quering global_temporary_tables causes server crash

[Aurimas Mikalauskas]

I've been running this very intensively to monitor temporary tables:

while [ 1 ]; do mysql information_schema -e "select * from global_temporary_tables"; done

However after running that for some time server crashed (and then again, and again). This is a production systems so I can't use it to work further on figuring out why this happenins, but I have a backtrace in logs which will hopefully help to diagnose it:

110329 16:14:03 - mysqld got signal 11 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help diagnose
the problem, but since we have already crashed, something is definitely wrong
and this may fail.

key_buffer_size=17179869184
read_buffer_size=131072
max_used_connections=107
max_threads=1000
threads_connected=102
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 18964067 K
bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x2ab178129c00
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0x528520e8 thread_stack 0x30000
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x874a3e]
/usr/sbin/mysqld(handle_segfault+0x351)[0x5b3161]
/lib64/libpthread.so.0[0x3b3480eb10]
/usr/sbin/mysqld(mi_status+0x26)[0x82dfe6]
/usr/sbin/mysqld(_ZN9ha_myisam4infoEj+0x2a)[0x846dfa]
/usr/sbin/mysqld[0x6b63ea]
/usr/sbin/mysqld[0x6b65f6]
/usr/sbin/mysqld(_Z24get_schema_tables_resultP4JOIN23enum_schema_table_state+0x1d0)[0x6b5d10]
/usr/sbin/mysqld(_ZN4JOIN4execEv+0x439)[0x62e4a9]
/usr/sbin/mysqld(_Z12mysql_selectP3THDPPP4ItemP10TABLE_LISTjR4ListIS1_ES2_jP8st_orderSB_S2_SB_yP13select_resultP18st_select_lex_unitP13st_select_lex+0x1b9)[0x630479]
/usr/sbin/mysqld(_Z13handle_selectP3THDP6st_lexP13select_resultm+0x1d5)[0x630df5]
/usr/sbin/mysqld[0x5becb1]
/usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x3e9)[0x5c1bf9]
/usr/sbin/mysqld(_Z11mysql_parseP3THDPcjPPKc+0x549)[0x5c7df9]
/usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0xa29)[0x5c8829]
/usr/sbin/mysqld(_Z10do_commandP3THD+0x120)[0x5c94c0]
/usr/sbin/mysqld(handle_one_connection+0x7cc)[0x5bb3bc]
/lib64/libpthread.so.0[0x3b3480673d]
/lib64/libc.so.6(clone+0x6d)[0x3b340d3f6d]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0x1a5b00c0): select * from global_temporary_tables
Connection ID (thread ID): 26285
Status: NOT_KILLED

Version: '5.1.55-rel12.6-log' socket: '/var/lib/mysql/mysql.sock' port: 3306 Percona Server (GPL), 12.6, Revision 200

[Aurimas Mikalauskas (aurimas-mikalauskas)]

Just to explain how severe it is, when I was running this, server crashed 3 times within half an hour. After I stopped the loop server never crashed ever since so this is definitely related.

[Alexey Kopytov (akopytov)]

I can easily reproduce it like this:

shell1> while true; do client/mysql -e "create temporary table t1(a int); create temporary table t2(a int); create temporary table t3(a int)" test; done

shell2> while true; do client/mysql -e "select * from global_temporary_tables" information_schema; done

Crashes in a few seconds.

[Stewart Smith (stewart)]

../client/mysqlslap --query "create temporary table t1 (a int); select * from information_schema.global_temporary_tables; drop table t1;" --create-schema test --number-of-queries=10000000 --concurrency=100 -S /tmp/rlZDDT6T4g/mysqld.1.sock

will reproduce.

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-331