Hello,
we have spent two weeks of testing until we found the root of the
problem (concurrent jobs) and we were 100% sure it is a bug. However
after I submitted it as a bug, it was closed as they couldn't
replicate it, claiming we have a hardware problem (which we proved is
not the case) and required us to send all of our hadrware (which I
guess was just a sarcasm or something).
At least we found a working solution (no concurrent jobs, because with
concurent jobs bacula was useless) hoping they will fix it sometime
when they receive enough proof that there IS a bug. You can reopen it
(as I'm not going to do it after I've got several times a response
"can't replicate, so there are no bugs") at bugs.bacula.org
Regards.
Saturday, September 8, 2007, 2:26:51 AM:
ST> On Fri, 7 Sep 2007, Doytchin Spiridonov wrote:
>> just to mention again that while there were suggestions this is a
>> hardware problem we did a lot of tests and proved the problem is not
>> hadrware but there is a bug (which however was closed with "unable to
>> reproduce" reason)
>>
>> As you are the next case, can you please test it again ("have deleted
>> all the backups, reinitialized bacula from scratch, run full backups
>> to different disk volumes, and tried a restore again") but w/o
>> enabling concurrent jobs (I bet you are) and see if it will be OK or
>> you will get the same problems?
ST> Very interesting.
ST> My entire set of backup volumes, currently about 4.2 TB, was rsync'd to a
ST> separate system (no errors) and a restore attempt made there from a fresh
ST> bacula installation (using the same catalog but otherwise different h/w
ST> and s/w). This restore attempt also failed in the identical fashion to
ST> that described previously. The backup volumes are evidently corrupted at
ST> backup time.
ST> On the original director, I then set "Max Concurrent Jobs" to 1 everywhere
ST> and ran a full backup of a 300 GB file system containing 151,155 files.
ST> Same hardware, same software, everything the same except for MCJ=1. I then
ST> restored all of these files without any error, which I have been unable to
ST> do with MCJ=2.
ST> The conclusion is that it looks like there may indeed be a bug.
ST> In any event, I shall run with MCJ=1 for now, and re-run full backups of
ST> all of my data (about 2 TB on this system), and then restore the whole lot
ST> to see what I get. If I get time I will take a peek at the source.
ST> Steve
ST> ----------------------------------------------------------------------------
ST> Steve Thompson E-mail: smt AT vgersoft DOT com
ST> Voyager Software LLC Web: http://www DOT vgersoft DOT com
ST> 39 Smugglers Path VSW Support: support AT vgersoft DOT com
ST> Ithaca, NY 14850
ST> "186,300 miles per second: it's not just a good idea, it's the law"
ST> ----------------------------------------------------------------------------
ST> -------------------------------------------------------------------------
ST> This SF.net email is sponsored by: Microsoft
ST> Defy all challenges. Microsoft(R) Visual Studio 2005.
ST> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
ST> _______________________________________________
ST> Bacula-users mailing list
ST> Bacula-users@lists.sourceforge.net
ST> https://lists.sourceforge.net/lists/listinfo/bacula-users
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
