Delete Restore Points

[wishr]

Hi gentlemen,

Thank you for your comments and feedback. We'll take a look at that.

Cheers!

[SBarrett847]

I keep running into this for Clients and I've yet to see any workable solution within Veeam to get a Job backup and running after a repository is filled. For example, currently I have a simple backup copy job with 1 VBK and 7 incrementals. The Repository is full and cant complete a job. Retention is reduced to 2 points in the job, but this can't be enacted because the backup must complete first.

I have no way to force the new shorter retention policy, nor can I manually delete incremental points to free space from within Veeam. If I manually delete the VIBs I risk corrupting the chain. I have no other Storage to place backups on.

It is a Chicken and Egg scenario. I have seen posts on the forum going back years highlighting the same issue, but no solution yet. If we could manually delete restore points in the GUI this wouldn't be an issue. If we could enforce a shorter retention policy, independent of the current backup state, this wouldn't be an issue.

Am I perhaps missing an obvious solution?

[Vitaliy S.]

Hi Stephen,

The solution would be to use a monitoring and capacity planning tool for backup repositories. This is the only way that would allow you to be in compliance with the configured retention policy and be alerted in advance when repository is close to its capacity limits. I do know this does not help in your particular scenario, but this is how it is usually solved by other customers and VCSPs.

In your case, it seems like you need to delete the latest incremental (not to corrupt the entire chain), lower the retention policy and re-run the backup job. Might do the trick, though a rescan option will be required as a subsequent job run will miss the deleted restore point.

Thanks!

[SBarrett847]

"In your case, it seems like you need to delete the latest incremental (not to corrupt the entire chain), lower the retention policy and re-run the backup job. "

Aha! Ok, that I can do. Much obliged.

[SBarrett847]

Ok Just for future reference (I'll probably find this thread again in search in a few weeks ) The trick is the Delete the last restore point, then exclude all the VMs in the Backup Job, then run the Backup. This way it goes straight to the Merging of the restore points, rather than running another backup first.

[Feature Request] A single command to go straight to merging of restore points and apply the new retention policy, without the need to run a backup first.

[DGrinev]

Hey Stephen,

That's very creative approach to force the retention shorten the backup chain, nice!
Here is a big discussion about deleting restore points, I've merged your thread to keep it in one place.

Thanks!

[SBarrett847]

OK, Further to the advice given to delete the last VIB in the chain - Deleting the last restore point breaks the Job when it tries to create the next restore point - it will not run again without that restore point.

Attempting to remove the missing Restore point in the GUI as per https://helpcenter.veeam.com/docs/backu ... l?ver=95u4
does not work - there is no option to remove the Restore point from the config.

Support have previously said to fix this you this you would .....
-Disable the Job
-Find the VBM file for that Job in the repository and add BAK to the file extension. (so "backupjob".vbkbak)
-Then Rescan the repository.
-Remove the BAK on the file extension
-Then rescan the repository.
-Remap the backup in the backup Job
-Re-enable the Job and run.

But this doesn't work for me.

Looks like a call to support

[SBarrett847]

Case 03462381

[foggy]

Support should be able to edit VBM file to reflect what is currently on disk.

[lobo519]

well - this has gone on for quite a while. I understand you guys say "plan" but we live in the real world and as they say - Stuff happens.

An option to run the job and apply retention at the beginning or a simple way to trim without running the job would be a really great feature at this point!

[mkaec]

Look how easy it could be!

[SBarrett847]

I would love to have that button.....

[lmbutault]

+1

[wishr]

Hello gentlemen,

The retention itself requires some space to be applied (remember merge, transform and other necessary operations), thus that "easy button" (I love the mockup, though ) will be useless in most situations when you lack free space. Do you have any other suggestions?

Thanks

[mkaec]

I think yes and no.

Forward Incremental Backup: Old restore points can just be deleted.
Reverse Incremental Backup: Old restore points can just be deleted.
Forever Forward Incremental Backup: Transform (and disk space) required to delete old restore points.

