Ralf, thanks for your answer.
As far as I can tell, the masquerading seems to work, the SD can contact both the FD and the DIR Here is a dump of such an error in my logfiles, in this one, you clearly see the error "The sizes do not match" <snip> 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3899: Begin pruning Jobs. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3899: Pruned 1 Job for client payclear.be-fd from catalog. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3899: Begin pruning Files. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3899: Pruned Files from 1 Jobs for client payclear.be-fd from catalog. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3899: End auto prune. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: Start Backup JobId 3904, Job=RelayFull.2008-12-01_23.05.42 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: Using Device "wicenas5.xxx.xxx.WICE" 01-Dec 23:33 relay.xxx.xxx-fd: DIR and FD clocks differ by 92 seconds, FD automatically adjusting. 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Volume "wicenas5wice0899" previously written, moving to end of data. 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Error: Bacula cannot write on disk Volume "wicenas5wice0899" because: The sizes do not match! Volume=3999950486 Catalog=5690031 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Marking Volume "wicenas5wice0899" in Error in Catalog. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: There are no more Jobs associated with Volume "wicenas5wice0606". Marking it purged. 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: All records pruned from Volume "wicenas5wice0606"; marking it "Purged" 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: Recycled volume "wicenas5wice0606" 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Fatal error: ../../stored/ label.c:481 Truncate error on device "wicenas5.xxx.xxx.WICE" (E:\Bacula \WICE): ERR=../../stored/dev.c:1971 Unable to stat device "wicenas5.xxx.xxx.WICE" (E:\Bacula\WICE). ERR=The handle is invalid.^M 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Marking Volume "wicenas5wice0606" in Error in Catalog. 01-Dec 23:30 wicenas5.xxx.xxx-sd JobId 3904: Fatal error: Job 3904 canceled. 01-Dec 23:33 relay.xxx.xxx-fd: RelayFull.2008-12-01_23.05.42 Fatal error: job.c:1665 Bad response to Append Data command. Wanted 3000 OK data , got 3903 Error append data 01-Dec 23:31 bacula001.xxx.xxx-dir JobId 3904: Error: Bacula bacula001.xxx.xxx-dir 2.4.3 (10Oct08): 01-Dec-2008 23:31:48 Build OS: i686-pc-linux-gnu debian 4.0 JobId: 3904 Job: RelayFull.2008-12-01_23.05.42 Backup Level: Incremental, since=2008-11-30 23:32:45 Client: "relay.xxx.xxx-fd" i386-pc-linux-gnu,debian, 3.1 FileSet: "RelayFullSet" 2008-05-28 14:28:11 Pool: "PoolWicenas5Wice" (From Job resource) Storage: "wicenas5.xxx.xxx.wice-sd" (From Job resource) Scheduled time: 01-Dec-2008 23:05:01 Start time: 01-Dec-2008 23:31:47 End time: 01-Dec-2008 23:31:48 Elapsed time: 1 sec Priority: 10 FD Files Written: 0 SD Files Written: 0 FD Bytes Written: 0 (0 B) SD Bytes Written: 0 (0 B) Rate: 0.0 KB/s Software Compression: None VSS: no Storage Encryption: no Volume name(s): Volume Session Id: 188 Volume Session Time: 1226699290 Last Volume Bytes: 1 (1 B) Non-fatal FD errors: 0 SD Errors: 0 FD termination status: Error SD termination status: Error Termination: *** Backup Error *** </snip> Here is another (I think, related) error, the volume wicenas5pix0641 does exist and has a size of 64K . <snip> 29-Nov 23:05 bacula001.xxx.xxx-dir JobId 3847: Start Backup JobId 3847, Job=Chap01Full.2008-11-29_23.05.56 29-Nov 23:05 bacula001.xxx.xxx-dir JobId 3847: There are no more Jobs associated with Volume "wicenas5pix0640". Marking it purged. 29-Nov 23:05 bacula001.xxx.xxx-dir JobId 3847: All records pruned from Volume "wicenas5pix0640"; marking it "Purged" 29-Nov 23:05 bacula001.xxx.xxx-dir JobId 3847: Recycled volume "wicenas5pix0640" 29-Nov 23:05 bacula001.xxx.xxx-dir JobId 3847: Using Device "wicenas5.xxx.xxx.Pixelixir" 29-Nov 23:18 chap01-fd JobId 3847: DIR and FD clocks differ by 429 seconds, FD automatically compensating. 29-Nov 23:08 wicenas5.xxx.xxx-sd JobId 3847: Labeled new Volume "wicenas5pix0640" on device "wicenas5.xxx.xxx.Pixelixir" (F:\Bacula \Pixelixir). 29-Nov 23:08 wicenas5.xxx.xxx-sd JobId 3847: Wrote label to prelabeled Volume "wicenas5pix0640" on device "wicenas5.xxx.xxx.Pixelixir" (F: \Bacula\Pixelixir) 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: User defined maximum volume capacity 4,000,000,000 exceeded on device "wicenas5.xxx.xxx.Pixelixir" (F:\Bacula\Pixelixir). 