The number of blocks (size) for a snapshot is always the same as the one for the base disk.
Original Message:
Sent: Dec 03, 2024 05:34 PM
From: Clay
Subject: Snapshots, Datastore Expansion and Broken VM
André,
This has worked, thank you. I have a couple questions, if you'd be so kind as to indulge me. I also created a descriptor of my own using vmkfstools, but it has two notable differences:
- The disk size I used (obtained from VM1-000001-sesparse.vmdk) resulted in a different extent integer than yours - RW 754974720 (yours) vs RW 617205568 (mine). How did you obtain this?
- Mine also contains the "ddb.longContentID" string at the end, the value of which I adjusted by referencing the appropriate CID. Yours does not have this, is that normal?
Thanks again. Always grateful to learn.
Original Message:
Sent: Dec 03, 2024 05:00 PM
From: a_p_
Subject: Snapshots, Datastore Expansion and Broken VM
I've created the missing file (see attachment) from the contents of the ones that you've provided.
André
Original Message:
Sent: Dec 03, 2024 04:39 PM
From: Clay
Subject: Snapshots, Datastore Expansion and Broken VM
That makes sense, this missing descriptor is breaking the chain. Thank you.
I'm eager to see the file you create so I can learn from this. Files attached.
Original Message:
Sent: Dec 03, 2024 04:21 PM
From: a_p_
Subject: Snapshots, Datastore Expansion and Broken VM
One of the descriptor files (VM1-000001.vmdk) is missing. It should be possible to re-create it, from the content of VM1.vmdk, and VM1-000002.vmdk.
Please compress/zip these two files, and attach the .zip archive to your next reply.
André
Original Message:
Sent: Dec 03, 2024 12:21 PM
From: Clay Anderson
Subject: Snapshots, Datastore Expansion and Broken VM
Hey everyone. I'm troubleshooting a single VM on a solitary ESXi 7 host.
A user called me saying "the VM won't boot" citing the error: "Unable to enumerate all disks". Apparently the datastore was expanded while running on a snapshot. I don't have any more information.
When trying to add the virtual disks back into the VM, all fields in the disk section appear blank, and I'm not able to choose "ok". I suspected a corrupt/missing descriptor, though I see one is present. I am unable to delete any snapshots, instead receiving the error "Failed to delete all snapshots from virtual machine 'VM1'."
Here is some relevant output from the host:
[root@localhost:/vmfs/volumes/65119fc8-8f1edd98-a818-c45ab1acf059/VM1] ls -lrht *vmdk*
-rw------- 1 root root 536 Jan 8 2024 VM1.vmdk
-rw------- 1 root root 360.0G Jan 19 2024 VM1-flat.vmdk
-rw------- 1 root root 294.3G Nov 14 13:17 VM1-000001-sesparse.vmdk
-rw------- 1 root root 332 Dec 2 16:32 VM1-000002.vmdk
-rw------- 1 root root 15.3G Dec 2 19:58 VM1-000002-sesparse.vmdk
[root@localhost:/vmfs/volumes/65119fc8-8f1edd98-a818-c45ab1acf059/VM1] ls -lh
total 485044416
-rw------- 1 root root 294.3G Nov 14 13:17 VM1-000001-sesparse.vmdk
-rw------- 1 root root 15.3G Dec 2 19:58 VM1-000002-sesparse.vmdk
-rw------- 1 root root 332 Dec 2 16:32 VM1-000002.vmdk
-rw------- 1 root root 24.0G Jan 19 2024 VM1-Snapshot1.vmem
-rw------- 1 root root 7.5M Jan 19 2024 VM1-Snapshot1.vmsn
-rw------- 1 root root 24.0G Nov 14 13:18 VM1-Snapshot4.vmem
-rw------- 1 root root 7.5M Nov 14 13:18 VM1-Snapshot4.vmsn
-rw------- 1 root root 360.0G Jan 19 2024 VM1-flat.vmdk
-rw-r--r-- 1 root root 264.5K Dec 2 16:33 VM1.nvram
-rw------- 1 root root 536 Jan 8 2024 VM1.vmdk
-rw-r--r-- 1 root root 765 Dec 2 20:32 VM1.vmsd
-rwxr-xr-x 1 root root 3.3K Dec 3 16:59 VM1.vmx
-rw-r--r-- 1 root root 28.0M Oct 12 07:01 vmware-15.log
-rw-r--r-- 1 root root 44.3M Nov 14 13:20 vmware-16.log
-rw-r--r-- 1 root root 24.1M Dec 2 16:32 vmware-17.log
-rw-r--r-- 1 root root 161.9K Dec 2 17:01 vmware-18.log
-rw-r--r-- 1 root root 56.8K Dec 2 17:01 vmware-19.log
-rw-r--r-- 1 root root 56.6K Dec 2 17:02 vmware-20.log
-rw-r--r-- 1 root root 56.9K Dec 2 19:58 vmware.log
Any direction would be appreciated.