Community Edit Backup Windows Server to NAS

[curt132]

Hello everybody,
i need some help on my hopefully easy to solve problem.

Im using a veeam community edition 12.2.0.334 on a Windows 2019 Server which is backing up local folders (size 2,1TB, very few but big files which are vhdx hyperv exports of vms)
to a NAS Qnap (free capacity 7,1TB). The Files i need to backup change completely each day (daylie vhdx export) so i dont know how much space it would need with incrementall or if the incremental is even working with vhds files but even with a full size backup with no compression this should work for 2-3 days i would think.

The Backup is working ok but i have the strange problem that the job is stopping after some days with error:
Error: Agent: Failed to process method {NasMaster.SaveJobBackupMeta}: not enough capycity, Failed to write data to the file [\\nas-xxx\backup\Backup-1.NAS\Backup-1.NAS.vstorecopy].

In the job settings under backup repository i even tried to set "Keep all versions last days" to just 1 day. But even 1 generation is filling the nas up after some days.
It seems to me that veeam is not deleting the old generation but i dont know why.

Maybe someone can help, Thnx in advance
greetings curt

[david.domask]

Hi Curt,

Before discussing the Unstructured Backup job error, can I ask, is there a reason to backup exported VHDX files as opposed to doing an image level backup with Veeam of the VMs?

Veeam integrates with Hyper-V and can do Application Consistent backups of your VMs, so I would advise using this method instead of the Unstructured Data backup -- the recovery from image level backups is much more straightforward and offers far more granular restore options.

As for the issue itself, regrettably we will not be able to troubleshoot this over the forums, so please open a Support Case as is noted when creating a topic. Free/Community Edition cases are eligible for support on a "best-effort" basis. Be sure to include logs for Support to review. (Use the 1st radio button and select the affected job itself)

If I had a guess, you're facing this pretty fast because Unstructured Backup jobs will backup new versions of files (files changed since last backup); the entire file is backed up when a new version is available, so very likely multiple versions of the same VHDX are getting backed up. These are _not_ incremental at the file level, if the file has changed, a new version of the file is backed up in its entirety.

Image level backups would help with this also, as we use Hyper-V native Resilient Change Tracking (equivalent of Changed Block Tracking), so your incremental backups will only include changed data since the last backup run.

[curt132]

Hi David,
thanx for you reply.
We use on some (of the most important) servers additional exports because we had in past cases with database backups that were created with application consistency, that should work but didnt work.
With an (offline) export there wont be an inconsistency. So its an additional fallback backup.

"If I had a guess, you're facing this pretty fast because Unstructured Backup jobs will backup new versions of files (files changed since last backup); the entire file is backed up when a new version is available, so very likely multiple versions of the same VHDX are getting backed up. These are _not_ incremental at the file level, if the file has changed, a new version of the file is backed up in its entirety."

I agree that because its vhdx every day there will be a need for full backup because the vhdx change (only 1time) daylie but if i have 2,1 tb data to backup and have 7tb space this should be ok
for 2-3 generations and even 1 gen is not working. What i dont understand is why there is trouble with deleting old versions. On other servers same kind of system this works ok.

i will try with the ticket. Thank you and have a nice weekend