On Wed, 28 May 2008, Tore Anderson wrote:

> > We make our users jump through hoops for restores - mainly because we
> > don't want them to become dependent on the things. Grabbing a file
> > off backup tape should be last resort. As part of this hoop jumping
> > we enforce a minimum waiting period for restore requests (a few
> > hours) which is usually long enough to allow for full execution
> > queues or higher priority jobs.
>
> Unfortunately I'm not in a position to enforce this, since I have to
> deal with my colleagues instead of (l)users...

Don't let on how easy individual file restores are, else you'll be
constantly handholding the worst idiot in your group.

You know the one - doesn't make backups, wiped all his files, wants them
back yesterday..... and then repeats the exercise every week.

> It also doesn't help solving the problem where I need to start jobs a
> few hours after all the others, and also make sure they finish as fast
> as possible, due to those jobs having a narrower backup window than the
> rest of the jobs.

Understood. For my situation I use different pools for those jobs and a
multiple-drive autochanger. That way the slow jobs can keep lumbering on
while the fast jobs execute on other hardware.

AB


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users