waiting for the system to cool. WARNING: Pressing Esc is NOT recommended as the system may shutdow

[Alexander Finkel]

Kenny,

I saw the same thing on one of my 385s and it has not recurred since disabling the ASR lockup detection.

I notice that the temps run a little higher on this box overall, but I have not had the temps go over the threshold since.

I would recommend that you disable the ASR Lockup detection and then keep an eye on the temps over at least a few days.

- Alex

[Tait Sanders]

I've got 6 * BL20p g3 blades with RH EL3 up7 installed. Also HP PSP 7.52. Every now and then one of the blades will 'kernel panic' and there are no entries in either the system or HP logs (IML, Insight Diagnostics). Then last night I saw an error on the terminal from a resulting kernel panic: "Uhhuh. NMI received. Dazed and confused, but trying to continue."

I've had a job logged with HP for months and am waiting for this issue to be resolved. This issue was also there when I had HP PSP 7.50 installed.

I've disabled the ASR on one of the blades and the only difference is that the server doesn't reboot (which is ASR's function) but just hangs with the kernel panic...

I'm wondering if it is the case that the HP PSP 7.52 version of the 'Insight Manager Health Driver' is buggy?

Has anyone seen a resolution to the kernel panick issue?

[Richard Rydström]

Hello guys
I've seen the same issue on 5 of our servers, and it was related to the HW combination SA6404 and FC1243 (Qlogic). After replacing the Qlogic based FC1243 with an Emulex based FC2243 problem was solved.