My first PSOD

[Derekius]

I've never seen a PSOD until today, and I have no idea how to diagnose it.  I am running vSphere Essentials 5.5 on a single HP DL380p Gen8 server.  I have a 2012 VM and a 2012 R2 VM.  A colleague was trying to install the SQL Analysis services on the 2012 R2 VM and we got the PSOD.  After rebooting the host, we retried the install of SQL Analysis services and it crashed in the same spot.  This could be coincidence, but regardless, I just Google'd my way into finding the dump file, but now I have a rough screenshot of the PSOD and a 60MB dump file that I don't know how to read.

I can't reach the guy that registered the license, so I can't open a support request with VMware until late tomorrow morning and I'm posting here in hopes I don't have to wait that long.  Ideas?

[Derekius]

As this was being posted, we tried the install one more time to see if we could generate another crash dump, just in case it was needed (better to break it intentionally, after hours, than by mistake, during business hours).  Naturally the install went smoothly and I have no idea what caused the crash.  I've used WinDbg to troubleshoot BSODs in Windows; is there a similar tool to analyze an ESXi crash dump?

[john23]

check this thread looks like a known issue:

https://communities.vmware.com/thread/465241

[Derekius]

While I any assistance with this problem, that link doesn't seem related to this problem.  Their PSOD specifically mentioned heartbeat issues and PCPU lockups, but none of that is in my PSOD.  The only commonality I see is that these are both happening on HP ProLiant Gen8 servers.

[grasshopper]

Hi Derekius,

If you have not done so already, please change the vNIC of the VM in question from e1000* to vmxnet3.

[GaneshNetworks]

Exception 14 is explained here - http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&docTypeID=DT_KB_1_1&e...

[Derekius]

The host crashed again, same error (PF Exception 14), so we took advantage of the downtime and switched the NICs to vmxnet3.  As for the link that GaneshNetworks posted, I already looked at that article before posting this.  It just says Exception 14 can be caused by a lot of things, and the only way to analyze the core dump is to open a case with VMware technical support.

While changing the NIC type, I noticed both my VMs are on hardware version 8.  I have searched, but can't find anything saying what OS's are supported for the various hardware types.  This is a standalone 5.5 server with no vCenter server, and the VMs on it are 2012 and 2012 R2.  My VIBs seem to be up to date, so I'm at a loss.  Any other ideas before we resort to opening a case?

[Derekius]

https://communities.vmware.com/message/2289019#2289019

https://communities.vmware.com/message/2299661#2299661

Changing the NIC to vmxnet3 seems to be a pretty common recommendation after looking at more threads.  And now we wait...

[admin · Accepted solution]

Hi,

Guess this should help

http://kb.vmware.com/kb/2059053 < ESXi 5.x host Experiences a purple diagnostic screen with errors for E1000PollRxRing and E1000DevRx  >

Thanks,

Avinash

[Derekius]

Not until after reading the link you posted did I notice my PSODs are riddled with 'E1000' lines.  Haven't seen the issue since changing the adapter, and that KB article's example PSOD is very similar to the issue I saw.  Thanks for that article!