Kern Sibbald wrote:
> The bscan problem that I found caused it to generate a JobMedia record in the
> database that had an end FileIndex one less than it should have been. This
> was the last record on a Volume, and the record was continued on the next
> Volume. When Bacula constructed a bsr, the "optimization" code had this one
> off problem, so when the restore job ran, the last record (partial) record on
> the first tape was ignored. When the restore job got the second tape up,
> after reading the first (partial) record, it realized that the first part of
> the record from the first Volume was not there, so my insanity check code
> aborted.
>
> What surprises me is that this never triggered before in all the years I ran
> it. I wish I had more time to devote to regression testing as I would
> develop a case that is 100% sure to exercise this problem ...
I recall finding (and fixing) a very similar bug in IBM's backup.exe
program that shipped with DOS 3.20. I spent about five hours on the
phone with a frantic New York stockbroker figuring out the problem and
walking him through patching his backup disks so that the backup would
restore properly.
--
Phil Stracchino [EMAIL PROTECTED]
Renaissance Man, Unix generalist, Perl hacker
Mobile: 603-216-7037 Landline: 603-886-3518
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
