This is a post moved over from my old blog — still relevant.
I love P2V. It works like a charm — except when it doesn’t…
When it fails, it usually fails in the strangest ways possible, the errors are obtuse, and finding the underlying cause is a nightmare. Well, I ran into the issue where the P2V would fail referencing an error with the snapshot and/or the disk ID. Well, I assumed this was related to the Source VSS Snapshot that is taken during the P2V process. After spending several hours tracking that down, and realizing I was barking up the wrong tree I started looking to the destination. Well, there was the problem. A snapshot in the target machine. I was pretty sure that the P2V process did not use snapshots on the target — I mean, why would it? Time to look elsewhere.
Well, turns out that elsewhere was the SAN that was housing the storage for the target datastore. The array, in this particular case a Nimble SAN, has a feature where you can quiescence the VMs on a LUN using the Native VMware snapshots to allow for better point in time recovery options with the SAN based protection. If this is on, it tries to take a snapshot of all the VMs on the target LUN. Now, if you are doing a small VM that will convert in between the snapshot window, no issue. If it is a larger machine — turn that feature off during the P2V window and save the headache.