Cleaning up backup repository

[ashrid]

Hello,

Unfortunately the location our Veeam uses for its backup repository ran out of disk space and the previous full backup and incremental files were deleted from the disk.

Unfortunately Veeam still lists the previous backup jobs as available restore points (which don't obviously work now) and I would like to 'reset' this show it only shows the backups currently on the drive.

I have tried using 'Remove from backup' and 'Delete from disk' on both the indivdual VM's and the whole datacentre but this doesn't work (and I dont want to delete the new backups from the disk anyway)

Can anyone tell me how I can clear down the restore list?

Thanks

[Gostev]

The only way do this without editing the database is to let job retention clean up the records belonging to the deleted restore points.

[ashrid]

Thanks for the reply.

The retention period for the backup job has always been set to 7 days but for some VM's I've got ~70 restore points (the job had previously been set to forever incremental which has now been changed).

[foggy]

So that's another problem of the retention settings not getting applied. It's hard to say what's actually happening without knowing your actual job configuration, the moment of switching to reverse incremental mode, and the moment of restore points deletion from disk. There's a little chance that after all this you have come up with corrupt backup chain due to some reason. If this is the case, let our support team have a look at your setup for further investigation.

[SunkistDavid]

I think same issue; using the primary B&R server has files dating back over 3 months. C:\backup\replicas - no cleanup seems to be performed

[Gostev]

David, these sounds like your replica metadata files, and not actual backups.

[SunkistDavid]

OK - when do those get cleaned-up as its showing 3 months worth and its filling-up my HD OS partition.

[Gostev]

Filling up the partition? They should be really small. They are required for the replication, and are created in the backup repository you have specified when creating the replication job. Technically, you can delete all of them at any time, and they will be automatically re-created at the next replication cycle (which will make the replication cycle to run longer than usual). If you believe some of them belong to no-longer-existing replication jobs, than we would want to investigate deeper why they did not get deleted along with the replication job.

[SunkistDavid]

I had 4GB in indicated folder dating back to January 2012. I deleted files up to March 1st freeing-up 2GB. Looks like each individual folder in the directory is between 16 to 40MB in size.
Sorry to co-opt this thread.

[ashrid]

I'm not too concerned about retaining the backup chain on disk as I backup the Veeam files to tape as well.

My next full backup is tomorrow, in which case is there an easy way to remove the entire Veeam backup history and start fresh with tomorrows full backup?

[foggy]

My next full backup is tomorrow, in which case is there an easy way to remove the entire Veeam backup history and start fresh with tomorrows full backup?

You may want to try to set retention to keep only 1 point. After creating the new full it should delete all the previous backups leaving this full alone. If some files remain on disk after that, you can clean them up manually. (Don't forget to set the retention settings to the required value after that.)

Though I'm not sure whether this helps in case of corrupt chain. If not, there's always a way to ask support for help in cleaning up the database.