Copy job to rotated drives: file does not exist

[maxxer]

Hi.

We have a running backup job to an iSCSI target, we'd like to have a second backup storage on rotated drives.

Instead of creating a new backup job I've configured a Backup copy job, so I can do it during work hours without causing load to the datacenter.
I added a new NFS repository checking the Rotated drive option. The first run went fine, but when we swapped the first disk the copy failed with:

Code: Select all

Processing FS Error: No such file or directory. NFS status code: 2 File FSD2021-07-08T135340_C50A.vbk doesn't exist Failed to open storage for read access. Storage: [10.10.10.18:/share/MyArchive1/Copia su Asustor/FS/FSD2021-07-08T135340_C50A.vbk]. Agent failed to process method {Stg.OpenRead}.
Error: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond

I thought this was because I configured the job to retain 2 restore points. After better reading the docs I learned that value should be the amount of restore points across all drives, having 10 disks to rotate daily I've set it to 10 restore points. But I'm still getting that same error.

As RPO Monitor I've set 16 days, I hope I used the value correctly.

Is this the correct approach? Or is it better to make a File copy job of the whole repository on the rotated disks?

Thanks

[PetrM]

Hi Lorenzo,

The approach seems to be correct but the behavior is not expected: if an external drive has been swapped and there is no full point, a new backup chain will be started automatically. Please contact our support team and provide us with a support case number as requested when you create a new topic about technical issue.

Thanks!

[foggy]

I agree, using a backup copy is the correct approach and the issue doesn't look related to the job settings.

[maxxer]

Opened case 04912998, thanks

[filippot]

Same problem here
Case ID: 04920569

[filippot]

Thanks to support the solution for us was to add this key
ForceCreateMissingVBK
1. Type: REG_DWORD (0 False, 1 True)
2. Default value: 0 (disabled)
under HKEY_LOCAL_MACHINE\SOFTWARE\Veeam\Veeam Backup and Replication\

https://www.veeam.com/kb1154

[maxxer]

The ForceCreateMissingVBK key value must be set to 0?

I'm having inconstant behavior, it seems the scheduled run always throws the same error, while if I run the job manually it works.

thanks

[maxxer]

We keep having problems with rotated drives. We found out if the drive is empty the backup copy goes fine, if the destination already contains backup data it still fails with

Code: Select all

7/30/2021 2:04:26 PM :: Processing ADC01 Error: No such file or directory.
NFS status code: 2
File DC01D2021-07-29T133136_ED43.vbk doesn't exist
Failed to open storage for read access. Storage: [10.2.1.1:/share/MyArchive1/Test/DC01/DC01D2021-07-29T133136_ED43.vbk].
Agent failed to process method {Stg.OpenRead}.

We double checked and the registry setting is set as suggested.

[veremin]

Might be worth reaching our support team for further investigation. Thanks!

[Gostev]

The only thing I noticed is that you're using NFS, which is a very rare type of rotated drive. So there's a chance there is some issue specific to this configuration. Since you already have a support case open, our engineers will get to the bottom of this.

[maxxer]

Thanks for the feedback. Unfortunately we're using a community installation so support is at best effort.
In the meantime we'll give a try using SMB repo. Thanks again

[foggy]

The ForceCreateMissingVBK key value must be set to 0?

Should be set to 1 to allow Veeam B&R to create new full after the disk is swapped.