Request: Disk skipped because of unsupported type is "only" a warning

[CKESSLER]

Dear People,
hopefully this is adressed on the correct location here in the Forums.

I have a VM with RDM disks connected (Virtual, Dependend).
During a resize of the disk within VMWare, the disk was reattached, but unfortunately as "Independent" which VBR does not support.

The next run of the backup job skipped this disk now, but only with a warning:
09.06.2022 21:46:17 :: Disk XXXX.vmdk has been skipped due to an unsupported type (independent raw device mapping disk in virtual compatibility mode)


In my opinion this should be an error, since this disk gets not backed up at all.

What do you think? Other opinions?

Maybe Veeam developers could adress this in next releases.

Regards,
Chris

[Mildur]

Hi Chris

as "Independent" which VBR does not support

It's not a VBR limitation, it's a VmWare limitation. VmWare does not allow to snapshot an independent disk.
All backup applications rely on snapshotting the vmdisks to create a backup of the vm. Veeam has to exclude this disk like all other vendors to backup at least the other disks from the vm.

In my opinion this should be an error, since this disk gets not backed up at all.
What do you think? Other opinions?

An error means, something is wrong and hasn't worked. But the data we can protect was successfully backed up.
In my opinion, nothing is wrong in case of excluding an independent disks. It has to be excluded because of the limitation I mentioned.

Thanks
Fabian

[soncscy]

Agree, this shouldn't be an error -- we have a lot of client setups where we know the RDMs aren't going to be protected and this is perfectly fine -- the data is protected in other ways and it's just important to protect the other configuration data on other disks for rapid restore of a system state if required.

The alternative is manually creating exclusion rules for each virtual machine, so in fact I'm very happy this is a one-off warning.

But, a small request, it'd be great if there was a way to report on unsupported disks on backups or reset the warning. There is the VM configuration report in VMware, but as best I know this just assesses the live virtual machine from the environment and doesn't warn if it's in a backup, so we have to write some logic to connect this report to the active jobs, which is pretty tricky when containers get involved. It's not uncommon that I have application owners who submit their server for backup and don't know about these limitations, the backup team maybe forgets to note to the owner about the warning, and then subsequent runs don't give a warning.

We already have mitigated this with powershell but it'd be really great to have an on-demand report for this built-in or the ability to reset the warning from the UI on a given job just in case we miss it. Ideally it would print the VM, the disks (VMware names), and the job the VM is included in. CSV format would be fantastic