We copied over some VMs from an older 3.5 ESXi system, to an NFS mount for later import.   2 of them failed.  One, I rebuilt the *.vmx file and this worked.  The fourth (and most important) one fails.  All of the data are there, they are correct to the best of my ability to check.  The *.vmx file is there, the Snapshot and delta files are there.

When I go to browse the directory (use existing disk) when creating a new VM, it doesn't even recognize that there is a valid vmdk/disk in there.  Corrupted *.vmdk files?

I read several articles about using the log to reconstruct the *.vmx file (it's not corrupted to begin with).  I'm very puzzled.

Since I created 4 identical hosts, I tried importing the vmdk disk (flat) into a working config, no go.  I think because it has a Snapshot and delta that it just wouldn't work anyway.

I also tried installing the vSphere API tools for the vmware-mount command and didn't get very far.

So I'm wondering what else I might be able to do to get at the data - there is some in there we need to retrieve (database).

I know there are only 2 partitions on the image: swap and /.

I literally tried everything I could read and think of.  Creating a new system, importing the disk files, with newly-created metadata, etc.  Nothing works.

Thanks.