Veeam Scale Out Repositories case id: 02811297

[ICT05]

Hi Veeam

We have this multi tenant environment with more than 80 clients.
I've got a primary job and a copy job. My copy job is where I use this Scale-Out repository this is made up of 8 Extent's each 10TB.

I can see that when the job run's it does not actually use a "Spill and fill" structure but rather a fill and spill architecture, this mean that when my drives fill up the next job will not merge as the space is already full.
Now for small backup's or small environments that's ok as you can just expand disks. But I'm already at disks with 10TB of data so for just put this into maintenance mode and evacuate this is a time consuming process.

If there was a proper spill and fill in place as the likes of "Commvault" does that would take my manual maintenance away.

Regards
ICT05

[foggy]

First of all, for more even distribution of backups between extents, it is recommended to use per-VM backup chains.

this mean that when my drives fill up the next job will not merge as the space is already full.

In theory, Veeam B&R always tries to avoid such situations and reserves the amount of space required for merge operations. But sometimes the heuristic space estimation mechanism might give wrong estimation - to minimize risk of such situation, you can increase space reservation with the help of SobrReserveExtentSpacePercent registry value (default is 1% of disk capacity).