There's only 1 backup method where the easy button wouldn't help get out of a free space jam, and two where it could be a lifesaver. If it were me, I'd make it available regardless of the backup method because sometimes admins just want to clear out old backups.

[wishr]

In the case with the Forward Incremental backup method, you cannot simply remove the "old" chain at any moment because the new chain might not have necessary number of restore points yet.

[SBarrett847]

Yes you should be able to, if you decide that is what you want to do. A simple popup warning stating that the current chain doesn't have the required Restore points is all that is needed.

[foggy]

Guys, just to remind, there are built-in space notifications and capacity planning reports in Veeam ONE to effectively prevent such situations. Those who ignore warnings seem to be armed with some secret techniques of handling space issues.

[mkaec]

That helps when Veeam ONE is installed. As was mentioned, if someone runs into space problems, they don't really want to be told "you shouldn't have let that happen".

I have not run into space problems with VBR. I have with VAW, and ended up losing the entire backup chain because I deleted the wrong thing. As for server backups, I have used retention management options in other applications even when there was not a repository space issue. Sometimes an admin wants to reduce retention without waiting several hours for a backup to complete first.

[foggy]

Sometimes an admin wants to reduce retention without waiting several hours for a backup to complete first.

But that would mean he already has some sort of space issues, right? Otherwise, he can wait. Or am I missing the use case?

[mkaec]

I guess the most common case *is* to do it in response to free space alerts. Even in that case, admins really don't want to have to wait to see the results. The most likely scenario is changing the retention and waiting until the next day because the admin won't want to kick off a backup during business hours. Then if the gains were not as much as desired, the admin will have to adjust the settings and wait another day. Rinse and repeat.

I know it's a little off-topic, this being the VBR forum, but I recently had an experience like that with VAW. I wanted to adjust the retention on the VAW backups I have on my home systems. Free space was getting a little low. I ended up deleting everything and starting over because I didn't want to go through a multi-day process to get things sorted out.

[foggy]

Thanks for the clarification, Marc.

[lobo519]

Guys, just to remind, there are built-in space notifications and capacity planning reports in Veeam ONE to effectively prevent such situations. Those who ignore warnings seem to be armed with some secret techniques of handling space issues.

Again - I will quote myself "I understand you guys say "plan" but we live in the real world and as they say - Stuff happens."

[TheWaterbug]

So what if I want to delete *specific* restore points.
For example, what if I want to keep the oldest one, but delete some specific incremental points in between?
Veeam has restore points not restore point age (aka 30 points not 30 days of backups) so it isn't always consistently aligned with my retention. Yes it's about storage space but I would prefer to be able to pick and choose "this was an important day, I want to delete backups to make space from the rest of the week but keep this day"etc.
Sometimes there are specific needs. Is there no way to browse restore points and delete a specific one for a specific machine?

Agreed. We just recovered from a ransomware infection on 11/26, so it would be nice to be able to delete all restore points between 11/24 - 11/28, but keep everything else.

As with some other folks here, I would strongly prefer to do it from within the Veeam GUI, so I don't screw it up!!!

[Richard D]

Hi,

We have a situation where a VM was added to a backup job (which already has 8 VMs) and this has filled up our backup repository to the point where the backup has failed. Is there any way I can remove this server from the backup incremental files without performing a delete and new full image?

Total backup storage: 1TB
Total backup size (incl incrementals) before VM was added: approx 400GB

After the VM was added (VM size is 3.2TB) we have ran out of space.

The job has failed but has filled up the backup repository drive with it trying to backup the large VM. Can we remove the partly backed up VM from the backup files to get back our 500GB?

[Natalia Lupacheva]

Hi Richard,

You are going to remove this VM from the job and keep running the job with "old" VMs as it was before, right?
In this case please take a look at this step-by-step instruction and the description what you have to do in general.
If you do not succeed with this instruction please contact Support to fix the problem with the last restore point deletion in your certain case.

Thanks!