How to safely stop the backup evacuation process?

[rlaovai]

Greetings.

We have a scale-out repository backed by 2 extents which are iSCSI volumes presented to Veeam. We started backups evacuation from one extent to another but we found evacuation estimate is not fit into our maintenance window. Is there any way to safely cancel evacuation process? Backup jobs failing with "Unable to allocate processing resources. Error: Some extents storing previous backup files are in maintenance mode"

Regards.
Alex.

[Shestakov]

Hello Alex and welcome to the community!
It`s totally safe to abort the evacuation. Already evacuated files will be kept on the new extents and all jobs will work fine.
Thanks!

[rlaovai]

Nikita, thank you for the help, but can you please explain how to abort evacuation. I see no such option in interface.

Regards,
Alex.

[Shestakov]

You stop it via the
History>System>Backups evacuation job > Stop

[williampu]

Is this safe to do on ReFS volumes?
My current SOBR is at 1.2PB with ~470TB consumed, however the written data is sized at ~6PB in size.
Will the consumption of each extent on the receiving end reduce to post ReFS data reduction sizes if the process is stopped?
I'm asking as I started the migration process late last week and its done 25% of one extent and the other extents are basically full now.

[mkretzer]

REFS block clone does not work for evacuations! Currently there seems to be no way to "transfer" the block clone benefits.

[Mike Resseler]

Hi William,
Mkretzer is correct. REFS block cloning cannot be simply copied from one volume to another. The block cloning mechanism is something that works per volume and if you move data to another volume, then you will see it is "rehydrated" (which is not really the correct term but I hope you understand what I mean). When data is on the new volume, only after a new synthetic full, the blocks can get aligned on that specific volumen and then you can start benefitting again.