Backup Exec 15-Hyper-V and FP3 Backup Checkpoints

[Rick_C]

As most of us have now realized...FP3 changed the way backups of Hyper-V VMs are down with the new "fast processing" mode.

If you have not read the following thread and would like to get some backgrond:

https://www.veritas.com/support/en_US/article.000100786

All well in good.  My backups *do* seem to be faster and do not appear to have any issues...except for one thing:

When it comes time that you need to expand a disk or do any type of "edit", that persistent checkpoint gets in the way!

I had to do this just today (expand).  I had to use the powershell commands (well documented in the referenced thread) to get rid of the checkpoint so that I could do the disk edit.

Am I doing something wrong?  The checkpoint reall are supposed to stay "forever"? Surely one is not supposed to edit the checkpoint file to expand... ?

Anybody else running into this?

[ABS_IT]

BE creates a new recovery snapshot when a backup starts and the faster processing option is enabled. If you observe the VM in Hyper-V Manager during an incremental backup, you will see two snapshots. When the backup is done, BE deletes the previous recovery snapshot (manual snapshots are not affected, as far as I can tell). So the snapshot does not "stay forever": it is a new snapshot every time. This fact is easily overlooked because the snapsnot names look similar.

Regards,

-Roger

PS. Personnally I love the new incremental backup method. It has made my incremental VM backups a lot faster at the expense of an additional disk space requirement. My only gripe is that the setting is BE server wide. I'd love to see it configurable on a per object basis.

[Colin_Weaver]

@Rick.C not being able to edit(resize) a Hyper-V disk if a checkpoint is present is a Microsoft limitation

The way the faster backup method works does use checkpoints that are maintained (and something other backup products using the same process would also do.)

Hence your experience is an expected consequence.

The fact the no one else has commented is probably because very few users have tried to resize disk sizes after running backups with the faster method.

@ABS IT - you should still see a checkpoint (after the first faster backup) when backups are no longer running, yes the next time a new backup runs then that checkpont is replaced with a new one so agree it is not a permanent "stay forever" checkpoint, however the end experience is that there will always be one checkpoint (possibly 2 during a backup)

[Mattythefatty]

This is a pain.  When this feature first came out, you would use powershell and specify which VM's you wanted to have participate in the faster processing method.   Now it's a global setting and worthless for the small VM's that can backup each night in full without issue.  

My method for extending the disk is to delete the checkpoint, reboot the VM so the .avhdx files will merge, extend the disk(s), and then run another full backup.   I don't believe I can just checkpoint the server myself to be back in the same place I started. 

Also, if you use Hyper-V replica like I do, you need to expand the virtual disks on replica exactly the same size or your replication will fail as well.

Matt