Hi, On 12/8/2006 9:38 PM, Forrest England wrote: > Hi there, I work with Rob Ostrander, the initiator of this thread. > I've also been working on the problem of bacula reporting that > the number of VolumeFiles mismatch between the volume and the catalog. > > After reading the manual thoroughly and poking around the web, I'm still not > sure > what VolumeFiles referrs to.
In short: These are simply special marks on the tape that the tape drive uses to position the tape. > an example: I did a test backup job of 101 files on a fresh tape which > completed successfully: > FD Files Written: 101 > SD Files Written: 101 > afterwards I attempted to do a second backup job on the same tape, but got > the familiar error message: > > 08-Dec 13:36 vtsan1-sd: test.2006-12-08_13.36.27 Error: I cannot write on > Volume "TestVolume" because: > The number of files mismatch! Volume=0 Catalog=1 > > I updated the VolumeFiles for this tape in the catalog to 0 and did a second > backup job. > I restored and verified both backup jobs without any problem. That was lucky... Bacula should have considered that tape as empty IMO. > what does VolumeFiles correspond to? It's obviously not the number of actual > files on the tape. > is it supposed to correspond to the number of jobs on a tape? > > anyway, many people seem to have this VolumeFiles mismatch problem. we have > it on 3 different servers > with 3 different tape changers, so I'm going to file a bug report. Make sure to include your SD device configuration and btape test output... Ok, a short description of the file mark problem: Bacula segments tape volumes by file marks. These are, in tape hardware language, called EOF markers. Bacula keeps the number of EOFs on tape in the catalog. When a tape is appended to, bacula lets the tape drive seek to the end of data on tape (called EOD). The tape drive counts the EOFs it crosses and reports that number as the current file position. That number is compared to the catalog data, and that comparison is the point where your error comes from. Unfortunately, EOF marks can be a little problematic. Some OS/tape drivers use two EOFs to indicate EOD. In such a case, before you can continue to write to that tape, you have to backspace one EOF mark. If you didn't, the data would be written to tape at a position that could not be reached by the normal positioning commands of the tape drive. For practical purposes this boils down to the advice to seriously test your settings of Two EOF, BSF at EOM, and the various Bachward/Forward Space File/Record variants. To give more detailed help, we would at least need to know which OS and which tape hardware you run. I'm quite certain that most OSes have their Bacula users, which might supply configuration examples. Arno -- IT-Service Lehmann [EMAIL PROTECTED] Arno Lehmann http://www.its-lehmann.de ------------------------------------------------------------------------- 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
