Hi,
23.11.2008 03:21, Maxime wrote:
> Hello,
>
> Recently I had to restore data from purged volumes that hadn't been
> recycled hopefully. It's files on hard drives.
> I rebuild the catalog with bscan.
> And I launch restoration jobs.
>
> In bacula-sd.conf, we can find :
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> Device {
> Name = FileStorage1
> Media Type = File
> Archive Device = /bacula/disk01/
>
> LabelMedia = yes; # lets Bacula label unlabeled media
> Random Access = Yes;
> AutomaticMount = yes; # when device opened, read it
> RemovableMedia = no;
> AlwaysOpen = no;
> }
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
> and in bacula-dir.conf
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> Storage {
> Name = backup01c-sd
> Address = backup01c.univ-avignon.fr
> SDPort = 9103
> Password = "xxxxxxxxx"
> Device = FileStorage1
> Media Type = File
>
> TLS Enable = yes
> TLS Require = yes
> ...
> }
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
>
> There is also FileStorage2 defined with /bacula/disk02/ partition
> (Storage backup01c-sd2).
> *All my old volumes are located on /bacula/disk02/*, but just before
> launching restore job, I could read :
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> The job will require the following
> Volume(s) Storage(s) SD Device(s)
> ===========================================================================
>
> Full-0113 backup01c-sd2 FileStorage2
>
> Inc-0263 backup01c-sd FileStorage1
>
> Inc-0373 backup01c-sd FileStorage1
>
> Inc-0406 backup01c-sd FileStorage1
>
> Inc-0412 backup01c-sd FileStorage1
>
> Inc-0466 backup01c-sd2 FileStorage2
>
> Inc-0486 backup01c-sd2 FileStorage2
>
> Inc-0546 backup01c-sd2 FileStorage2
>
> Inc-0566 backup01c-sd2 FileStorage2
>
> Inc-0072 backup01c-sd2 FileStorage2
>
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Hmm... restoring from two storage devices is not expected to work
without manual intervention. You might have to separate the bootstrap
file manually, so that you can run a number of jobs each using only
one storage device.
For disk based volumes, it might be easier (temporarily) assigning the
volumes to one common pool and soft-link them to a common directory,
i.e. storage device.
> But all volumes had been read from /bacula/disk02/ since job report
> displays :
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> 22-Nov 18:35 backup01c-sd JobId 2924: Ready to read from volume
> "Inc-0263" on device "FileStorage2" (/bacula/disk02/).
> 22-Nov 18:36 backup01c-sd JobId 2924: End of Volume at file 2 on device
> "FileStorage2" (/bacula/disk02/), Volume "Inc-0263"
> 22-Nov 18:36 backup01c-sd JobId 2924: Ready to read from volume
> "Inc-0373" on device "FileStorage2" (/bacula/disk02/).
> 22-Nov 18:36 backup01c-sd JobId 2924: End of Volume at file 0 on device
> "FileStorage2" (/bacula/disk02/), Volume "Inc-0373"
> 22-Nov 18:36 backup01c-sd JobId 2924: Ready to read from volume
> "Inc-0406" on device "FileStorage2" (/bacula/disk02/).
> 22-Nov 18:37 backup01c-sd JobId 2924: End of Volume at file 2 on device
> "FileStorage2" (/bacula/disk02/), Volume "Inc-0406"
> 22-Nov 18:37 backup01c-sd JobId 2924: Ready to read from volume
> "Inc-0412" on device "FileStorage2" (/bacula/disk02/).
> 22-Nov 18:39 backup01c-sd JobId 2924: End of Volume at file 2 on device
> "FileStorage2" (/bacula/disk02/), Volume "Inc-0412"
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
>
> I think bacula is confused because there are non purged volumes named
> Inc-0263, Inc-0373, Inc-0406, and Inc-0412 on FileStorage1 :
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> *list volume=Inc-0406
> +---------+------------+-----------+---------+---------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> | MediaId | VolumeName | VolStatus | Enabled | VolBytes | VolFiles
> | VolRetention | Recycle | Slot | InChanger | MediaType | LastWritten
> |
> +---------+------------+-----------+---------+---------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> | 406 | Inc-0406 | Used | 1 | 9,337,460,938 | 0
> | 2,592,000 | 1 | 0 | 0 | File | 2008-10-21
> 08:27:27 |
> +---------+------------+-----------+---------+---------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> *list volumes jobid=2542
> Jobid 2542 used 1 Volume(s): Inc-0406
> +-------+------------+---------------------+------+-------+----------+-------------+-----------+
> | JobId | Name | StartTime | Type | Level | JobFiles |
> JobBytes | JobStatus |
> +-------+------------+---------------------+------+-------+----------+-------------+-----------+
> | 2,542 | job_web01c | 2008-11-04 01:48:10 | B | I | 2,982 |
> 559,944,881 | T |
> +-------+------------+---------------------+------+-------+----------+-------------+-----------+
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
> Dates don't match...
>
>
> What brings me to the points :
> 1) There are purged volumes on a partition (storage device) and non
> purged volumes identically named on the other partition. Is my bacula
> configuration weird or is it a normal state?
If you've got two volumes with the same name that is definitely a problem.
If those volumes have different IDs in the catalog, you /only/ have a
user interface problem :-)
If the volumes represent different historical snapshots of the same
volume you'd better use bextract to specify which files to use.
bscanning old states of volumes into the catalog will surely scre up
your catalog as that will override the current inventory of things on
those volumes.
> 2) If everything is fine, bacula console shouldn't have displayed it
> will use volumes from backup01c-sd FileStorage1. So is it a bug?
Hardly. I rather suggest to use bextract to get the files you need...
setting volumes to an older state and expecting Bacula to correctly
handle two versions of the same volume is probably a bit too much :-)
Arno
> Thanks for your comments
> --
> Maxime CHARPENNE
> Université d'Avignon et des Pays de Vaucluse - CRI
> Tél. : +33 4 90 16 26 05
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Bacula-users mailing list
> Bacula-users AT lists.sourceforge DOT net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
--
Arno Lehmann
IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück
www.its-lehmann.de
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
