Hello We're using bacula on our productiv systems. It works well except for one thing which is annoying. As we took a closer look to the tranfered filesize, the filesize of the differential backup was bigger than the fullbackup itself. But not many files has really changed since then. So we traced down the problem. We started bacula-fd in the debug mode and analized the log output[1].
As in protocol specification the director transmit the date of last fullbackup or incrementalbackup (depending on the backuplevel) to the file daemon. So far so good, but it seems, that the date is the really beginning of the job not, as expected, the date it saves the first file. So all files created or changed by running "RunBeforeJob" (this job runs for about 10 minutes) has a *newer* date as the backup, and they will backed up again in the next backup and so on. In a incrementel this is less problematic than in a differential, as in a differential all changes since the last fullbackup will be saved. So is this planned or still a bug? How do we avoid this problem? An ugly workarround would be to set the modify date of the files *after* backup with "RunAfterJob" (e.g) back behind the really start of the last backup. But I think that should no be the case of the user as of the program. [1] The log is really huge, but if you wish a full log, please remark this and we'll publish them. kind regards Berner Martin ------------------------------------------------------- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnk&kid7521&bid$8729&dat1642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
