Deleted reverse incremental (manually).

[lohelle]

Hello
Some reverse incrementals where deleted when urgently needing some HDD space
We decreased the number of restore points for the job AFTER this, but the transform job crashes when not finding the old backup files.

TextFromTar failed Client error: File does not exist. File: [G:\high\high_vb_long\high_vb_long2014-01-07T001713.vrb]. Failed to restore file from local backup. VFS link: [summary.xml]. Target file: [MemFs://Tar2Text]. CHMOD mask: [0].
Error: Client error: File does not exist. File: [G:\high\high_vb_long\high_vb_long2014-01-07T001713.vrb]. Failed to restore file from local backup. VFS link: [summary.xml]. Target file: [MemFs://Tar2Text]. CHMOD mask: [0].

Is it possible to force the job to not care about the missing VRB's and just apply retention policy. All needed files (for the new retension policy) should be there. Only the oldest files (now outside to the retention period) was deleted.

[Vitaliy S.]

Hello Lars,

Do you run forward incremental backup job with synthetic full and transform option? I don't think there is a way to force job not to care about restore points that are still present in metadata and configuration database. In order to fix your job, you should run active job pass or continue running incremental job passes and waiting for another transform operation (not sure the second variant would work due to missing files).

P.S. Deleting files manually is highly not recommended. These kind of actions lead to job failures like you've posted above.

Thank you!

[tsightler]

If you were just using reverse incremental, then you'll get the error above when it tries to implement the retention, but it shouldn't be catastrophic in any way. Once your retention moves past the files you've manually deleted things should return to normal.

If you absolutely want to fix this there is a way, but it's somewhat involved and probably not worth the risk if the backup job is still running and it's only the retention portion that is giving an errorl.

[lohelle]

I think I will start a new backup job next weekend. Will take a couple of days (remote location, 100mbit link)

The backup job is running fine, but the transform fails also (due to this error). So I get a long chain of "normal" incrementals.
I'm pretty sure that deleting old reverse incrementals worked pre-v7. I did that a few times when migrating raids etc

[foggy]

Does this "long chain of normal incrementals" already contain the required number of restore points specified in the job settings?