Failed to download disk. End of file failed to upload disk

[kjstech]

Hello, I have a backup that failed with this error:
Error: File does not exist. File: [\\fnbackup1\backups\vnxfs2\vnxfs22015-01-10T220108.vib]. Failed to download disk. End of file Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}.

I fear that it is because this weekend when the backups ran, on Saturdays the backups are huge files, then during the week they are smaller incrementals. So what happens is the backup repository \\fnbackup1 in this case filled out of space. Its a FreeNAS box so its copy on write. In order to get the disk space back I had to ssh to the freenas repository and cat /dev/null > to the largest and oldest full backup file in the folder.

I am thinking because the device ran out of space, the last full backup file failed this weekend and the full backup file that was created ended up becoming corrupted.

Two questions.
1. How do I recover from this error so backups will function the rest of the week until this weekends full?
2. Since its very tough to have consecutive space on any particular device, can full backups be put in one repository and incrementals in another?

[foggy]

Most likely by deleting the VBK file you've corrupted the entire chain, since all subsequent restore points are dependent on it. Manual deletion of restore points outside Veeam B&R console is not recommended. To recover from this, you need to create an active full backup to start the chain from scratch, but consider having enough space for that. No, it is not possible to store fulls and incrementals separately.

To avoid such situations in future, consider lowering retention or switching to forever incremental backup method. On a side note, there are built-in space notifications that allow to track free space on the repository and also Capacity Planning for Backup Repositories report in Veeam ONE that allows to monitor current space consumption and estimate the date when you run out of free space on repository.

[kjstech]

Active full takes a long time right? I hope I can get it done in one evening. I will start it at 5:30 PM instead of 11 PM.

Do I have to delete or move all the backups that currently exist in this \vnx2fs folder? There are two vbk files, one is 1 tb another is 1.8tb, there are 6 vib files between them and 1 vib file after the last vbk file. All in all 11 items at 3.18TB. If I clear that folder out (except for the vnxfs2.vbm file) and start fresh, would that work?

[foggy]

You can move existing backups to reclaim space in the repository. Active full does not depend on previous restore points, so it will work fine.

[kjstech]

Ok thank you. I temporally changed the advanced section of the job to make Tuesdays active full days. On Saturday's we have a synthetic full scheduled.

[foggy]

An easier way to perform active full is just right-clicking the job and selecting corresponding command on the shortcut menu.

[kjstech]

Ok i will likely be gone for the day by that time so I have to schedule it, but its good to know that option is there.

[Bennon Maina]

Alright, had the same issue with Windows Agents backup to an SMB Share.
I had migrated the backup files (windows copy) to a new location but the backup job still failed with this error:
16/03/2018 10:52:10 :: Error: Shared memory connection was closed. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Exception from server: Storage not initialized. Failed to download disk.

It turns out to be the .VBM file which stores information about the VM/Node being backed up. I had to retrieve it from the previous destination (a QNAP share with Network Recycle Bin enabled), copied it to the new location and boom!
Backup with CBT ran without any issues.

Hope this helps your situation .

[shivadiniz]

Hey guys im with the same problem. Bennon Maina which file do you copy and where do you copy?

Thank you for the help

[foggy]

I strongly recommend contacting support for assistance with this kind of errors.