SOBR extent (almost) full then backups failing

[rjwolf82]

Hi,

We are using Veeam Backup&Replication with backup to local storage (NAS) and/or Cloud Connect. Some tenants only have Veeam Agent for Windows to cloud backup.
We have configured SOBR in Azure VM with 8 extents, 8 Azure managed disks of 4 TB, so 32 TB in total.

We use two servers with each 4 extents, server 1 has extent F:\, G:\, H:\ and I:\ and server 2 has extent F:\, G:\, H:\ and I:\

SOBR Backup file placement is configured as 'Data locality' and 'Per-VM backup files' is checked.

The problem is that if one extent has not enough disk space all backup jobs that try to backup to that extent will fail, it will not switch to a any of the other extents that have enough disk space.
Is that how it works? Doesn't Veeam check the free disk space and if it's not enough it will switch to another extent?

In my opinion that is a big issue, because we now have multiple backup jobs that are failing because one extent has not enough disk space!

Anyone who can tell me if that is how it works? We are still using Veeam Backup & Replication 9.5 Update 3, but I don't think that Update 4 will change the way SOBR works regarding extents that are (almost) out of disk space.

[ejenner]

Data locality will try to keep the incrementals with their respective full backup files. So if you could try manually moving a folder full of backup files for a particular job onto different disks within the SOBR that will change which disk new data is sent to.

[Dima P.]

Hello rjwolf82,

Kindly take a look at Backup File Placement Help Center article to learn more about data placement in Scale-out backup repository. Cheers!