Feature Request: Ability to map VM's in the backup job

[ratkinsonuk]

Within the Backup Job Virtual Machines tab, you can hit recalculate and it will show you any VM's that no longer exist (0 Bytes). Often, this is because the VM has been restored or cloned.

The problem is, to add the VM back into the job, we have to delete it and then add it back in so it picks up the new UID. In turn, we then have to make sure any server-specific settings get re-applied, which can be a chore at best.

Could we have a 'Map' button instead that allows us to link the existing entry to a server. This would work in a similar way to how you can map backup jobs to an existing repository when backup files get copied or cloned (https://helpcenter.veeam.com/docs/backu ... ml?ver=120)

Just to add, I found this vmware-vsphere-f24/manually-updating-mo ... 61851.html entry, but I think it should be something native to the GUI and supported.

Thanks, Rob.

[Mildur]

Hi Robert

Just to add, I found this vmware-vsphere-f24/manually-updating-mo ... 61851.html entry, but I think it should be something native to the GUI and supported.

For vCenter migration where the virtual machines remain the same object with the same vDisks, we have since v12.3 a supported PowerShell cmdlet.

Often, this is because the VM has been restored or cloned.

A restored or cloned VM is always a new object with new vDisks. Mapping it to a backup chain from a different VM could be risky.

The problem is, to add the VM back into the job, we have to delete it and then add it back in so it picks up the new UID. In turn, we then have to make sure any server-specific settings get re-applied, which can be a chore at best.

May I ask how often that workflow (recovery or clone) is performed? Additionally, do you typically use dedicated settings (such as guest processing and disk exclusions) for each protected VM?

Best,
Fabian

[ratkinsonuk]

Morning Fabian. I've had a look through your response, and the set of Powershell tools you've developed. They will certainly be helpful if we do another vCentre migration. I think we've done 3 major migrations in the past from memory. I'm reading the notes that it wouldn't specifically help with my current scenario.

If a mapping tool is risky, could a re-sync tool be developed instead. Basically, it would do what we have to do manually - delete the existing VM called 'xxxxxx' from the backup and re-add it with the same settings?

In response to your question on frequency and impact, I guess we have to reset the UID on servers once or twice a month. That's for an estate of around 600 VM's. We do use specific guest processing and disk settings across all of our servers, but try and put comments in the description field for anything non-standard such as the one below for the backup job of the Veeam B&R servers.

Code: Select all

Veeam Backup & Recovery application servers
- Proxy servers only backup system and swap disks to stop 'Hot Add' disks being backed up
- Amazon Gateway, only backup system and swap disks to stop 'offload' data being backed up
- VEEAM_SERVER_01 includes a backup of the Metadata disk (E: drive)

I hope this helps you to understand the requirement better.

Thanks, Rob.

[ratkinsonuk]

Sorry, I meant to add. I could probably create a Powershell script to collect the existing job settings, delete the server from the job, and then re-add it back in using those settings. I already have most of the PS code as I use scripts to create the backups in the first place.

But, that wouldn't help the community as a whole who I'm sure run into similar problems. A global solution would be really good.

Thanks, Rob.