A Question about Housekeeping...

[Bob_Manning]

Hi All,

This matter is the subject of a support case (#05104583).

Environment is 2 * Veeam B&R V10 (10.0.0.4461.P1), writing to HPE StoreOnce 5200 - 130TB configured.

About 10 days ago, I had what I thought was the bright idea of switching on encryption in a selection of my backups. I noted the fact that a new Active Full would be taken and proceeded on the basis that I had about 25TB free on one StoreOncearray and 40TB on the other.

On Friday last I got an alert from HPE to say that one array was at 95% capacity, and when I checked it out it had indeed dropped from 25TB to 4.88. You can imagine the surprise

I immediately suspended backups and went looking for ways to reclaim space, and a colleague and I spent time on Saturday stepping down restore points from 28 to 21 and restarting the non-encrypted jobs.

This had the effect, over the following 36 hours, of getting us back about 5TB, but obviously we're still running on fumes.

I'd like to delete any old objects on disk that have exceeded their retention policy of 30 days and been written to tape - i realise I can do this in the "Backups\Disk" context by selecting an object and clicking 'Delete from Disk'.

However, some questions I have raised with Support have to do with objects in the "Disk (Imported)" context; specifically, what are these objects and is it okay to delete them? I note that most refer to backups that have exceeded disk retention periods, and while I've been able to interrogate some for Guest Files recovery, others throw an error that suggests the relevant backup object no longer exists on the backup array.

These objects have always been there, and I guess I haven't given them much thought up to now, but it's time to do a bit of housecleaning and I just want to be certain I'm not going to get rid of something I might later wish I hadn't.

I'm really just looking for some guidance on the best way to free up space - I've seen a number of other threads but none seem to mirror my situation.

Anyway, if you've made it to the end of the thread (a bit long-winded, sorry), I'd appreciate any advice anyone might have.

Thanks,
Bob

[Gostev]

Hi Bob

Indeed encryption kills external deduplication... the only way to combine deduplication and encryption is to have Veeam manage everything and use ReFS/XFS repositories. We're aware which blocks are matching even if they are encrypted, so we can still reuse them normally in newly created restore points.

Backups under Disk (Imported) are normally safe to delete. These can only represent temporarily imported backups, or backups not belonging to any active job. If you had V11 installed, you would have separate entries for these two backup types thanks to the new Disk (Orphaned) node.

Thanks!

[Bob_Manning]

Hi Gostev,

Great stuff! Thanks for the swift reply

As it happens, I saw a video on YouTube that showed the Disk (Orphaned) area, but I thought it was one of those items that was only created in the GUI as the need arose, etc. Definitely worth looking into!

I'll get on with the housework immediately and look into the encryption issue separately.

Thanks again for your help - have a good one!

Cheers,
Bob