Hi all,
Bit of a strange one here. We have an alert to show any VM's that have a snapshot. I noticed that one of our servers was showing this alert and when checking the datastore there were in fact some snaphots that hadn't consolodated properly. No problem with this, I simply created a new snapshot and then deleted all snaphosts which resolved the issue. The VM was still alerting to the snaphot so I browsed the datastore and could see the following
I've tried the following to resovle this:
None of the above has worked. When i sVmotioned the VM to another datastore I selected the option to provision the .vmdk's as thin but when checking after the sVmotion it's still showing as thick. When attempting to sVmotion with the VM powered on I recevie the following error message:
Anyone has any other suggestions?
Thanks
Michael
What you also could test is clone your vm or use the converter to build it new.
That should also remove all of your snaps.
Frank
Hi,
please take a look at that post
Also you could not do a sVmotion when there are Snapshots on the vm.
Frank
Frank,
Thanks for the reply. I actually dont think any snapshots exist as I can sVmotion and there are no delta files in the directory. I think the snapshot alarm is a red herring:
What's confusing me is the 3 VMDK files with the -XXXXXXXXXXXXXXXXXX
The VM is using the base disks rather than the -XXXXXXXXXXXXXXX disks but I'm not to sure if I can just delete these disks
Michael
Because the files have current time stamps, it looks like they are indeed in use. I'd recommend you take a look at the VM's latest vmware.log to see whether these files show up there.
André
are you using any backup software for taking the vm backup? these can also create snapshots.
@ André - I think the timestamps are showing current as this was when the sVmotion was performed. I'll check the log files and report back
@ Gkeerthy - Yes, we use Backup Exec 2010 but I don't think it'sa snapshot issue
Michael
I know this is not the correct place to be making this post, but I am getting constant emails everytime a new post is posted on this forum. I have tried to adjust my settings so that I don't get emails for every new post, but surprisingly I can't find the option to change this preference. I've spent the last 3 weeks accepting the fact that I'll have to delete each email as it comes in, which is very annoying.
Can anyone poit me directly to the place to change this preference? Thanks.
Dan
I've checked the latest vmware.log file and can see the following references to the strange disks:
@danw76
You can check your email notifications when you open "Profile" --> "Manage Email Notifications"
The direct link for you should be: http://communities.vmware.com/people/danw76?view=watches&numResults=15&watchObjectType=14
André
Not sure what caused this issue (I'd guess the backup application did), but your current virtual disks have snapshots with the names in an unusual order, where the delta files have the names which are usually assigned to the flat file!? In addition to this the virtual disk size
NEXDBTV01.vmdk -> NEXDBTV01-delta.vmdk
NEXDBTV01-ac1eb6efaa73e4a.vmdk -> NEXDBTV01-ac1eb6efaa73e4a-flat.vmdk
NEXDBTV01_1.vmdk -> NEXDBTV01_1-delta.vmdk
NEXDBTV01_1-9ace2ad6c00b0c6.vmdk -> NEXDBTV01_1-9ace2ad6c00b0c6-flat.vmdk
NEXDBTV01_2.vmdk -> NEXDBTV01_2-delta.vmdk
NEXDBTV01_2-c90359261cca5b0.vmdk -> NEXDBTV01_2-c90359261cca5b0-flat.vmdk
What I would do is to run "Delete All" from the "Snapshot Manager" to merge the snapshot files into the base disks and then clean this up by either "Storage vMotion" the VM, which - depending on the version - should rename the files or clean this up manually (let me know if you need assistance with this).
André
I've created a temp snapshot and "deleted all" and I'm currently in the process of sVmotioning the VM to another datastore with a different block size to ensure VAAI is not used in the copy.
I'll update tomorrow with the results. It looks like the disks are finally changed to thin but the -flat file is still showing as -XXXXXXXXXXXXXX.
Michael
After creating a new snapshot and sVmotioning the VM to another datastore with a different block size the VM is still showing the same disks
Anyone got anything else I can try before raising it with VMware support?
Michael
Did you run the "Delete All" with the VM powered on? In this case the vmware??.log file should contain information what happened while deleting the snapshots.
André
Hi,
take a look at this post.
http://networkadminkb.com/KB/a347/how-to-fix-virtual-disk-is-in-link-cloned-mode-error-message.aspx
Frank
What you also could test is clone your vm or use the converter to build it new.
That should also remove all of your snaps.
Frank
Jimknopf's suggestion of using converter/cloning should work since the delete all snap doesnt
@ JimKnopf99 - cloning worked.
Very strange but I thought the sVmotion would have produced the same reslts as a clone operation but obviosuly not.
Thanks for all your help guys.
Much appreciated.
Michael