Excuse me, I did send this mail to Tom directly, but as everybody should be able to read it, I do post it to bacula-users again.
[EMAIL PROTECTED] (Eric Böse-Wolf) writes: > Tom Meiner <[EMAIL PROTECTED]> writes: > >> Hi! >> >> I have set up a bacula 2.0.3 server with an autochanger and several >> clients. My default pool directs to the tape. So my backup lasts a long >> time because all clients have to wait while one client does his backup. >> To speed up the backup I consider to create an additional disk pool as >> the default pool to get more then one client backed up simultaneous and >> than migrate the files to the next tape which lays on the tapes. >> >> Although I heave read http://www.bacula.org/rel-manual/Migration.html >> and all the referenced parts of the manual I haven't found an answer to >> one of this questions. >> >> >> 1. Is bacula aware to update the bootstrap files >> (/var/spool/bacula/client.bsr) after migrating the files from disk >> pool to tape pool? > > The manual says "write bootstrap" is only for backup jobs, but I would > test if it doesn't work in a migration job too. > >> 2. Is it possible to begin to migrate the files automatically from disk >> pool to tape if the disk pool grows up to i.e. 60%? > > "Migration High Bytes" and a migration job with PoolOccupancy selection > type, which runs each hour or every ten minutes as you like. > >> 3. What is the best way to create this disk pool (2 files sized 1 GB >> each?)? > > Pool { > Name = <your pool name here> > Pool Type = Backup > Recycle = yes > AutoPrune = yes > Volume Retention = 100 days > Label Format = "Volume-" > Maximum Volume Bytes = 1073741824 #1Gb per Volume > Maximum Volumes = 2 #Max 2 Volumes > Storage = <your storage resource here> > Next Pool = <your tape pool here> > Migration High Bytes = 1288490188 #60% of 2GB > } > > As I'm new to bacula, too, I would take that with a grain of salt :-) > Maybe you wait until some grown up confirms of neglects this. > >> 4. How big has this disk pool to be? What happens with the backup job if >> the disk pool becomes full. Does is automatically start the migration >> and after that continues backing up that client? > > I don't know, but for the questions you asked, I would imagine that you > don't need migration but data spooling. With data spooling the > configuration is much easier, it happens automagically and if the spool > device is full it automagically begins to despool data. Hmm, spooling is > done on a per job basis, so may it's not what you want, as each client > has again to wait until the previous job despooled. Again: I'm new to > bacula too and this information may be inaccurate. > > mfg > > Eric -- Das Briefgeheimnis sowie das Post- und Fernmeldegeheimnis sind unverletzlich. (Grundgesetz, Artikel 10, Abs. 1) http://www.gnupp.de http://www.gnupg.org http://www.stud.uni-karlsruhe.de/~uwaru/
pgpSq0sJg2zIb.pgp
Description: PGP signature
------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. 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
