Thick provisioned disks acting like Thin

[Micric]

I have had a strange situation this morning and I hope someone can shed some light for me.

I have a virtual server with multiple large HDD's (300GB and 200GB).  Both HDD's were created as thick provisioned, as far as I remember though they show as thick provisioned in the server properties.  The data is stored on an iSCSI SAN.

This morning I came in to find one of my virtual servers had frozen as it had "Insufficient disk space to write".  I looked at the store and found the although the disks were provisioned as thick they are infact behaving more like thin disks.  In the store they show as 110GB and 202GB instead of 200GB and 300GB.

Any thoughts?

[a_p_]

Can you please post a screen shot of the datastore browser windows showing all files and details (name, extension, sizes, time stamps)? It would also be helpful if you'd attach a current vmware.log file from the VM's folder (make sure it doesn't contain any usernames/passwords in the annotations)

André

[Micric]

As you can tell the logs are not showing as being close to current.

I will endevour to get some data from the current vmware.log for this system though.

Message was edited by: a.p. - moved pasted log output to a txt file for better readability

[a_p_ · Accepted solution]

From the screen shot it looks like your VM has 3 virtual disks (60GB, ~256GB, 350GB). From the size of the snapshot files (the screen shot does not show the time stamps), I assume there has been an issue some time ago (one of the snapshots is already ~67GB in size) with deleting the snapshots after a backup operation.

You should immediately take action to get rid of the snapshots! However, to find out whether it is save to just delete the snapshots from the snapshot manager (it may currently not display any snapshots although you have some) I need to know which version/buils of ESX(i) you are currently running and how much free disk space do you have on the different datastores (the virtual disks are located on different datastores)!

Does the VM currently run without issues?

Please attach (don't paste) the latest vmware.log file from the VM's folder (see your screen shot).

André

[Micric]

Bingo, the HDD are located on another store.  I had a good look through and have discovered the HDD. 

The snapshots seems to be part of Veeam.  As such they have to remain as they are.

Thank you for your help in this.

[a_p_]

The snapshots seems to be part of Veeam.  As such they have to remain as they are.

Although I don't know your setup, I'm 100% sure the shown snapshots exist due to an issue with deleting them after a backup operation. Usually snapshots should only exist while the backup process is running. From the size of the snapshots in the screen shot (up to ~67GB) this is definitely not the case.

What date/time is shown for the base vmdk's in the datastore browser? This is most likely the time when the issue occurred.

André

[Micric]

The snap shots were old, April and May, so they were removed and everything seems to have returned to it original state.

The server was running perfectly even with this problem.  Just a little suprised, but I now know to keep an eye out for those errant restore points.

Thank you for your help Andre.