[Bacula-users] purging/pruning errored jobs

[John S. <jo...@st...>]

Hi all,

I've been running v2.2.8-? on Debian, using the debian packaged
version and suffering from randon crashed of the bacula-sd daemon,
which emails me a useless GDB traceback.  Oh well, it looks like a
known bug.

But now I'd like to go back and purge/prune all my various backups
which failed due to these crashes and used up a varying amount of
tape on various volumes.  Basically, I've got a bunch of full backups
which all have a status of 'Error' so I figure I can't restore data
from them, so I may as well just blow them away and start from
scratch.

Heck, I'm tempted to just recycle *all* my volumes and start over from
scratch, because it's so hard to know what data is good and which
isn't.

Sigh... so now I've got all kinds of crap mixed in.  Should I just

a) use the sledgehammer and just recycle the entire tape?

b) can I prune/purge by jobid and just get rid of those savesets, and
   then migrate good data from mostly empty tapes to other tapes?

c) how do I generate reports showing this easily?  This is where
   bacula relly falls down, is in useful and conscise reports.  I
   really dislike how the raw results of mysql queries are returned
   from various commands.  Like:

	list jobs
	list volumes
	status storage
	update slots

   All of these commands could be so more more useful if they were
   formatted better.  


Thanks,
John