Hello, Several points:
1. You have apparently modified the test. What you show below does not at all correspond to the virtual-backup-test that we are currently running. 2. In a number of the regression tests, there is a need for more testing and verification. This is a project that we (mostly Eric) are working on, but it is not at all easy to use portable shell scripting to check the details of a job. 3. If you want to submit some code that improves the testing, we would be happy to see it, but one needs to take a lot of care to make sure it is portable and will run on all Unix flavors ... Regards, Kern On Monday 20 October 2008 11:32:03 Graham Keeling wrote: > Hello, > It looks like tests/virtual-backup-test is working well for me now, so I > tried tests/virtual-backup2-test. > > At first glance, it appeared to work. It would output something like this: > > === Starting virtual-backup-test at 10:07:11 === > > > > ===== virtual-backup-test OK 10:09:00 ===== > > However, on closer inspection, it turns out that the two Virtual Full > backups failed (JobIds 6 and 9). Here is the 'list jobs' table from the end > of the test, when I run it with REGRESS_DEBUG on (tidied up for ease of > reading): > > +---+--------------+---------------------+-------+-------+----------+------ >---+ > > | J | Name | StartTime | Level | Files | JobBytes | > | JStatus | > > +---+--------------+---------------------+-------+-------+----------+------ >---+ > > | 1 | Vbackup | 2008-10-20 10:01:20 | F | 12107 | 90453446 | T > | | 2 | Vbackup | 2008-10-20 10:01:46 | I | 43 | 931145 | T > | | 3 | Vbackup | 2008-10-20 10:01:50 | D | 86 | 3648605 > | | T | 4 | Vbackup | 2008-10-20 10:01:55 | I | 44 | > | 2717489 | T | 5 | RestoreFiles | 2008-10-20 10:02:00 | F | > | 12107 | 90453446 | T | 6 | Vbackup | 2008-10-20 10:02:14 | F > | | 0 | 0 | f | 7 | RestoreFiles | 2008-10-20 10:02:19 | > | F | 12107 | 90453446 | T | 8 | Vbackup | 2008-10-20 > | 10:02:33 | I | 43 | 2717460 | T | 9 | Vbackup | > | 2008-10-20 10:02:38 | F | 0 | 0 | f | 10 | > | RestoreFiles | 2008-10-20 10:02:42 | F | 12107 | 90453446 | T | > > +---+--------------+---------------------+-------+----------+-------+------ >---+ > > Further back in the debug output, I find these: > > 20-Oct 10:02 localhost-dir JobId 6: Fatal error: Job canceled. Attempt to > > read and write same device. > > Read storage "DiskChanger" (From Job resource) -- Write storage > > "DiskChanger" (From Storage from Pool's NextPool resource) > > > > 20-Oct 10:02 localhost-dir JobId 9: Fatal error: Job canceled. Attempt to > > read and write same device. > > Read storage "DiskChanger" (From Job resource) -- Write storage > > "DiskChanger" (From Storage from Pool's NextPool resource) > > It seems wrong to me that the thing that it is trying to test (Virtual Full > backups) is failing completely, and yet it reports success. > > If it can't detect that a job has failed, perhaps it can do something like > purging the jobs/volumes that came before the Virtual Full. That means that > when it restores after the Virtual Full, it will definitely be restoring > from the Virtual Full, rather than the original backups. > > Graham. > > > ------------------------------------------------------------------------- > 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-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/bacula-devel ------------------------------------------------------------------------- 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-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/bacula-devel
