Feature request: Manual restore point management gui

[kkuszek]

Hello,
I think Veeam B&R could really benefit from the ability to browse and manipulate the repository in a way that's friendly to veeam, probably another tree view where I can expand a repository, then a job to see the VM's within it's retention and then expand the VM to see each restore point. Then I could say click a restore point and delete it. If it's a full and subsequent incrementals exist then maybe it could warn me about deleting the whole chain. This could easily just tree out under the backup infrastructure tab.

Here is my dillema. I ran extra active full's this month and my repository is large enough to fit x number of days worth of backups. Since tossing in a couple extra active fulls makes a big difference in how much repository you use up I don't have space to accommodate. Now I have to see about manually deleting files and it's a messy way to do it.
I tried changing my retention days but I have no way to re-evaluate repository based on retention settings (maybe this should exist too). In the end I have no clean and clear way to pick and choose what I can afford to lose and ensure backups actually happen.

I could see regulatory benefit here too for other users/industries, maybe allow a specific backup to bypass retention limits & counts from that window. You could have 30 restore points but also keep your eocy full on disk by setting it to manual delete only.

Veeam generally does a good job managing things but you either have to set your number of restore points really low to leave lots of headroom or not do too much out of band. I don't see those as reasonable workarounds.

[Gostev]

I think Veeam B&R could really benefit from the ability to browse and manipulate the repository in a way that's friendly to veeam, probably another tree view where I can expand a repository, then a job to see the VM's within it's retention and then expand the VM to see each restore point. Then I could say click a restore point and delete it. If it's a full and subsequent incrementals exist then maybe it could warn me about deleting the whole chain. This could easily just tree out under the backup infrastructure tab.

Hi Kurt, we have this view today (right-click backup, select Properties). And we have actually added something very similar to the above in v9, but it can only invoked on missing backup files in Backup Properties. This was added to allow users easily clean up unusable chains, and it automatically determines and deletes all dependent backup files starting from selected. I'll log a CR for a future version to make this functionality available on any restore point, including valid ones. Thanks!

[rreed]

Awesome for both. We run into the same here were we're about to consume all or at best nearly all our repository - but if we could just chose that last backup chain and delete, we'd be in good shape. The workaround of setting a job to lower retention and run a backup so it'll delete expired chains/files won't work when you're full/nearly full. B&R getting smart enough to delete bad/unusable chains/files would also be awesome. So yes please add the manually delete a chain/file functionality!

[kkuszek]

Gostev,

I wanted to reply to this now that I have upgraded to V9 and comment.

I do see what you mean where you have the ability to right click and say delete from disk, however it only allows you to delete either all backups for a vm from disk, or all files for the job.

You do not have the ability to selectively delete by restore points.

A better more useful handling would be if right clicking and selecting delete from disk came up with a secondary menu confirmation combination, one which allows you to select based on criteria (all but last successful, all over x restore points, all over x days old or before x date, all successfully archived to tape or other secondary copy, etc).
Alternatively, delete from disk could spawn an expanding subtree with these selections where you then utilize a normal confirmation (unless it requires additional details like number of points, etc).


I never want to clear out an entire job if it's not related to a recent restore, but I do like to cherry pick based on the expected needs of the business.

[dellock6]

But if you delete an incremental restore point in the middle of the chain, all the following points are non usable since the backup is incremental, while your description would work if the backup was differential, where you need for each restore point the full and the single differential restore point.
To make your idea successful, we should merge the deleted restore point into the previous incremental before confirming the deletion itself...