On Fri, 19 Jan 2007, Kern Sibbald wrote:
> Given that this doesn't seem to be creating enormous problems
I had noticed it, but hadn't been looking closely enough to file a
detailed report.
Kern, can you please revisit the definition of "failed" job upgrading?
With large filesets there is a distinct possibility that a full backup may
stil be running when an incremental is scheduled.
If concurrency is enabled, but max concurrency for any single Jobname set
to 1, Bacula has been deciding the backup job in progress had failed and
would upgrade the incremental to full before queuing it. This can and does
result in an endless sucession of full backups - highly undesirable and
can chew up all available tape in a very short period of time.
Ideally:
1: Only do the test at the time the job actually starts running,
not when it's added to the director queue
and
2: Don't define a running job as "failed" for the purposes of testing to
see if an upgrade is required.
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
