Now that we know there is a way to use multiple disks without adding the
'moving symlink' script (thanks Kern),

is there by any chance an undocumented trick to write two copies of the
same job at once? (aka RAIT.) Like maybe listing two Pools in the JobDef?

The goal is to have 2 copies of the volume file on 2 different disks so
that if the "main" disk decides to shed magnetic layer off its platters,
I can remount the "copy" disk as bacula's backup Device and keep going.

The closest I can see in the docs is a Copy job, and that has a few
problems:

1. Documentation is sketchy.
2. I'm not sure what to make of "when a job record is purged (deleted)
from the catalog, it will promote the copy": I'm looking for an exact
copy I can just swap in if the original dies of media failure. No
metadata tweaking required.
3. There seems to be room for media failure between the time main jobs
finish and the copy job kicks in. It also seems a media failure in the
middle of the main job will kill the whole run. With RAIT, you write two
copies at once and if one device fails you just keep going, writing to
the other one.

So am I better off just rsync'ing the disks from RunScript?

TIA
-- 
Dimitri Maziuk
Programmer/sysadmin
BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu

signature.asc
Description: OpenPGP digital signature

------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and 
their applications. This 200-page book is written by three acclaimed 
leaders in the field. The early access version is available now. 
Download your free book today! http://p.sf.net/sfu/neotech_d2d_may

_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users