Am Donnerstag, 1. März 2007 15:51 schrieb Arno Lehmann: > Hi, > > On 3/1/2007 2:55 PM, Thomas Franz wrote: > > Am Mittwoch, 28. Februar 2007 22:02 schrieben Sie: > >>>>>>>On Wed, 28 Feb 2007 13:41:25 +0100, Thomas Franz said: > >>> > >>>Hello, > >>> > >>>I am running bacula for a lot of month, without any problems. > >>>But yesterday I've got this message: > >>>............ > >>>27-Feb 11:05 dssbkup-sd: Volume "A00040" previously written, moving to > >>>end of data. > >>>27-Feb 11:06 dssbkup-sd: Backup.2007-02-27_11.05.00 Error: I cannot > >>> write on Volume "A00040" because: > >>>The number of files mismatch! Volume=456 Catalog=457 > >>>27-Feb 11:06 dssbkup-sd: Marking Volume "A00040" in Error in Catalog. > >>>................... > >>> > >>>OS: FreeBSD 6.1 > >>>Bacula: 1.38.9 > >>>TAPE : LTO-3 (IBM Ultrium-TD3) > >>>Library: Exabyte > >>> > >>>It was the first run after a system restart. ( But this was not the > >>> first time bacula runs after a restart) . > >>> > >>>I did a "bls -c bacula-sd.conf -j -V A00040 /dev/nsa3 : > >>>------------------------------------------------------------------------ > > ... > > >>>27-Feb 22:41 bls: bls Error: block.c:945 Read error at file:blk 457:0 on > >>>device "LTO-3" (/dev/nsa3). ERR=Operation not permi > >>>tted. > >>>27-Feb 22:41 bls: End of Volume at file 457 on device "LTO-3" > >>>(/dev/nsa3), Volume "A00040" > >>>27-Feb 22:41 bls: End of all volumes. > >>>------------------------------------------------------------------ > >>> > >>>Job 750 was the last backup at this time ( it was the backup of the > >>>Catalog itself) . I could restore this file and it seems to be OK. > >> > >>Since you got a read error at the end, it looks like something is wrong > >>with the data on the tape. A few possibilities I can think of: > >> > >>1) The EOT model in Bacula now doesn't match the EOT model in Bacula when > >>the tape was written. > >> > >>2) The EOT model in the drive doesn't match the EOT model in Bacula (or > >>didn't match when the tape was written). > >> > >>3) The EOT model in the drive is not supported by the drive correctly. > >> > >>4) The EOM handling is not working and the tape is 100% full. > > > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > >^ Hello Martin > > > > thank you very much for your help. > > There was fact I did not knew until now. It was not a regular restart of > > the server. The system crashed because of a power blackout. > > Ok, than this sort of problem can be expected. > > > In this case I could image that this is a problem for bacula to find the > > correct end of the tape (AlwaysOpen = yes). > > Well, kind of... I'd phrase it differently: In this case this is a > problem because the file mark needed is not on the tape. > > > I will try to manipulate the end-of-file marks at the end of this tape. > > I'd recommend against that... it's possible you do more damage, and so > I'd simply set the volume status to 'Used', allowing Bacula to use the > tape for restores (which probably is one of the essential things) and to > recycle the volume normally. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Nevertheless I did it. I wrote a second filemark after the last and now all seems to be OK. But of course , I made a security copy of this tape before. I used the FreeBSD utility "tcopy" and this program has already eliminated this "end of tape" problem. The copied tape was already OK.
Thomas > > Unless you absolutely need the remaining capacity of that tape (which > would be serious problem, in my opinion :-) that should be ok. > > Arno > > > Thanks a lot. > > > > Thomas > > > >>>And I did this btape-run: > >>>------------------------------------------- > >>>Tape block granularity is 1024 bytes. > >>>btape: butil.c:272 Using device: "/dev/nsa3" for writing. > >>>28-Feb 09:43 btape: 3301 Issuing autochanger "loaded drive 0" command. > >>>28-Feb 09:43 btape: 3302 Autochanger "loaded drive 0", result is Slot 3. > >>>28-Feb 09:43 btape: 3301 Issuing autochanger "loaded drive 0" command. > >>>28-Feb 09:43 btape: 3302 Autochanger "loaded drive 0", result is Slot 3. > >>>btape: btape.c:338 open device "LTO-3" (/dev/nsa3): OK > >>>*rewind > >>>btape: btape.c:438 Rewound "LTO-3" (/dev/nsa3) > >>>*status > >>> Bacula status: file=0 block=0 > >>> Device status: file=0 block=0 > >>>btape: btape.c:1759 Device status: 1. ERR= > >>>*eod > >>>btape: btape.c:490 Moved to end of medium. > >>>*status > >>> Bacula status: file=456 block=0 > >>> Device status: file=456 block=0 > >>>btape: btape.c:1759 Device status: 1. ERR= > >>>*quit > >>>Orphaned buffer: 16 bytes allocated at line 276 of btape reserve.c > >>>------------------------------------------------ > >>> > >>>The content of the catalog for this volume : see attachment > >> > >>It looks like the catalog value matches what bls prints, except for the > >>final read error. The btape eod doesn't match, which again indicates an > >>EOT/EOM problem. > >> > >>Have you run the btape tests with you current settings? > >> > >>__Martin > > > > ------------------------------------------------------------------------- > > Take Surveys. Earn Cash. Influence the Future of IT > > Join SourceForge.net's Techsay panel and you'll get the chance to share > > your opinions on IT & business topics through brief surveys-and earn cash > > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > _______________________________________________ > > Bacula-users mailing list > > Bacula-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/bacula-users ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
