VMware ESXi 5.5 PSOD -- PCPU locked up -- What's causing this?

[triggerim]

Dear community,

My company has been experiencing some difficulties running ESXi 5.5 on HP DL160 hardware.

We have always been running on Dell hardware but since we have a lot of "old" HP hardware lying we would like to use it in building up more infrastructure, it's still good hardware.

I had till machine installed for some month ago and it has since then crashed about 3 times and now I'm pretty tired of it.

I have patched ESXi 5.5 with all the newest updates I could find, updated BIOS firmware to the latest I could find but it still crashed on me on a 1 ½ week basis.

The third time was today and now It's really stretching my patience too maximum.

I captured the attached image on how the crash looks, it has said "PCPU locked up 4", "PCPU locked up 6".

I'm running 2 x Intel Xeon E5620 CPU's with about ~74 GB RAM.

All hardware is green under the "Health Status" in VMware vSphere -> Configuration.

I installed ESXi with HP-ESXi-5.5.0-iso-5.72.27, I then updated this with all new ESXi 5.5 updates I could find.

It crashed both before and after these updates, so I upgraded the BIOS firmware to the latest version I could find.

Is it possible that any of my two CPU's is broken?

Thanks for any help!

[sgunelius · Accepted solution]

Could it be something as simple as incompatibility?  Which generation of the DL160 are you using?  If you're not seeing hardware-related issues displayed at POST, have you tried booting from the latest Service Pack for ProLiant (SPP) and run diagnostics to make sure the hardware is fully functional?

[triggerim]

Hello Sgunelius,

Sorry for my late reply.

I have had a lot of things to deal with lately, but now I have been able to try fixing it again.

I have used the HP 2014 SP3 Firmware CD to upgrade the server and have yet not experienced any issues, this was done yesterday.

I'm running a DL160 G6 for your information, If I run into this issue again after this upgrade I will see if I can get my hands on some kind of diagnostics tools.

I have no error messages about the hardware displayed at the POST so I'm hoping it should go well after this upgrade, otherwise I will run the diagnostics after that I think the only possible reason would be a CPU error/fault.

Thanks for your help!

[ramkrishna1]

Hi

Welcome to communities.

From you description come to know that your company wan to utilize old retired hardware  ,

If so please user esx 4.1 instead of latest version for compatibility point of view .

[triggerim]

Hi,

I would not consider a HP DL160 G6 with 2 x Xeon E5620 and fresh memory as retired hardware, it's not new but not that old to be called retired.

Should also be fine considered it's running on the HP version of ESXi, also I don't see how ESXi 4.1 would have better compatibility, I really hope that VMware development isn't going backwards.

Best regards

[triggerim]

Hello,

This happened once again today and I was hoping I had fixed the issue

What I did was upgrade the BIOS Firmware, replace some memory and it was running fine for about 3 weeks... now I got the same message again "PCPU locked up".

The only thing I can think of now is that one (or both) of the CPU's is bad, so I will start with removing one of them and wait and see.

If it crashes; I will remove/replace the other one too and see if that one is the problem.

If anybody has anything to add, please do so.

Thanks

[DJRammY]

This looks like incompatibility issues.

I also have a HP DL60G6 and I also experience this PSOD once in a while.

[triggerim]

It's sad

[DJRammY]

Check this topic:
i updated the driver... i hope it's solved now...

[triggerim]

Hello again,

I have switched one of the CPU's out and was running it for about 3 weeks without issue.

However it crashed again, but this one looks a little bit different than the last one.

I'm probably gonna try switching out the other one too, but I suspect it won't help...