[EMAIL PROTECTED] writes: > I've had similar failures--due to client problems--in the midst of multi-TB > backups, and was extremely interested to hear about what might have been a > solution to resume the backup without starting from the beginning.
I think most backup programs encounter this at one point or another. It is just worse if you are backing up a long data set. > interesting thread, in terms of making bacula more robust. I can see many > applications for the idea of resuming an interrupted backup This is a feature which would make Bacula much more robust. It may be complex to solve, but it would make Bacula significantly more helpfull. I don't know if it has changed in the latest 2 releases, but in previous releases it was not possible to restore from failed backups. So if I was backing up 10 million files and the backup finished 9 million files.. I was unable to restore from the 9 million files that had gotten backed up. I find that even worse than been able to continue a backup. > Hmmm.... I don't know anything about the internal structure of bacula, or much > about databases, but it seems to me that this is a serious weakness. I would not call it a weakness, but something which could be done better. I don't know of any other backup that has the feature. > Would it be possible to use the existing algorithms in bacula to insert the > File > records into a temporary db table--and also write File records to the > temporary I am sure the solution is likely not easy, otherwise it would have been done, but instead of giving an idea of why it is difficult or simply just saying "it is complex to solve".. Kern took the possibly least usefull route. >read the real table corresponding to the media >read the media >read the temporary table >reconcile the differences (ie., any filesets written to media, with An over-simplification of why the problem is complex. It is not easy not even to define what "resume" would mean. Would it mean tracking directories and continue where the backup failed? How about the directory it was working on? What if any indicators should exist to point out what files were done at a later date. When you do a full backup you have a number of expectations concerning the date when it was done. Allowing a backup to "continue" after failure will break many of those expectations. So the first step on getting something done would likely be to even define what one would expect bacula to do. One possible, semi-fix, may be to allow to mark failed backups to a new state which would indicate a manual change to completed. Then allow the next backup to be differential. ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
