Left behind snapshots are not removed by following job runs

[zak2011]

Support has to still find out exactly the reason why the left over snapshots were not removed. They have investigated the logs both from veeam as well as well as the host logs and strangely it doesnt show anything about the failure to remove Veeam snapshots. They have asked me to upgrade the build of vcenter from the previous version to the new version.

Now i have another two VMs showing up as warnings in Veeam One saying ' snapshot age'. However when i go to vsphere i dont see any snapshots for these two VMs.
Any reason why Veeam One shows snapshot age and vsphere doesnt show any snapshots?
Thanks!

[veremin]

However when i go to vsphere i dont see any snapshots for these two VMs.

Have you tried to take instead a putty session to the host and see whether there are some existing snapshots of this VM on the datastore?

Hope this helps.
Thanks.

[zak2011]

Have not taking the putty session to the host. Were you referring to the following KB?
http://kb.vmware.com/selfservice/micros ... Id=1005503
I think this may require me to power down the VMs inorder to avoid generating much I/O. Since they are production servers, i wont be able to do that now.
I have checked the properties of the hard disk of these two vms and it just shows up as normal vmdks. Wondering why Veeam one shows the warnings though.

Thanks.

[veremin]

It might happen that under certain circumstances vSphere-client doesn’t show snapshots clearly, so, it's SSH-client that should be used in this case.

The virtual machine should be powered off at this point because having the virtual machine running generates more I/O, causing the snapshot files to continue to grow.

In fact, checking the presence of snapshots on the datastore via Putty utility or other SSH-client doesn’t require powering off VM and the abomentioned note is related primarily to storage space considerations.

Hope this helps.
Thanks.

[zak2011]

I have tried running the command from powercli

Get-VM | Get-Snapshot | Select Created, VM

It doesn't show that any snapshots are present though!

[tsightler]

Yes, I have seen a couple of cases where the information reported by vCenter is incorrect, almost like some type of strange locking issue. Even the information shown in the File Browser was not consistent with the host. Easiest thing to do was to enabled the local or remote service console and login, although another alternative might to connect with the vSphere Client directly to the host rather than vCenter and see what if that shows something different.

[veremin]

I have tried running the command from powercli.

Information received by PowerClI is taken from the vCenter or ESXi host to which PowerClI is connected (Connect-VIServer). Therefore, I’d still recommend verifying it by one of the aforesaid methods.

Hope this helps.
Thanks.

[zak2011]

I tried connecting directly to the host with the vSphere client, however it still doesnt show any snapshots.
I will try the SSH method as Vladimir mentioned and update the post.

[Vitaliy S.]

Any reason why Veeam One shows snapshot age and vsphere doesnt show any snapshots?

Veeam ONE takes VM snapshot information from the MOB (Managed Object Browser). If you do not have any snapshot files when looking at the datastore with vSphere Client -> Datastore Browser, then I would suggest contacting our support team to get the instructions on how to check snapshot presence via MOB, this would allow to isolate the issue.

[zak2011]

Thanks Vitaliy. The Veeam One SE was able to check for snapshot presence via MOB and confirmed the state of the snapshot was unset.
What happened was the login credentials to the vcenter on Veeam One had expired due to which it was showing old data. Once i inputted
the credentials again the warnings disappeared from the console showing the updated information.
Thanks!

[zak2011]

However the original case of snapshots being left still is a mysetry.
The issue started out with left over snapshots from exchange server. Once the patch was released on Dec24 , seemed like it resolved the problem.
After this twice i got left over snapshots from Veeam, and support has indicated after weeks of investigation vcenter should be upgraded from its current build.
According to what support says, they cant seem to find any indication of these snapshots both in the Veeam logs as well as the hostd logs of the esxhost.
Not exactly what I was expecting!