Hi folks,
in the definition of a Job there is the "Run=" option for making
cloned Jobs. It is said in the documentation that by this you can
create the same job again (or a different one), with some changed
parameters (like the storage target).
So, practically this would be the way to do tape mirroring.
At first approach this seems to work (with some lesser pitfalls
which I will come to soon). It results in two objects saved to the
storage media, one from the "primary" job and one from the "cloned".
The big surprise comes when at a later time the created jobs shall
be migrated from one storage to another. That migration will work,
but after it has completed, the old cloned job will be run again
with its original setup!
So what do we have:
1. primary job -> primary object on media 1a
2. cloned job -> mirror object on media 1b
and at a later time
3. migration a -> move primary object to media 2a
4. migration b -> move mirrored object to media 2b
So far that would be what is intended.
But then happens without reason and out of schedule:
5. cloned job -> new "mirror" object on media 1b
When this is an ordinary filespace backup, then it is just a nuisance
and fills up media and disturbs schedule.
It is worse when it is some special "intelligent" job doing
some interlocking, like moving out database redo-logs which are
not redundant and MUST exist COMPLETELY.
In any case, I consider it a bug.
rgds,
PMc
--
..having tossedaway Microsoft(tm) already in 1991!
-------------------------------------------------------------------------
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@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
