On Wed, 2007-07-11 at 10:15 +0100, Martin Simmons wrote:
> >>>>> On Tue, 10 Jul 2007 10:50:46 -0700, Craig White said:
> >
> > I have a 3 week rotation with a full backup each Friday and differential
> > backups every weekday.
> >
> > Last night, upon execution, the 1_Monday_Week_1 AND 1_Monday_Week_2 were
> > both purged and it shouldn't have purged 1_Monday_Week_2
> >
> > from my backup logs...
> >
> > 09-Jul 20:00 SRV1: Start Backup JobId 91,
> > Job=SRV1Backup.2007-07-09_20.00.00
> > 09-Jul 20:00 SRV1: Pruned 3 Jobs on Volume "1_Monday_Week_1" from
> > catalog.
> > 09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
> > "1_Monday_Week_1"; marking it "Purged"
> > 09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
> > "1_Monday_Week_2"; marking it "Purged"
> > 09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1"
> > 09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1" on device
> > "DDS-4" (/dev/nst0), all previous data lost.
> >
> > My MondayPool is defined as...
> > Pool {
> > Name = MondayPool
> > Pool Type = Backup
> > Recycle = yes
> > AutoPrune = yes
> > Volume Retention = 19d
> > Volume Use Duration = 4d
> > Maximum Volume Jobs = 5
> > }
> >
> > and according to list media...
> > | MediaId | VolumeName | VolStatus | Enabled | VolBytes |
> > VolFiles | VolRetention | Recycle | Slot | InChanger | MediaType |
> > LastWritten |
> > +---------+-----------------+-----------+---------+----------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> > | 2 | 1_Monday_Week_2 | Purged | 1 | 25,815,250,944 |
> > 26 | 1,641,600 | 1 | 0 | 1 | DDS-4 | 2007-06-25
> > 22:07:04 |
> >
> > My 1_Monday_Week_2 was last written on 6/25/2007 which is clearly not 19
> > days of Volume Retention.
> >
> > Is there something wrong with my understanding of this?
> >
> > # rpm -qa|grep bacula
> > bacula-mysql-2.0.3-1
> > bacula-mtx-2.0.3-1
> > bacula-wxconsole-2.0.3-1
> > bacula-gconsole-2.0.3-1
>
> Do you have previous logs still? If so, check for other messages about
> 1_Monday_Week_2, because it looks like all the jobs had been removed before it
> ran the above. Also, maybe they were removed with the delete command?
----
I don't - but I think I figured out why this happened. Two weeks ago, I
ended up running a 'Full' backup on Monday instead of the Differential
because I changed the FileSet selectors (actually the 'Exclude' portion)
and that made it run onto a second tape. After I had another 'Full'
backup on Friday, I deleted 2_Monday_Week_2 and internally, Bacula must
have figured that since the continuation tape for that particular 'Full'
set was deleted, then the first tape of that set was worthless too so it
just purged it.
I guess I don't mind, but it was clearly unexpected. I did notice that
the same thing didn't happen last night on the TuesdayPool nor did it
happen at any other time so it definitely caught me by surprise. That
Bacula had reached ahead a week in time and 'purged' a backup without
cause didn't make sense given the objective of Bacula to hold on to all
possible backups as long as needed which is also why I have set the
'retention period' to be 19 days on a 3 week rotation.
Thanks
--
Craig White <[EMAIL PROTECTED]>
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
