On 13/02/2007 Steve Langasek wrote: > On Tue, Feb 13, 2007 at 06:46:22PM +0100, Jonas Meurer wrote: > > I already did this 'stack trace': > > > [ copied the 0x... hex digits to mysqld.stack] > > # cp /usr/share/doc/mysql-server-5.0/mysqld.sym.gz > > # gzip -d mysqld.sym.gz > > # resolve_stack_dump -s mysqld.sym -n mysqld.stack > > 0x81c0649 handle_segfault + 681 > > 0xb7cfb947 _end + -1352585609 > > 0xb7cfd0c9 _end + -1352579591 > > 0xb7d30fda _end + -1352366838 > > 0xb7d3889f _end + -1352335921 > > 0xb7d38942 _end + -1352335758 > > 0x8484df3 free_root + 67 > > 0x81dbc5d _Z16dispatch_command19enum_server_commandP3THDPcj + 509 > > 0x81dd188 _Z10do_commandP3THD + 136 > > 0x81ddb94 handle_one_connection + 2308 > > 0xb7f640bd _end + -1350060563 > > 0xb7d9e93e _end + -1351917970 > > Could you try running mysqld under valgrind? It's likely in this scenario > that the bug lies far away from the point where glibc detects it; valgrind > is usually the best way to find out where that is.
The bug appeared on a production machine, so debugging is somewhat difficult. One admin considered that the mysql server crash was produced by a particular zope application, but we were not able to reproduce it yet. The crash didn't appear in the last three days, by the way. So if you like, we could lower severity of the bug to important and tag it as unreproducible for now. ... jonas -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