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: End of medium on Volume "wicenas5pix0640" Bytes=3,999,937,719 Blocks=62,003 at 29-Nov-2008 23:10. 29-Nov 23:12 bacula001.xxx.xxx-dir JobId 3847: There are no more Jobs associated with Volume "wicenas5pix0641". Marking it purged. 29-Nov 23:12 bacula001.xxx.xxx-dir JobId 3847: All records pruned from Volume "wicenas5pix0641"; marking it "Purged" 29-Nov 23:12 bacula001.xxx.xxx-dir JobId 3847: Recycled volume "wicenas5pix0641" 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: Fatal error: ../../stored/ label.c:481 Truncate error on device "wicenas5.xxx.xxx.Pixelixir" (F: \Bacula\Pixelixir): ERR=../../stored/dev.c:1971 Unable to stat device "wicenas5.xxx.xxx.Pixelixir" (F:\Bacula\Pixelixir). ERR=The handle is invalid. 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: Marking Volume "wicenas5pix0641" in Error in Catalog. 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: Fatal error: Job 3847 canceled. 29-Nov 23:10 wicenas5.xxx.xxx-sd JobId 3847: Job write elapsed time = 00:01:28, Transfer rate = 45.42 M bytes/second 29-Nov 23:20 chap01-fd JobId 3847: Fatal error: ../../filed/backup.c: 892 Network send error to SD. ERR=Input/output error 29-Nov 23:20 chap01-fd JobId 3847: Error: ../../lib/bsock.c:306 Write error sending 65536 bytes to Storage daemon:82.175.193.69:9103: ERR=Input/output error 29-Nov 23:12 bacula001.xxx.xxx-dir JobId 3847: Error: Bacula bacula001.xxx.xxx-dir 2.4.3 (10Oct08): 29-Nov-2008 23:12:59 Build OS: i686-pc-linux-gnu debian 4.0 JobId: 3847 Job: Chap01Full.2008-11-29_23.05.56 Backup Level: Incremental, since=2008-11-24 23:09:36 Client: "chap01-fd" 2.4.3 (10Oct08) Linux,Cross- compile,Win32 FileSet: "Chap01FullSet" 2008-11-21 23:05:00 Pool: "PoolPixelixir" (From Job resource) Storage: "wicenas5.xxx.xxx.pixelixir-sd" (From Job resource) Scheduled time: 29-Nov-2008 23:05:00 Start time: 29-Nov-2008 23:11:38 End time: 29-Nov-2008 23:12:59 Elapsed time: 1 min 21 secs Priority: 10 FD Files Written: 223 SD Files Written: 223 FD Bytes Written: 3,997,232,773 (3.997 GB) SD Bytes Written: 3,997,002,772 (3.997 GB) Rate: 49348.6 KB/s Software Compression: None VSS: no Storage Encryption: no Volume name(s): wicenas5pix0640|wicenas5pix0641 Volume Session Id: 159 Volume Session Time: 1226699290 Last Volume Bytes: 64,748 (64.74 KB) Non-fatal FD errors: 1 SD Errors: 0 FD termination status: Error SD termination status: Error Termination: *** Backup Error *** </snip> Thanks a lot for having a look at this, it is very appreciated Regards, Sébastien On 04 Dec 2008, at 10:42, Ralf Gross wrote: > S?bastien Delneste schrieb: >> >> I'm administrating an heterogeneous hosting platform and we have all >> kinds of servers which we backup on 2 different storages using one >> director. >> >> One storage is running great but the second one is giving me problems >> when I try to re-use volumes (which is the plan, of course) >> >> Here is the kind of error I get : >> Error: Bacula cannot write on disk Volume "wicenas5wice0747" because: >> The sizes do not match! Volume=3999942850 Catalog=1135996289 >> >> My director is running 2.4.3 on a Debian Linux with a MySQL backend >> My clients are different versions on Windows and on Linux and they >> all >> have the problem >> My Storage is running 2.4.3 on Windows 2003 Server. >> >> There are two things I can think of which could be slightly different >> from a standard setup >> >> 1?) I do some masquerading when connecting to the storage >> 2?) I use the same device wicenas5.xxx.xxx twice in my storage >> definitions >> >> The first question is, of course, if there is any way to fix a >> corrupted volume or to recalculate its size but the other question >> (more important to me) is how do I avoid such a problem ? > > > Anything in baculas logfile? Regarding masquerading, if I'm right the > SD needs to send the attributes to the DIR. Is this possible with > your setup? > > Ralf > > > ------------------------------------------------------------------------- > 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 > [email protected] > https://lists.sourceforge.net/lists/listinfo/bacula-users ------------------------------------------------------------------------- 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 [email protected] https://lists.sourceforge.net/lists/listinfo/bacula-users
