Change datastore, VM not backed up

[rsberzerker]

Using B&R 8. I moved one of my VMs to a new datastore, although the host remained the same. In fact, nothing else changed. The next backup job with that VM failed (on that VM only). I was able to edit the job and delete the "old" VM and re-add it, but now B&R sees it as a totally different VM in the backup restore points. Is there anyway to consolidate the restore points and/or avoid this behavior? If I move it back and re-add it, will it continue using the old restore points?

[the_mentor]

how did you move the VM to a new datastore? storage vmotion?

[PTide]

Hi Robert,

how did you move the VM to a new datastore? storage vmotion?

DeMentor's question makes sense. With Storage vMotion MoRef ID on the host still changes.

Thank you.

[foggy]

How VM was added to the job (via vCenter or via host directly)?

The "old" VM will be subjected to deleted VMs retention.

[rsberzerker]

How did I move the VM?

In vsphere, right clicked on VM, picked migrate. Then change datastore, picked the new datastore, then finish. VM was on at the time, so I think that is vmotion.

How VM was added to the job (via vCenter or via host directly)?

Via the host. I right clicked on the job, selected VMs, clicked add, selected the host then selected the VM, then clicked add (and finish).

[Vitaliy S.]

When moref ID of the VM changes, it is treated as new one. If you have vCenter Server, then it is recommended to add VMs via vCenter Server connection to make your jobs "vMotion-aware". Moreover, adding VM containers to the job would be a better approach (rather than adding VMs explicitly). In this case all discovered VMs (from this container) will be automatically picked up by the backup job.

[rsberzerker]

I do have a Vcenter server. If I change my jobs to use that, will I run into the same issue (veeam sees it as a new VM), but now for all my VMs?

The VM containers sounds interesting. Never heard of it. Unfortunately, my google-fu has failed me. Do you have a link where I could read more?

[foggy]

I do have a Vcenter server. If I change my jobs to use that, will I run into the same issue (veeam sees it as a new VM), but now for all my VMs?

Yes, but you will avoid similar issues in future.

The VM containers sounds interesting. Never heard of it. Unfortunately, my google-fu has failed me. Do you have a link where I could read more?

You just need to add some VM container to the backup job (VM folder or resource pool, for example) instead of adding individual VMs.

[rsberzerker]

Thanks. I now understand what I have to do. Sadly, this looks like it is going to take a while as extra disk space is limited and I'm going to have to move the VMs to vsphere over time. At least I have a fix.