Alan Brown wrote:
> On Wed, 28 May 2008, Tore Anderson wrote:
> 
>> Waiting for one job to finish (or having to cancel only one job) before
>> the restore process starts is OK.  Having to cancel all running jobs
>> (which might have been at it for hours) to get the restore job going
>> sucks.  The best would be if there was always a free slot dedicated to
>> restore processes, come to think of it.
> 
> Given your situation (SLA requirements), you really would be better off
> making sure there's always a free drive in the changer.
> 
> Or another changer/drive available for restores.

Not that I have direct experience in this, but I've always thought a 
dedicated drive for restores is a good idea.  Depending on a given 
situation, perhaps more than one drive.

-- 
Dan Langille

BSDCan - The Technical BSD Conference : http://www.bsdcan.org/
PGCon  - The PostgreSQL Conference:     http://www.pgcon.org/

-------------------------------------------------------------------------
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