B&R 9 - creating new folders for jobs with updated nameset?

[kkuszek]

I ran my first B&R9 job last night and when I go into my repository noticed it created a new folder for a new backup chain and ran a full vs building an incremental off my existing chain.
Previously in v8 I had renamed a job but it retained the folder name it previously held.

Unfortunately for me that added an unscheduled full.

Anyone else see this?

[veremin]

Can you elaborate on that, please? You'd renamed the job and that led to the new full backup? Thanks.

[kkuszek]

Hi V.Eremin,

in vbr 8 I created a job called: "2k12R2 machines"
I later renamed it to: "Backup - 2k12R2 machines"
The backup chain was initially created in E:\Backups\2k12R2 machines\ and remained in that same location even after renaming the job, although the file names within had been changed to reflect the new job name..
I have also in the past manually deleted files, then re-imported them with a datastore rescan. These files for this job are listed under disk (imported).

When I upgraded to v9 and ran the job for the first time the location "E:\Backups\Backup - 2k12R2 machines" was created and the scheduled incremental ran as a full having been created in an empty folder. My original mismatch folder remains alongside and untouched.

If I look at the job I see only a few available restore points.

[dellock6]

The job itself may have few restore point, but since each VM still has its own morefID, the said VMs have more restore points, by combining the ones in the old backup and the ones in the new backup. At the end you do care about restore points of VMs, not the jobs.

[veremin]

When I upgraded to v9 and ran the job for the first time the location "E:\Backups\Backup - 2k12R2 machines" was created and the scheduled incremental ran as a full having been created in an empty folder. My original mismatch folder remains alongside and untouched.

If prior to that you've manually deleted backup files, the described behaviour is expected - instead of running incremental cycle the backup job performs a new full backup. Thanks.

[kkuszek]

Hi V.Eremin,

Just to clarify:
files deleted manually are ones older than the most recent full backup so it should not necessitate a new chain and took place awhile ago on v8. The last backup prior to v9 was incremental.

I think it might be worth an advisory on the upgrade notes if this is expected behavior that any previous jobs that have ever been renamed will require storage for an additional out of scope full after upgrading to v9 if using forward incremental mode with regular active full (not sure if any other modes would do the same).

[veremin]

Thanks for clarification. The behaviour described by you does not look expected, so, kindly, open a ticket with our support team and let them confirm that. Thanks.