Last days I was busy with recovering data from a VMFS-volume that looked completely blank. The VMFS-volume was attached to a local RAID controller and the system was running ESXi 5.0.
The day before the problem there were about 20 vmdks on that volume.
I tried a force mount and resignature without success.
Next I tried vmfs-fuse and UFS-explorer but they did not see any vmdks either.
So I had to carve out the most important vmdks - that worked quite well as the customer only used thick provisioned vmdks and so I was able to recover 16 of the 20 vmdks.
Before I started the carving I did a check with a ESXi 5.1 LiveCD to be able to use Voma.
It reported an Ondisk error and mentioned a size problem of the volume - something like "On disk error. Size is x but should be y.
Question for next time:
Can I use that size detail to somehow make the volume be usable to ESXi again ?
As the data was still there I think there should be a way to get back the data that is less painful than what I had to use ...