mysql got signal 11

[Aaron C. de Bruyn]

Binary package hint: mysql-server-5.0

A client is trying to transfer a database dump from an old server to a new server. The dump was created in phpmyadmin and imported through phpmyadmin on the new server.

The old server is Ubuntu 6.06 running mysql-server-5.0 version: 5.0.22-0ubuntu6.06.10.
The new server is Ubuntu 8.04 running mysql-server-5.0 version: 5.0.45-1ubuntu3.3.

When the client tries to import the database, it gets partially created and then mysqld dies with this in syslog:

May 26 14:20:39 tiefighter mysqld[20598]: 080526 14:20:39 - mysqld got signal 11;
May 26 14:20:39 tiefighter mysqld[20598]: This could be because you hit a bug. It is also possible that this binary
May 26 14:20:39 tiefighter mysqld[20598]: or one of the libraries it was linked against is corrupt, improperly built,
May 26 14:20:39 tiefighter mysqld[20598]: or misconfigured. This error can also be caused by malfunctioning hardware.
May 26 14:20:39 tiefighter mysqld[20598]: We will try our best to scrape up some info that will hopefully help diagnose
May 26 14:20:39 tiefighter mysqld[20598]: the problem, but since we have already crashed, something is definitely wrong
May 26 14:20:39 tiefighter mysqld[20598]: and this may fail.
May 26 14:20:39 tiefighter mysqld[20598]:
May 26 14:20:39 tiefighter mysqld[20598]: key_buffer_size=16777216
May 26 14:20:39 tiefighter mysqld[20598]: read_buffer_size=131072
May 26 14:20:39 tiefighter mysqld[20598]: max_used_connections=13
May 26 14:20:39 tiefighter mysqld[20598]: max_connections=100
May 26 14:20:39 tiefighter mysqld[20598]: threads_connected=9
May 26 14:20:39 tiefighter mysqld[20598]: It is possible that mysqld could use up to
May 26 14:20:39 tiefighter mysqld[20598]: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_connections = 233983 K
May 26 14:20:39 tiefighter mysqld[20598]: bytes of memory
May 26 14:20:39 tiefighter mysqld[20598]: Hope that's ok; if not, decrease some variables in the equation.
May 26 14:20:39 tiefighter mysqld[20598]:
May 26 14:20:39 tiefighter mysqld[20598]: thd=0xb0b96a70
May 26 14:20:39 tiefighter mysqld[20598]: Attempting backtrace. You can use the following information to find out
May 26 14:20:39 tiefighter mysqld[20598]: where mysqld died. If you see no messages after this, something went
May 26 14:20:39 tiefighter mysqld[20598]: terribly wrong...
May 26 14:20:39 tiefighter mysqld[20598]: Cannot determine thread, fp=0xb0e13a48, backtrace may not be correct.
May 26 14:20:39 tiefighter mysqld[20598]: Stack range sanity check OK, backtrace follows:
May 26 14:20:39 tiefighter mysqld[20598]: 0x81f04a0
May 26 14:20:39 tiefighter mysqld[20598]: 0xb7eb2550
May 26 14:20:39 tiefighter mysqld[20598]: 0x825a1d0
May 26 14:20:39 tiefighter mysqld[20598]: 0x825f275
May 26 14:20:39 tiefighter mysqld[20598]: 0x8209a03
May 26 14:20:39 tiefighter mysqld[20598]: 0x820d3a0
May 26 14:20:39 tiefighter mysqld[20598]: 0x820d98b
May 26 14:20:39 tiefighter mysqld[20598]: 0x820ed97
May 26 14:20:39 tiefighter mysqld[20598]: 0x820f940
May 26 14:20:39 tiefighter mysqld[20598]: 0xb7eb046b
May 26 14:20:39 tiefighter mysqld[20598]: 0xb7cca6de
May 26 14:20:39 tiefighter mysqld[20598]: New value of fp=(nil) failed sanity check, terminating stack trace!
May 26 14:20:39 tiefighter mysqld[20598]: Please read http://dev.mysql.com/doc/mysql/en/using-stack-trace.html and follow instructions on how to resolve the stack trace. Resolved
May 26 14:20:39 tiefighter mysqld[20598]: stack trace is much more helpful in diagnosing the problem, so please do
May 26 14:20:39 tiefighter mysqld[20598]: resolve it
May 26 14:20:39 tiefighter mysqld[20598]: Trying to get some variables.
May 26 14:20:39 tiefighter mysqld[20598]: Some pointers may be invalid and cause the dump to abort...
May 26 14:20:39 tiefighter mysqld[20598]: thd->query at 0xb0bea9d0 is invalid pointer
May 26 14:20:39 tiefighter mysqld[20598]: thd->thread_id=16069
May 26 14:20:39 tiefighter mysqld[20598]: The manual page at http://www.mysql.com/doc/en/Crashing.html contains
May 26 14:20:39 tiefighter mysqld[20598]: information that should help you find out what is causing the crash.

I followed http://dev.mysql.com/doc/mysql/en/using-stack-trace.html and came up with the following:

root@tiefighter:~# resolve_stack_dump -s ./mysqld.sym -n mysqld.stack
0x81f04a0 handle_segfault + 768
0xb7eb2550 _end + -1351137256
0x825a1d0 select_insert::select_insert(st_table_list*, st_table*, List<Item>*, List<Item>*, List<Item>*, enum_duplicates, bool) + 256
0x825f275 mysql_insert(THD*, st_table_list*, List<Item>&, List<List<Item> >&, List<Item>&, List<Item>&, enum_duplicates, bool) + 1285
0x8209a03 mysql_execute_command(THD*) + 14947
0x820d3a0 mysql_parse(THD*, char const*, unsigned int, char const**) + 368
0x820d98b dispatch_command(enum_server_command, THD*, char*, unsigned int) + 1259
0x820ed97 do_command(THD*) + 167
0x820f940 handle_one_connection + 2656
0xb7eb046b _end + -1351145677
0xb7cca6de _end + -1353135706

This only happens when he turns on delayed inserts in phpmyadmin and has been duplicated 3 times by him on my server at my request.

If I try to install mysql-server-5.0-dbgsym I get the following:
The following packages have unmet dependencies:
  mysql-server-5.0-dbgsym: Depends: mysql-server-5.0 (= 5.0.45-1ubuntu3) but 5.0.45-1ubuntu3.3 is to be installed
E: Broken packages

Is this because the latest mysql package was pushed out and debug symbols are not ready yet?

Unfortunately due to confidentiality requirements I am unable to post the content of the .sql file that causes the crash.

[Chuck Short (zulcss)]

If possible can you provide us a way to reproduce this bug?

Thanks
chuck

[Aaron C. de Bruyn (darkpixel2k)]

Unfortunately I was not able to reproduce the problem with the data stripped out or anonymized, and they don't want to release the contents of the SQL file. We have worked around the issue and they are fine with that resolution.

I wish I could be more help.

I'll mark this bug as invalid.