That is what I was suspecting.
Followup to this, my client went and created a single node MS Cluster under VM, and used vmdk disk for the MS Cluster. They got away with this because it was a single node cluster. They then executed a snapshot manually via VDP and said it worked (I am just now thinking maybe it was not via VDP). Next they then configured supposedly the same exact thing in VDP using VDP Appliance Proxy. When executed, this resulted in an immediate MS Error 1038 - Physical Disks Lost and the cluster was unusable and took many hours to recover, eventually restoring via TSM the system state from several days earlier and then taking the disk resources out of the MS Cluster.
The client is still not convinced the snapshot via VDP should have anything to do with this occurring. My general logic is a snapshot needs to communicate to put the resources like disk in a quiet state and take control of the disks to start the snapshot and then hand them back , all in a defined controlled manner. If VDP thinks it needs to do that with VMware owning the disk, but actually the MS Cluster "owns" those resources, VDP very well could (and did) cause the MS Cluster to lose the disks (i.e. the 1038). I know the actual mechanics may be different, but something along those lines.
If you or anyone can shed any light on what were the actual technical aspects that would cause VDP executing a "Backup / Restore VM" command as shows in the VDP Log, that would cause an immediate MS Err 1038 - Physical Disks Lost (my general description), I would greatly appreciate.
