ESXi 5.5 hosting Server 2012 - IO operation errors

[ATG123]

My Windows Server 2012 which is hosted by ESXi 5.5 keeps restarting unexpectedly.  I don't believe the host is restarting, only the VM.  ESXi logs do not indicate a restart.  This is the error I am getting in my server 2013 event logs when the restart occurs.  ESXi is complaining about a scratch partition not being configured, but I have never had this configured.  This installation is a few years old and things have been fine until now.  Any ideas would be appreciated. I think these errors are related:

Windows 2012 server event log error:

The Windows The IO operation at logical block address 0x3a38ead0 for Disk 1 was retried.

ESXi error:

Device or filesystem with identifier mpx.vmhba32:C0:T0:L0 has entered the all paths down state.

[dbalcaraz]

Hi,

Seems a misunderstaing about what's happening, read about APD (All paths down) and PDL: VMware Knowledge Base

The problem can be with the storage attached to the ESXi host and it seems more clear in the W2k12 when it says DIsk 1 was retried.

Please check /var/log/hostd.log and review the errors.

[dbalcaraz]

Forgot to say that obviously the HBA attached to the host could be affected.

Try to vMotion this VM to another host and see if you have the same problem.

Check the ESXi's HW in order to see if there is any error.