VBR Not respecting retention policy

[josh_swr]

I have a backup job which was originally set with a 14 restore points retention policy and a schedule to back up once per day. A few weeks ago I changed this job to backup every 3 hours between 6 AM and 7 PM, with a 14-day retention policy.

The issue is that it doesn't seem to be respecting this retention policy. I've currently got incremental backup files going all the way back to the 6th of March, which is well beyond the 14-day retention. Can anyone advise why this would be happening? It's causing my backups to use more disk space than anticipated. This is a Hyper-V backup job if that makes any difference.

[Mildur]

Hi Josh

Can you tell us a little bit more about how the jobs are configured? Does your job creates regular full backups (synthentic or active fulls)?

With monthly active full backups, you will see restore point on your backup repo with a retention policy of 14 days for the last 0.5 - 1.5 months.
This is how a Forward Incremental backup chain works. The entire old backup chain can only be deleted after all restore points from this chain are outdated.
This behavior is explained here: Forward Incremental Backup Retention Policy

If you don't have monthly fullbackups configured, then please open a support case as required for any technical issue. The retention period should be respected. HyperV Jobs are not different in this case from other jobs like VmWare or Managed Agent Jobs.

[josh_swr]

My job is configured to do a full active backup on the last Sunday of the month, so I suppose that explains it.

Looking at the backup files I can see that I have full backup files from 28/03/22 and 17/03/22 with 8 days worth of incremental backups earlier than that. Does that sound right?

Also the last set of incremental backup files jump from 06/03/22 to 27/02/22, and the ones from 27/02/22 don't appear in the repository properties menu. Are these ones safe to just delete?

[Mildur]

My job is configured to do a full active backup on the last Sunday of the month, so I suppose that explains it.

Yes, if you just do monthly fulls and no weeklies, that explains it why you have more restore points on disk as the configured 14 days. If you only want to have 14 days of restore points, you have two options:
1) configure weekly synthetic fulls. This will give you max 3 weeks of backups.
2) disable the regular fulls and do forever forward incremental backups. This is the only way to have exactly 14 days of restore points on your backup repository.

I don't recommend option 2, I would use option 1. With reFS/xfs Fast Clone enabled backup repositories, synthetic full backups are space less and much faster than active fulls. But this depends on the typ of backup repository you are using.
If you choose to implement one of this two options I provided, veeam will automatically remove all outdated restore points after 2-3 weeks.

Looking at the backup files I can see that I have full backup files from 28/03/22 and 17/03/22 with 8 days worth of incremental backups earlier than that. Does that sound right?
Also the last set of incremental backup files jump from 06/03/22 to 27/02/22, and the ones from 27/02/22 don't appear in the repository properties menu. Are these ones safe to just delete?

I don't see them (fullbackups) on your provided screenshot. That's strange. With just monthly full backups you shouldn't have the weekly full from the 17/03/22. The 28/03/22 makes sense, but it should be removed already. I would kindly ask you to open a support case for this two questions. I don't have access to your environment and your job configuration. So I cannot say what files can be safely deleted and which are not.