Hi, First: I'm (still) happily running 1.36.1, and the more I use Bacula, the better I find it.
Following the good advice to have 'Rerun Failed Levels = yes' in my Job definition along with a a 'Run Before Job' that check if the host is up and running, I encountered the following problem. I had a FD that was shutdown for a long time, thus each of its jobs were failing as shown by this SQL request: mysql> select JobId,Level,ClientId,JobStatus,StartTime,PoolId,FileSetId from Job where name='pierre'; +-------+-------+----------+-----------+---------------------+--------+-----------+ | JobId | Level | ClientId | JobStatus | StartTime | PoolId | FileSetId | +-------+-------+----------+-----------+---------------------+--------+-----------+ | 41 | F | 3 | T | 2005-03-16 01:10:40 | 1 | 4 | | 45 | I | 3 | T | 2005-03-17 01:08:36 | 3 | 4 | | 51 | I | 3 | E | 2005-03-18 01:08:12 | 3 | 4 | | 57 | I | 3 | T | 2005-03-18 09:56:06 | 3 | 4 | | 61 | I | 3 | E | 2005-03-18 21:33:09 | 3 | 4 | | 68 | D | 3 | T | 2005-03-21 21:36:29 | 2 | 4 | | 75 | I | 3 | T | 2005-03-22 21:33:58 | 3 | 4 | | 82 | I | 3 | T | 2005-03-23 21:33:57 | 3 | 4 | | 89 | I | 3 | E | 2005-03-24 21:34:09 | 3 | 4 | | 96 | I | 3 | E | 2005-03-25 21:32:29 | 3 | 4 | | 103 | D | 3 | E | 2005-03-28 21:36:04 | 2 | 4 | | 110 | I | 3 | E | 2005-03-29 21:35:13 | 3 | 4 | | 117 | I | 3 | E | 2005-03-30 21:33:17 | 3 | 4 | | 124 | I | 0 | f | 2005-03-31 21:35:42 | 0 | 0 | | 131 | I | 0 | f | 2005-04-01 21:35:05 | 0 | 0 | | 138 | F | 0 | f | 2005-04-04 22:56:38 | 0 | 0 | | 145 | I | 3 | T | 2005-04-05 21:36:32 | 3 | 4 | +-------+-------+----------+-----------+---------------------+--------+-----------+ I added the 'Run Before Job' on 3/31/2005, and the Error code, changed from E (error) to f (failed). Two days ago, the backup was a Full backup, but the host was down. Yesterday the backup was an Incremental backup and this particular FD was up and running, I expected it to upgrade its backup to full, but no: 05-Apr 21:36 arsenic-dir: Start Backup JobId 145, Job=pierre.2005-04-05_21.30.03 05-Apr 21:36 pierre-fd: Since time adjusted by -1 seconds. 05-Apr 21:36 arsenic-sd: Volume "Inc-0005" previously written, moving to end of data. 05-Apr 21:46 arsenic-dir: Bacula 1.36.1 (26Nov04): 05-Apr-2005 21:46:43 JobId: 145 Job: pierre.2005-04-05_21.30.03 Backup Level: Incremental, since=2005-03-23 21:33:57 Client: pierre-fd FileSet: "Pierre Set" 2005-03-15 17:18:12 Pool: "Inc-Pool" Storage: "File" Start time: 05-Apr-2005 21:36:32 End time: 05-Apr-2005 21:46:43 FD Files Written: 864 SD Files Written: 864 FD Bytes Written: 2,720,456,391 SD Bytes Written: 2,720,642,779 Rate: 4452.5 KB/s Software Compression: 18.8 % Volume name(s): Inc-0005 Volume Session Id: 16 Volume Session Time: 1112282457 Last Volume Bytes: 3,153,354,566 Non-fatal FD errors: 0 SD Errors: 0 FD termination status: OK SD termination status: OK Termination: Backup OK After reading the code, I found that the problem was indeed the following: * The failed Job because the 'Run Before Job' returned an error have a Job status of Failed and no CliendId nor FileSetId. * The 'Rerun Failed Levels' part of get_level_since_time() in fd_cmds.c does not promote the job to Full or Differential because it searches for a Job that has a valid ClientId. Is it a design limitation of the Run Before Job or is it a bug ? I can file a bug report if necessary. Does anyone have a workaround (besides removing my 'Run Before Job') ? And finally how can I fix this and force an upgrade to Full for this FD ? Thanks for your answer, -- Brice Figureau Days of Wonder ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
