Bug #37859 system test stopped with a crash of mysqld.
Submitted: 4 Jul 2008 8:39 Modified: 11 Aug 2008 7:29
Reporter: Horst Hunger Email Updates:
Status: Duplicate Impact on me:
None 
Category:MySQL Server Severity:S1 (Critical)
Version: Ver 6.0.6-alpha-debug OS:Linux (suse-linux on x86_64, 2.6.16.46-0.12-smp)
Assigned to: CPU Architecture:Any

[4 Jul 2008 8:39] Horst Hunger
Description:
The system test crashed with the following info:

Core was generated by `/data0/bzr/mysql-6.0-system-test/sql/mysqld --no-defaults --basedir=/data0/bzr/'.
Program terminated with signal 11, Segmentation fault.
#0  0x00002b5774271ea3 in pthread_kill () from /lib64/libpthread.so.0
(gdb) bt
#0  0x00002b5774271ea3 in pthread_kill () from /lib64/libpthread.so.0
#1  0x0000000000d35d91 in my_write_core (sig=11) at stacktrace.c:307
#2  0x00000000006f74fb in handle_segfault (sig=11) at mysqld.cc:2638
#3  <signal handler called>
#4  0x0000000000d3a7dc in safe_mutex_lock (mp=0xa5a5a5a5a5a5a5bd, try_lock=0 '\0',
    file=0xfa1a6f "thr_lock.c", line=525) at thr_mutex.c:99
#5  0x0000000000d385e2 in thr_lock (data=0x2aaab06bc100, owner=0x4a42280,
    lock_type=-1515870811) at thr_lock.c:525
#6  0x0000000000d398b2 in thr_multi_lock (data=0x2aaaafc83238, count=1,
    owner=0x4a42280) at thr_lock.c:990
#7  0x00000000006edb09 in mysql_lock_tables (thd=0x4a41518, tables=0x4a607f48,
    count=1, flags=27, need_reopen=0x4a607ddb) at lock.cc:306
#8  0x0000000000755a75 in open_ltable (thd=0x4a41518, table_list=0x4a607ea0,
    lock_type=TL_WRITE_CONCURRENT_INSERT, lock_flags=27) at sql_base.cc:4042
#9  0x0000000000755b72 in open_performance_schema_table (thd=0x4a41518,
    one_table=0x4a607ea0, backup=0x4a608420) at sql_base.cc:7891
#10 0x00000000007cbcaf in Log_to_csv_event_handler::log_general (this=0x1e00be0,
    thd=0x4a41518, event_time=1215041201,
    user_host=0x4a608570 "systuser[systuser] @ localhost []", user_host_len=33,
    thread_id=689, command_type=0xe7240d "Query", command_type_len=5,
    sql_text=0x4499a00 "EXECUTE sel_single USING @next_val", sql_text_len=34,
    client_cs=0x134dee0) at log.cc:391
#11 0x00000000007ca672 in LOGGER::general_log_write (this=0x135ff40,
    thd=0x4a41518, command=COM_QUERY,
    query=0x4499a00 "EXECUTE sel_single USING @next_val", query_length=34)
    at log.cc:1046
#12 0x00000000007ca722 in general_log_write (thd=0x4a41518, command=COM_QUERY,
    query=0x4499a00 "EXECUTE sel_single USING @next_val", query_length=34)
    at log.cc:4091
#13 0x00000000007120fa in dispatch_command (command=COM_QUERY, thd=0x4a41518,
    packet=0x4378809 "EXECUTE sel_single USING @next_val", packet_length=34)
    at sql_parse.cc:1042
#14 0x0000000000713686 in do_command (thd=0x4a41518) at sql_parse.cc:724
#15 0x000000000070022b in handle_one_connection (arg=0x4a41518)
    at sql_connect.cc:1153
#16 0x00002b577426d143 in start_thread () from /lib64/libpthread.so.0
#17 0x00002b577489f74d in clone () from /lib64/libc.so.6
#18 0x0000000000000000 in ?? ()
(gdb)   

More info can be found on the machine "dl360-g5-d" in "/data0/bzr/mysql-6.0-system-test". The var directory is available as "system-crash.tar.gz".

How to repeat:
The system test executes sql in an arbitrary order and run about 1 hour. So, reproduction is difficult. The analysis of general_log.CSV or master.log may help to find the sequence of sql to reproduce the crash.
[11 Aug 2008 7:29] Sveta Smirnova
Closed as duplicate of bug #38099, because they are same and later contains dixcussion.