Snapshot issue

[sheehanje]

I actually have a couple questions, but first my most important. Is there anyway to get a job to abort a snapshot removal process automatically. I just went in to restore a guest level file for someone on a file server and noticed it hasn't been backed up in almost a week. I usually check for failed jobs every weekday, so started investigated why this happened. Apparently we had a job running for 140 hours, and 139 of that was a stuck snapshot removal on one virtual machine. The task ended up failing eventually, but it was running for days. I didn't get a failed mail on the job because it hadn't actually officially failed yet. This particular job backs up 8 of our file servers - and I feel very vulnerable with this type of issue.

I checked the VM in question - and it actually didn't have a snapshot on it - so I'm wondering if it's an issue with VMWare not reporting back at times to VEEAM.

I just added a backup window during the work week - and hope this actually cancels the snapshot removal - but I know from past experience that I've had to actually reboot the server at times when snapshotting has problems.

We are running Veeam Backup & Replication 6.5 on VMWare Vsphere/ESX 5.1

Thanks in advanced for any suggestions on resolving this issue.

John

[veremin]

First of all, please be aware that the problem of snapshot removal isn’t usually related to Veeam. Therefore, it seems reasonable to reproduce this situation in order to understand whether these issues should be addressed by Veeam or by VMware. So, you can take the snapshot of your production VM manually, keep the snapshot open for long enough time before deleting it, similar to time it takes to backup the VM, and see if you experience the aforesaid issue or not.

Is there anyway to get a job to abort a snapshot removal process automatically.

No, I don’t think so. Assuming it were possible, there would be a variety of problems connected to it.

However, there are a couple of options available fo you:

• Specify backup window.
• If you have Veeam One installed, there is a predefined alarm that tracks job duration. You can specify the required time interval and set a PowerShell script as a post-alarm action to terminate the backup/replication job. Previously, we (me and Vitaliy) have created a detailed plan how it can be achieved, so let me know if you need any assistance.
• Create a custom Powershell script, that will check in a cycle a job duration and terminate it in case job exceeds the specified interval.

Anyway, if you haven’t deployed VeeamOne yet, it might be worth doing it now, since it provides you with up-to-date information regarding your Virtual Infrastructure, such as obsolete snapshot presence, resource utilization and many other.

Hope this helps.
Thanks.

[sheehanje]

Thanks for the insight. The one problem I have noticed is that a snapshot isn't even present on the machine it says it's removing a snapshot from. I just installed Patch 3 - so hopefully that may resolve the issue.

Last night all jobs executed properly - but I will keep a close eye on the jobs for the next few weeks and see if the issue presents itself again.

Thanks,
John

[veremin]

Glad that nowadays everything works as expected.

Anyway, keep us updated about the results you get. Thanks.

[sheehanje]

Well, still stuck in the same boat. Had another job hang with one VM supposedly not removing a snapshot - but according to vCenter no snapshot was being removed.

Also, it seems my backup window mattered none - seeing the job was stuck for 149 hours before I killed it.

This is a huge issue as it's not just affected the 1 VM in question, but the whole job - as well as other jobs.

Hopefully we find a solution soon.

[ecowgill]

Hello John,

I encourage you to create a support request here so the support team can review the log files. Please follow-up by posting your support case ID here.

[veremin]

I’m also along the same line in recommending you to open a support case; since this situation looks exactly as the one where deep log investigation is required.

Thanks.