Backup copy resume backup if the job runs out of the window?

[csinetops]

I just want to verify this is how Backup Copy jobs work. I have a VM with a very large disk of 3.2 TB. If the backup copy job doesn't finish in it's 24 hour window, it will fail but then pick up copying that disk where it left off, correct?

Also, can you process more than one disk at a time in a Backup Copy job? I've had copy jobs running on smaller VM's and never really looked at the jobs until thsi larger VM and noticed its only processing one disk at a time. It would make the job run much faster if it would process more than one disk at a time.

Thanks!

[veremin]

If the backup copy job doesn't finish in it's 24 hour window, it will fail but then pick up copying that disk where it left off, correct?

Correct.

I've had copy jobs running on smaller VM's and never really looked at the jobs until thsi larger VM and noticed its only processing one disk at a time.

Was it a direct or WAN Accelerated backup copy job?

Thanks.

[csinetops]

It's a direct job from our primary backup storage to our NetApp AltaVault, No WAN accelerators used.

[foggy]

In case of direct data transfer, VM disks are processed in parallel (Veeam B&R starts one data mover per source disk to retrieve and transfer data to the target data mover). If per-VM backup chain option enabled, Veeam B&R starts one target data mover for each VM and at this stage each VM disks are sequentially written to the target storage.

[veremin]

In this case, the said behaviour looks unexpected, as VMs should have been processed in parallel, as Foggy's said. So, it might be worth reaching our support team and letting them investigate it directly. Thanks.

[csinetops]

Hey v.Eremin, let me clarify. The VM's are processed in parallel during the job but unlike a normal backup job the disks are processed one by one. I'm sure that is by design and for a reason. It would however be nice on bigger jobs where we have a VM with 4 2+TB drives to process more than one at a time to speed up the job.Just a observation.

[foggy]

As I've mentioned above, VM disks should be processed in parallel (unless, for example, the concurrent tasks limit comes into play).