Hi, could you please try db5.1 (and optionally db5.2 in experimental)?
O. On Wed, Dec 14, 2011 at 12:55, Stephan Sürken <abs...@debian.org> wrote: > Source: db > Severity: normal > Tags: upstream patch > > Dear Maintainer, > > when you happen to have a (BTREE) Berkeley DB (w/o environment) > broken in a certain way [1][2], trying a salvage dump like so > > db4.8_dump -r -f database.dump database.db > > leads to an endless loop inside libdb in all Berkeley DB > versions >= 4.7. > > This is due to the attached upstream patch [3]; reversing this > patch "fixes" the issue, and it works again (like in Berkeley DB > 4.4). > > I am not at all into Berkeley DB internals, however I guess the > introduced loop misses a guard against "pageno"s > loop-referencing (same two pagenos usually reappear enlessly > when debugging). > > Reverting the patch definitely fixes a regression for my use > case. Don't know what it might break, though. > > Hope you can forward this upstream, and/or maybe considering a > Debian patch for this. > > Thanks! > > Stephan > > -- > [1] The attached database.db is such an example. > [2] The attached tar.gz has a program to create "suchly broken" > dbs -- basically an unclean shutdown after a number of put() and > del() does the trick. To use it, try > ? make > ? make find-broken > [3] The attached patch shows the loop introduced in 4.7 upstream. > > -- System Information: > Debian Release: wheezy/sid > APT prefers unstable > APT policy: (500, 'unstable') > Architecture: amd64 (x86_64) > > Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores) > Locale: LANG=en_GB.utf8, LC_CTYPE=en_GB.utf8 (charmap=UTF-8) > Shell: /bin/sh linked to /bin/dash > > _______________________________________________ > pkg-db-devel mailing list > pkg-db-de...@lists.alioth.debian.org > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-db-devel -- Ondřej Surý <ond...@sury.org> -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
