On my server I have several VMs on drive D:
. I'm using Windows Server Backup 2012 to back them up.
When I'm asked to select the items to backup, what's the difference between choosing the entries from the Hyper-V tree vs. choosing the partition, where the VMs are located on, itself?
I found out that when I leave drive D:
unticked, the backup will fail on most days (but not always).
So, is there a downside when I don't select the Hyper-V tree but just drive D:
instead?
Since all VHD(X) files and all settings are stored on drive D:
, this shouldn't make a difference, should it?
Best Answer
Backing up the D: drive simply attempts to back up all the files on the D: drive.
The Hyper-V specific methods of backing up VMs use the Volume Shadowcopy Service and VSS requestors and writers to allow the guest OS to intelligently participate in its own backup process. The success of this method is highly contingent upon the Integration Services/tools being properly installed and enabled on the guest OS.
If you do not backup the VMs through Hyper-V, another way to get good backups of your VMs is to simply set up Windows Server Backup on each one of your guest VMs themselves.
You will not get good backups of your VMs if you simply copy the VHD[X] files while the VM is running, because the VMs are in a volatile state, with data in RAM, etc.
As for why the backup is failing when set to a certain configuration, we would need all the relevant events and errors.