nas backup problem

[ousturali]

Hi to all
we use the veeam10 ,with NAS backup and vmware backups.
for the last 2 months there were no problems ,but recently we got this error.

"Failed to perform applying backup retention policy for restore point 8/5/2020 9:00:32 PM Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 0005
Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 0005"

and backup creates errors, any ideas??

There are different nas backups on the same veeam server, only one of them creates this error.
(nas backups are from netapp filer )

[Natalia Lupacheva]

Hi,

it does not seem to be a common issue, please open the Support ticket and post its ID here (it's mandatory when you post a technical issue according to the forum rules).

Thanks!

[ousturali]

Ok then ,

i have created the case , the number is as Case #04402771

[roel.loots]

I have the exact same issue case #02063736

30/09/2020 0:53:48 :: Failed to perform applying backup retention policy for restore point 6/09/2020 21:29:58 Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 0004

In our case we have a nutanix files fileserver that we are trying to connect to.

[Natalia Lupacheva]

Hi Roel,

I've checked Orcun's case and I see the problem can be solved with update 10a installation.

As a workaround, you can clone the job and start a new backup set.

Thanks!

[roel.loots]

I will install the update this afternoon and will report back tomorrow.

[roel.loots]

Not all backup jobs finished in time, update will be postponed. I'll report back after upgrade.

[Natalia Lupacheva]

Hi Roel,

Thanks a lot for the update, will wait for the results.
Also, the case number is 04415361, right?

Thanks!

[roel.loots]

Hi Natalia, that is correct. That is my case number.

I wasn't able to update Veeam to version 10a on the scheduled time, but I was able to do this yesterday evening.
Unfortunately installing v10a did not solve the issue for me.

9/10/2020 1:45:47 :: Failed to perform applying backup retention policy for restore point 6/09/2020 21:29:58 Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 0004
9/10/2020 1:46:40 :: Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 0004

I'm waiting for further instructions from Veeam support team.

[mtellier]

I have the exact same issue case #04358073

This ticket was created on August 27 (>5 weeks ago) and I am not getting any traction.
I applied hotfix 10a then another fix 247640 for 10a to my VBR server, all proxies and rebooted but didn't resolve the issue.

10/11/2020 9:08:07 PM :: Failed to perform applying backup retention policy for restore point 7/26/2020 9:00:19 PM Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Failed to find item 000e

This failing job is for a NetApp SMB Share.

I also have a separate VBR deployment, NFS share on NetApp also failing with a different error, details in the same ticket:
10/4/2020 1:27:14 PM :: Failed to perform applying backup retention policy for restore point 8/29/2020 7:50:48 AM Error: Agent: Failed to process method {NasMaster.RetainShortTerm}: Invalid insert result [false]

I was informed that the hotfix 247640 will resolve this problem, I just didn't let the backup run long enough. The first run looked like it was stuck at 39% after 36 hours, next time 51% for a longer period - I stopped both jobs. I decided to run a health check on ~ 5TB it's still running after 95 hours and only at 72% complete.

Both backup jobs were running fine, then just stopped working, at different times. We have many more NAS shares to backup, but unable to move forward because of these unresolved issues.

[roel.loots]

So I got back in touch with Veeam Support. The feedback I got was the following:

- Veeam is investigating if this is a bug.
Since our issue occurred on the 6th of September, I'm pretty sure it's a bug, because we did not touch the job in over 2 months
- There are too many restore points for the share, these cannot be deleted manually (unless you delete all of them)
- Running a health check might fix the issue, but is very likely to also fail since there are way too many restore points that need to be comitted (85 restore points and we only keep 21)
- Best practice (that is not publicly documented): 1 job per share (and do not chain backup jobs)
- Easy fix is to just clone the job: this triggers a new dataset (active full) so you need to clean up the old data.

I'm running the Health check right now and I'm probably going to reconfigure our job to match the 1 job per share principle.
Although this might be a best practice, I'm convinced that this is not a fix for the actual problem.

[Dima P.]

Hello folks,

Thank you for sharing the updates. We will review both cases with RnD folks. Cheers!

[mtellier]

Here is the solution I received from Veeam Support.

The metadata for the backup chain had become corrupt, the new cumulative patch should prevent it from happening again but does not correct the existing metadata. Solution is to create a new backup chain as follows:

Hold down <Ctrl>, right click on NAS backup job, select 'Start new backup'
This will move existing backups to 'Disk imported' then begin a new backup chain.

[Natalia Lupacheva]

Hi Mark!

Thanks a lot for sharing, appreciate it.

[roel.loots]

The health check did not fix the metadata, so I ended up creating new/separate jobs as adviced.
These jobs are running fine.

The feedback from mtellier also is a good tip if you want to keep your existing jobs.

[Natalia Lupacheva]

Hi Roel,

Thank you for coming back with the results!
Glad you've finally got a working solution.

[storageguy]

Here is the solution I received from Veeam Support.

The metadata for the backup chain had become corrupt, the new cumulative patch should prevent it from happening again but does not correct the existing metadata. Solution is to create a new backup chain as follows:

Hold down <Ctrl>, right click on NAS backup job, select 'Start new backup'
This will move existing backups to 'Disk imported' then begin a new backup chain.

If I do this will it preserve my existing restore points? We keep one version on-prem and the rest of the older versions will be sent to object storage for archiving for 3 years. So if I do this method will it keep all of our existing older versions in the object storage?

I have a similar issue by the way.

[Dima P.]

Create new backup chain option starts new backup and detaches your existing backup chain from the backup job under the 'Disk - Orphaned' node (in v10 that was Disk - Imported). Retention for such backups wont be performed and you have to delete those manually. Hope that helps!