VDisk showing as failed on UCS c240 offline diafnostics

[Andrew Beardsley]

We have a newly rebuilt UCS C240 M4SX which is showing a PSOD in vSphere ESXi 6.0.

When I ran offline diagnostics on the server, it is showing failed status for all of the VDisks (apart from the boot disk). Tests are passing for all of the physical disks backing the vdisks. The vdisks are raid 0 containing 1 physical disks.

The diagnostics log gives:

SEL/MCE log indicates no error

Test failed

Any pointers where I should be looking for what is causing this? My searching so far has not shown what would cause this error.

[Evan Mickel]

PSODs can have a wide variety of causes, typically what we would look for would be:

1)  Contents of the OBFL log from around the time of the incident.

2)  An image of the PSOD output shown on the KVM of the server.

3)  Contents of the vmkernel logging from around the time of the incident.

Searching for VMware KB articles against lines from the PSOD output can help in such a situation.  This could be related to the controller physically or the driver.  There could also be a hardware fault against the controller, cables, riser, slot etc...

The answer unfortunately for the moment is that there isn't quite enough data here to provide an estimate as to what it might be with any accuracy.  I would recommend opening a service request against the server.  From there TAC can assist in narrowing the situation down for you.

Thanks!

[Andrew Beardsley]

Thanks for the reply. I did in fact log a service request. Turned out to be a faulty CPU.

[Evan Mickel]

Glad to hear it got resolved!  Thanks for getting back to close the loop here, have a great weekend!