Mysterious shutdown problems of many Z620 Workstations

atledale · ‎05-24-2013

For about 6mnths now we have been chasing the root cause of why 12 out of 15 workstations randomly shuts down. They even usually do not restart. Just shuts down. The log has the famous Event id 6008 ("unexpected shutdown").

Since the problem affects so many workstations, it is tempting to think this is the PDU/APC Power system, but why does it then not affect all of them?

Appreciate any feedback.

I finally thought this could be due to some thermal alarm that shuts it down, but I could not find its settings in the BIOS.

I also assume these workstations do not use ASR.

MSLEEE · ‎05-24-2013

What is the bios version? If it's not v3.15 that was leased this week, please update it. If the problem still exists, please call HP Support and escalate it.

atledale · ‎05-24-2013

hi,

it was not the latest version from May. I have upgraded 3 machines to the latest BIOS as well as the Intel storage drivers. But in the release notes for the BIOS, it does not mention a fix of shutdown problems, so I am a bit pessimistic...

Dan_WGBU · ‎05-24-2013

atledale, try the latest BIOS posted on hp.com and let us know if this makes any difference.

I am an HP Employee.
My opinions are my own, and do not express those of HP.

Please click "Accept as Solution" if you problem was solved. This helps other forum readers.

atledale · ‎05-27-2013

2 machines shut down this weekend. They both had the latest BIOS 3.14. One of the machines even had its own power supply going outside of the APC PDU.

The strange thing is that they never seem to write a memory.dmp or minidump. But today I found one by coincidence from 25 April. I will look for a tool to analyze it. What is recommended that is free? The one from Microsoft? Hopefully I will find the answer through this analysis.

atledale · ‎05-27-2013

seems like the attachment is the culprit, at least in the one machine I have the minidump for.

I deduced from the minidump using a tool that the crash is caused by a Windows corrupt system file:

rspcrash64.sys

and it can perhaps be fixed with sfc /scannow or the file can simply be replaced.

What is recommended in this case? What is this file?

Here is a link to the use of sfc:

http://www.sevenforums.com/tutorials/1538-sfc-scannow-command-system-file-checker.html

I verified the Windows system files (Win 7 Enterprise) integrity by: sfc (verifynow. It did find problems. I will now proceed to go ahead and fix them using the above command.

But a bit strange so many machines are affected.

Update: Sfc /scannow could not repair the files. I will now go ahead and replace this file.

atledale · ‎05-30-2013

I could not replace this file, rspcrash64.sys, it was not in the win7 Enterprise CD from 2009 (without SP1). If noone have a different suggestion, we will have to re-image all these PCs.

dima999 · ‎06-04-2013

I had random restarts when I allowed teh factory intsall to automatically update itself....reinstalled and prohibited any updates....been no restarts ever - hope this helps!

atledale · ‎06-04-2013

ok, thanks, but we use Microsoft SMS to update. Anyway, I conclude something must be wrong in the image. From what I gather, these PCs may have a cloned image on them, so now I am reinstalling a couple to see if it helps.

dima999 · ‎06-04-2013

yes.....I used the factory install cd which came with teh workstation and abolished any hint of updating (first time it installed like 20000 updates which might have interfered wit the factory setup...