ESXi 5.5
Size of VM = 285 GB
Datastore = 2TB free space = 465GB
Veeam 8.0
For the last 2 weeks I have a VM that fails to consolidate the snapshot (during Veeam backups). It retries (varying amount of retires depending upon which night) and is eventually successful. Other VM's on this Datastore backup, snaphshot and consolidate without a problem. The server is not busy at this time of night!
I would like to troubleshoot why this is happening in case I have an underlying problem. Can someone advise what how to troubleshoot?
Thanks.
Have you tried a manual consolidation from vCenter?
What is the nature of the machine? windows/linux? Application server ? DB?.
Error message you get in veeam/vcenter when the task runs.
Also share the output of directory list form VM folder from command line.
Does the VM's vmware.log show any details about the failed snapshot consolidation?
Please attach the current vmware.log to a reply post.
André
First we need to know which kind of OS it is. Also if you are using quiescing or not for the backup job. That will help us to determine what quiescing mechanism is being used.
Some time you can have errors related to VMware tools, or even a faulty Microsoft VSS writer for a specific application.
Also when in the process this is happening, is it in the beginning of the backup job or after it has copied the data but can not commit/consolidate the snapshot during the final stage.
Windows 2008R2
The Veeam back up job is not set to quiesce.
The job fails at the commit\consolidate stage, see my screenshot in previous post from vcentre.
I restarted the VMTools service early in the week, to no effect.
Server was last rebooted 25/3/16
Thanks
I recommend trying checking the VSS writers in the OS : https://kb.vmware.com/kb/1007696
Then I recommend upgrading to Veeam 9.0 and use the awesomely name feature Snapshot Hunter: https://helpcenter.veeam.com/backup/vsphere/snapshot_hunter.html
But then again its more like a workaround rather than a fix.
Hope this helps.
I moved the VM to another datastore the backup ran that night and no errors in respect to the snapshot. Moved the VM back to the original datastore and errors returned. There over are other larger VM's on this datastore and they do not error.
Investigated within the VM and event viewer (system) has errors relating to ntfs. See VM KB2006849