Options for backing up a VM that gets recreated daily?

[sid6point7]

So I have a process that deletes and recreates a VM every day. This means that while the VM Display Name remains the same, the moref ID changes daily. I set up the backup job to backup based on VM tag and it has been backing up up every night successfully. The problem is, the VM shows up multiple times with only one restore point each. This means that the older VM restore points will never age out of the repository since I have a 30 restore point retention.

What are my options here?

Thanks for the help guys!

[DaveWatkins]

My immediate question is "why?"

Why not revert to a known snapshot state instead of deleteing and recreating the VM, then it'll always be the same VM, but you get back to your known state (which I'm assuming is the goal)

[sid6point7]

@DaveWatkins - The VM is a clone that comes in from a different datacenter each night. We do not want to use replication for this.

[veremin]

Hardly can we do something about constant morefID changes. Each time it will be a new VM, from our perspective, and you will get exactly what you're getting at the moment - one VM with one restore point.

However, may I ask you why you're using cloning as replication means? Do you find it more convenient or more reliable to clone VM, delete previous clone, etc., rather than just to replicate it?

Thanks.

[nefes]

Since it is full backup every time, I suppose VeeamZIP with retention could fit your goal.

[foggy]

If you're using reversed incremental or forward + transform, the deleted VMs retention could also do.

[sid6point7]

Hardly can we do something about constant morefID changes. Each time it will be a new VM, from our perspective, and you will get exactly what you're getting at the moment - one VM with one restore point.

However, may I ask you why you're using cloning as replication means? Do you find it more convenient or more reliable to clone VM, delete previous clone, etc., rather than just to replicate it?

Thanks.

@v.Eremin Unfortunately the VM comes in from a separate environment that is not licensed for Veeam so we can't do the replication. Doing the clones, I feel give us these offline recovery points were I dont have to worry about internal VM corruption being replicated if that happens.

[sid6point7]

If you're using reversed incremental or forward + transform, the deleted VMs retention could also do.

@foggy - This seems like it could be a good fit! We are using forward incremental backups with synthetic fulls being created every Saturday.



Please forgive my ignorance, but can you fill me in on the Transforms, what they are and why I should do them?

Thanks for all the help! I've been using the forums only for a couple of days but the help is amazing. You guys rock!

[foggy]

It literally transforms the forward incremental backup chain into reverse incremental one along with creating a synthetic full.

[jmmarton]

I'd go much simpler. Since the source VM which is used for the clone isn't licensed for Veeam, why not use Veeam Backup Free Edition to backup at the source and not mess with the clone at all?

Joe

[tvl]

I had an issue with a host and restored the host from backup using the Entire VM option.

The restore creates a duplicate of the host object, could that be the reason backup of the host resulted in failure?

[PTide]

Hi,

Every time you restore a VM its morefid gets changed so you should remap the backup job that was used to back up the restored VM.

Thanks