diagnose purple screen of death ESX 6.0

Question

Our server keeps crashing with a purple screen of death. I have tried interpreting the crash dumps but to no avail. I have an HP Proliant DL360 G7 with ESXi 6.0 running on it and I have other identical servers with the same software with no issues, but this one just keeps crashing about once a week with the following error.

Machine Check Exception: Fatal (unrecoverable) MCE on PCPU5 in world 32801:coalesceWorl System has encountered a Hardware Error - Please contact the hardware vendor

I have all of the crash dump logs if someone would be willing to go over them and help me figure out exactly what the cause of the problem is. I can't have this server in production when it keeps crashing all the time.

You can view the latest crash dump here. http://pastebin.com/JvziBPtA

Please any help would be greatly appreciated.

If the server is still under warranty, contact the vendor, as suggested. — Gerald Schneider, Jan 16 '17 at 17:41
Presumably, you have an active support contract with VMware and HP. If so, have you contacted VMware and HP Support? — joeqwerty, Jan 16 '17 at 17:41
I do not have an active support contract and there is no warranty on the server. — Kyle Vaughan, Jan 16 '17 at 17:44
Since you have other identical servers, I would suggest swapping the processors in the problem server with a working server. If the other server quits working, you know you need a new processor. — B00TK1D, Jan 16 '17 at 17:55
Unfortunately that is not an option as the servers are a few hours away at our other location. IS there any chance someone could narrow down the culprit by examining the dump log I provided? — Kyle Vaughan, Jan 16 '17 at 18:02

score 2 · Answer 1 · answered Jan 16 '17 at 19:40

2

I suggest troubleshooting or following the guidance of the error message.

Things you can do:

You're running ESXi build #3620759 from March 2016. UPDATE YOUR ESXi INSTALLATION!!
HP servers have comprehensive diagnostic messages and logging.
Look at your ILO3 interface and open the HP IML log. That will tell you what is wrong.
If you're using the HP-specific ESXi install, look at your hardware status interface.
Run diagnostics from the HP Service Pack for ProLiant bootable DVD.
Call HP support.

answered Jan 16 '17 at 19:40

ewwhite

197,159
92
443
809

I have tried a newer installation, it gives the PSOD even more frequently, rather than it occurring once a week or so, it happens every few hours. iLo contains no useful information. The Hp specific hardware status interface shows that everything is normal. Running the hardware diagnostic seems to be my only option but will be difficult as I am a few hours away from the server. – Kyle Vaughan Jan 16 '17 at 23:16

diagnose purple screen of death ESX 6.0

1 Answers1