Veeam B&R 11a deleting backups to external rotated drive 'Removing BACKUP per retention policy'

[AlexLeadingEdge]

Case #05371282

Hi guys,

We were wondered why our backups on our external daily rotated drives were all being deleted after upgrading to Veeam B&R 11a. In the past we have B&R set to 14 restore points, but now it is deleting them down to 2-4 backups on our five rotated backup drives.

The 14 restore points was calculated so that there would be the maximum number of backup points as possible on a drive while leaving enough overhead for the next backup and the backup merge to be able to complete.

On the drives that haven't had a new backup on since the upgrade it clearly shows that there are 14 backups on the drive.

Is this new to Veeam Backup & Replication 11a?

Am I now to go 5 drives x 14 restore points = 70 restore points?

We're a little annoyed because we have effectively lost 14 weeks / 3.5 months worth of air-gapped backup history.

When you specify retention settings for a backup job or a backup copy job targeted at a backup repository with rotated drives, you must define the total number of restore points that you want to retain on all drives in the set. For example, if you set retention to 14, the job will keep the total of 14 restore points across all drives.

https://helpcenter.veeam.com/docs/backu ... ml?ver=110

[HannesK]

Hello,
the behavior in 11a is as described in the user guide. Looks like a bug in the version you used before.

Best regards,
Hannes

[AlexLeadingEdge]

Hi HannesK,

We have been using Veeam B&R since version 8 and gone up every stable version that have come out. We have B&R installed at four different companies and they all have had the same setup with the number of restore points set for the amount that can fit on the rotated drives, not the total number we want.

I think this is a policy change in 11a, otherwise we wouldn't have rotated backup drives with 14 restore points from when it was running 10a.

We also wouldn't be seeing groups of this in the logs:

2/04/2022 2:30:29 a.m. :: Removing 'R:\Backups\Copy All Physicals to R (REMOVABLE 10TB)\Copy All Physicals to R (REMOVABLE 10TB)D2022-03-28T023000_CA0E.vib' per retention policy
2/04/2022 2:30:29 a.m. :: Removing 'R:\Backups\Copy All Physicals to R (REMOVABLE 10TB)\Copy All Physicals to R (REMOVABLE 10TB)D2022-03-27T023000_6737.vib' per retention policy
2/04/2022 2:30:29 a.m. :: Removing 'R:\Backups\Copy All Physicals to R (REMOVABLE 10TB)\Copy All Physicals to R (REMOVABLE 10TB)D2022-03-26T023000_7763.vib' per retention policy
2/04/2022 2:30:30 a.m. :: Removing 'R:\Backups\Copy All Physicals to R (REMOVABLE 10TB)\Copy All Physicals to R (REMOVABLE 10TB)D2022-03-25T023000_9EAC.vbk' per retention policy

If I look at one company with version 10.0.1.4854 (B&R 10a) I see the retention is set to 4 restore points and there are 4 on the rotated drive.

Question: If I set the number of restore points to 2, and have five drives, how many restore points should I expect to see on each drive?

[Gostev]

This question does not make much sense because backup job retention is not "per drive". By setting the number of restore points to 2, you're telling the job to have its retention policy delete all restore points beyond the most recent two it has created, regardless of where those restore points are physically located. As Hannes said, if it worked any different in some previous versions, it was not intentional but rather a bug.

We're a little annoyed because we have effectively lost 14 weeks / 3.5 months worth of air-gapped backup history.

I'm sorry, but how could you possible lose air-gapped backups? Veeam cannot magically delete backups residing on disconnected external hard drives. So I really don't follow you here.

[AlexLeadingEdge]

I'm sorry, but how could you possible lose air-gapped backups? Veeam cannot magically delete backups residing on disconnected external hard drives. So I really don't follow you here.

We have rotated through all the air-gapped rotated backup drives before we noticed that the backup copies were disappearing. There were 14 backup copies on each drive, now there are 1 to 3 backup copies.

I have increased the number of restore points from 14 to 70, hopefully that will fix this issue.

On further analysis I note that the other three clients that have had no issues the number of backup points is set to a number lower than the number of drives, so we wouldn't have noticed an issue anyway.

I'm reading this about 10A, and it agrees with what HannesK said:

When you specify retention settings for a backup job targeted at a backup repository with rotated drives, you must define the total number of restore points that you want to retain on all drives in the set. For example, if you set retention to 14, the job will keep the total of 14 restore points across all drives.

https://helpcenter.veeam.com/archive/ba ... s_hiw.html

[Mildur]

You should configure retention type to days (98 days in your case) instead of a restore point count.
Then you will have restore points for the last 14 weeks, as your business requirement is.

[AlexLeadingEdge]

I may be wrong, but if I go for "days" then if there are two backups on one day it gets counted as 1 backup, which could end up with too many backups on a rotated drive. As mentioned above, we have calculated how many backups can fit on a rotated drive so 'restore points' makes it easy for us to work with. I do get your logic, however.

I may also be wrong, but Veeam B&R doesn't seem to monitor how much free space is available and just keeps writing until the drive is 100% full, at which stage the only way to get out of this situation is to delete backups and drop the number of backup points.

[foggy]

You're correct with both assumptions.

[AlexLeadingEdge]

I had set the backup copies to 70 points, but on the Friday backup drive, which gets left in all weekend, the drive has 34 backup points and has almost run out of free disk space. I have changed it to 60 days and will continue to monitor over the next couple of weeks, expanding that number as possible.

[AlexLeadingEdge]

Hi guys,

Now we have the opposite issue: I've set it to 60 days, but on one of the five drives we now have 29 backups and the drive is about to run out of free space. This is the Friday drive, which stays in all weekend. The oldest backup and the newest backup are 57 days apart, so in that regard it seems to be working.

What I really want is 10 to 12 backups on each drive, I don't want a total number of days or restore points across all drives, but I don't seem to have this option.

I've reduced the number of backup 'days' down to 50.