On Tuesday 06 February 2007 09:57, David Romerstein wrote:
> On Tue, 6 Feb 2007, [EMAIL PROTECTED] wrote:
> > I want to backup a larger amount of data (approx. 800GB and a total of 2
> > million files) with a library of Ultrium-3 tapes.
>
> That, in essence, is why I first installed bacula - to backup roughly 2TB
> of data just once (my file count is *significantly* higher than yours,
> though - on the order of 200 million files across 10s of thousands of
> directories).

        I see no problem with what you describe. I would recommend, only, 
performing 
more than one backup, or duplicating your tape(s) after the backup, so that 
you have more than one copy of your files, possibly stored in different 
locations.

> > After a succesful backup, the data will be removed from the disk storage
> > and the tapes will be taken offsite. Probably, it will be necessary to
> > restore certain files from the backup, but this case should not occur.
> > In other words, I want to backup a file set only once (perhaps do a
> > complete restore and compare the files), store the catalog offsite,
> > remove the used tapes from the library and wipe the disk storage and
> > perhaps clear the catalog.
>
> If, by 'clear the catalog', you mean 'erase the bacula database entries',
> I would not recommend doing that. Otherwise, this should be dead simple.

        It would seem to me that the principle advantage of using Bacula for 
your 
backup would be that you had a history of your backup(s) in the database for 
future reference. If you purge the database, why go to all the bother of 
setting up Bacula and, instead, just use tar or cpio to create your backup 
tapes.

> What I did was created a set of jobs in my bacula-dir.conf file (I had
> already discovered that backing up the entire fileset takes longer than
> bacula will allow for a single backup job). Each of these jobs were
> defined individually, although I probably could have created a JobDef
> entry for most of this. I did *not* use the DefaultJob JobDef, as that has
> a weekly schedule. My jobs were all variations on the following:
>
> Job {
>    Name = "BackupFileStore_pt1"
>    Type = Backup
>    Level = Full
>    FileSet="xRaid1_pt1"
>    Storage = Ultrium
>    Client = srv01-fd
>    Messages = Standard
>    Pool = Default
>    Priority = 5
> }
>
> I then dumped to a text file a listing of all directories on the filestore
> I was backing up. I broke that list into 4 pieces, and used those pieces
> to generate FileSets in my bacula-dir.conf:
>
> FileSet {
>    Name = "xRaid1_pt1"
>    Include {
>      Options {
>        signature = MD5
>      }
>      File = "</data/staging/xRaid1/part1"
>    }
> }
>
> Then, all I had to do was put in a tape, run bconsole, and start the job.
>
> -- D

Cheers!

cmr
-- 
Debian 'Etch' - Registered Linux User #241964
--------
"More laws, less justice." -- Marcus Tullius Ciceroca, 42 BC

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to