vmfs3 datastore keep shrinking

[esbee]

Hello Guys,

I've noticed several times that the vmfs3 datastore keeps shrinking. There are 3 windows 2k8 servers running on the (2) hosts (not using VC)

I think the problem is that they all are using thick disk provisioning (I thought this is the standard option when you create an new VM, I didn't take the time to figure oit what was the best choice.... so now I've a problem )

During VM creration I gave al the 3 VM's a disk size, one VM with a provisioned size from 50GB, maximum size usable space is 118,96GB,

the second VM with a provisioned size from 250GB, maximum size usable space is 256GB and the last vm with a provisioned from 60GB, maximum size is 128,96GB. At this time there is only 68.96GB space left (last week 69.96)

How can I prevent that the VM's are taking every week a litle bit more storage? Do I have to migrate the disk into thin disks?

Can anyone please help me how to solve this problem so I'm not running out of diskpace within a certain time?

regards,

Steven

[idle-jam]

Hi, welcome!

Think provision will preallocate the unused space hence the VMDK will fill up the value defined in the VMDK. This should not see an increase of VMDK as it has being "fixed". the only reason i see is that the space could be used up by snapshots.

Can you check that the VM does not have snapshot turn on? if so please delete/commit it and you will see the space coming back.

iDLE-jAM | VCP 2, VCP 3 & VCP 4

If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points

[a_p_]

I agree with , if the disks are thick provisioned, there are only two situations in which additional disk space is used.

First - and most likely - is the use of snapshots. Right click the VM in the inventory and open the snapshot manager, if you see a snapshot and you don't need it, use the Delete button to commit the data to the base disk.

The second cause for additional disk space would be, if you have issues with the VM's and core dumps are created. Open the datastore browser and take a look at the VM's folder. If you see "core" files, then you should start troubleshooting.

If you don't see any snapshots in the snapshot manager, but you see vmdk files with e.g. "...-000001.vmdk" in the datastore browser, then take a screenshot and attach it to your next post.

André

[esbee]

Hello,

In attachement you can see a screenshot from from the VM folders and this on both esxi 4.1 hosts.

I don't see any core dumps in the datastore, so hopefully somebody can help me to throubleshoot?

If I need to post the content from the logfiles, let me know it.

regards,

Steven

[idle-jam]

Did not see any snapshots being turn on. btw can you tell us how much of space are you loosing in your VMFS say daily and weekly?

iDLE-jAM | VCP 2, VCP 3 & VCP 4

If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points.

[esbee]

There are no snapshots on the VM's, I checked this on both hosts.

Last week it thought it was around 2GB, this week 1GB. The past weeks I can't remember...

(during the weekend I started to write down the storage decrease in an excel sheet to get a better overview)