the server was crashed while some queries were executed concurrently

Bug #984605 reported by yinfeng
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
MySQL Server
Unknown
Unknown
Percona Server moved to https://jira.percona.com/projects/PS
Triaged
High
Unassigned

Bug Description

i had run the test case(http://bugs.mysql.com/file.php?id=17153) and soon server was crashed nearly 30 mins later

server version: Percona-server 5.5.18

here is part of error log:

12026 120418 14:33:51 - mysqld got signal 11 ;
12027 This could be because you hit a bug. It is also possible that this binary
12028 or one of the libraries it was linked against is corrupt, improperly built,
12029 or misconfigured. This error can also be caused by malfunctioning hardware.
12030 We will try our best to scrape up some info that will hopefully help diagnose
12031 the problem, but since we have already crashed, something is definitely wrong
12032 and this may fail.
12033
12034 key_buffer_size=8388608
12035 read_buffer_size=131072
12036 max_used_connections=17
12037 max_threads=1000
12038 thread_count=12
12039 connection_count=12
12040 It is possible that mysqld could use up to
12041 key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 180418 K
12042 bytes of memory
12043 Hope that's ok; if not, decrease some variables in the equation.
12044
12045 Thread pointer: 0x7f806008cb70
12046 Attempting backtrace. You can use the following information to find out
12047 where mysqld died. If you see no messages after this, something went
12048 terribly wrong...
12049 stack_bottom = 0x412390f8 thread_stack 0x80000
12050 /u01/ps.orig/bin/mysqld(my_print_stacktrace+0x39)[0x7caaa9]
12051 /u01/ps.orig/bin/mysqld(handle_segfault+0x36d)[0x4ebe7d]
12052 /lib64/libpthread.so.0[0x3e6380e7c0]
12053 /lib64/libc.so.6(memcpy+0x15b)[0x3e6307bf0b]
12054 /u01/ps.orig/bin/mysqld(_mi_rec_pack+0x2bc)[0x93b2bc]
12055 /u01/ps.orig/bin/mysqld(_mi_write_blob_record+0x9d)[0x93e0cd]
12056 /u01/ps.orig/bin/mysqld(mi_write+0x86f)[0x95702f]
12057 /u01/ps.orig/bin/mysqld(_ZN9ha_myisam9write_rowEPh+0x43)[0x9259d3]
12058 /u01/ps.orig/bin/mysqld(_ZN7handler12ha_write_rowEPh+0x5f)[0x689cff]
12059 /u01/ps.orig/bin/mysqld(_Z23fill_schema_processlistP3THDP10TABLE_LISTP4Item+0x44d)[0x5c019d]
12060 /u01/ps.orig/bin/mysqld(_Z24get_schema_tables_resultP4JOIN23enum_schema_table_state+0x211)[0x5ac931]
12061 /u01/ps.orig/bin/mysqld(_ZN4JOIN4execEv+0x406)[0x5a7ba6]
12062 /u01/ps.orig/bin/mysqld(_Z12mysql_selectP3THDPPP4ItemP10TABLE_LISTjR4ListIS1_ES2_jP8st_orderSB_S2_SB_yP13select_resultP18st_select_lex_unitP13st_select_lex+0x192 )[0x5a9ee2]
12063 /u01/ps.orig/bin/mysqld(_Z13handle_selectP3THDP3LEXP13select_resultm+0x1cb)[0x5aa90b]
12064 /u01/ps.orig/bin/mysqld[0x566510]
12065 /u01/ps.orig/bin/mysqld(_Z21mysql_execute_commandP3THD+0x17b0)[0x5692d0]
12066 /u01/ps.orig/bin/mysqld(_Z11mysql_parseP3THDPcjP12Parser_state+0x3ad)[0x56e7dd]
12067 /u01/ps.orig/bin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x13ca)[0x56fbba]
12068 /u01/ps.orig/bin/mysqld(_Z10do_commandP3THD+0x106)[0x56ff16]
12069 /u01/ps.orig/bin/mysqld(_Z24do_handle_one_connectionP3THD+0x125)[0x60ca65]
12070 /u01/ps.orig/bin/mysqld(handle_one_connection+0x4c)[0x60cb8c]
12071 /lib64/libpthread.so.0[0x3e638064a7]
12072 /lib64/libc.so.6(clone+0x6d)[0x3e630d3c2d]
12073
12074 Trying to get some variables.
12075 Some pointers may be invalid and cause the dump to abort.
12076 Query (0x7f80600498c0): select id from information_schema.processlist where id<>connection_id() and info like 'select%' order by rand() limit 1 into @a
12077 Connection ID (thread ID): 300588
12078 Status: NOT_KILLED
12079
12080 The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
12081 information that should help you find out what is causing the crash.

Changed in percona-server:
status: New → Confirmed
Revision history for this message
Laurynas Biveinis (laurynas-biveinis) wrote :

The linked upstream bug deals with InnoDB internal deadlocks, is that the correct bug linked?

Revision history for this message
Alexey Kopytov (akopytov) wrote :

Yes, according to the discussion in the upstream bug, the same test case leads to either a deadlock in 5.1.x or a crash in 5.5.18.

Revision history for this message
yinfeng (yinfeng-zwx) wrote :
Download full text (4.0 KiB)

here is full stack

(gdb) bt
#0 0x0000003e6307bf0b in memcpy () from /lib64/libc.so.6
#1 0x000000000093ae0c in _mi_rec_pack (info=0x7fff90011f60, to=0x7fff03fff028 "", from=0x7fff90015c4f "")
    at /tmp/Percona-Server-5.5.18-rel23.0/storage/myisam/mi_dynrec.c:997
#2 0x000000000093dc1d in _mi_write_blob_record (info=0x7fff90011f60, record=0x7fff900159a0 "\370\v\252\002")
    at /tmp/Percona-Server-5.5.18-rel23.0/storage/myisam/mi_dynrec.c:289
#3 0x0000000000956b7f in mi_write (info=0x7fff90011f60, record=0x7fff900159a0 "\370\v\252\002")
    at /tmp/Percona-Server-5.5.18-rel23.0/storage/myisam/mi_write.c:142
#4 0x0000000000925523 in ha_myisam::write_row (this=0x7fff90011940, buf=0x7fff900159a0 "\370\v\252\002")
    at /tmp/Percona-Server-5.5.18-rel23.0/storage/myisam/ha_myisam.cc:789
#5 0x0000000000689a0f in handler::ha_write_row (this=0x7fff90011940, buf=0x7fff900159a0 "\370\v\252\002")
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/handler.cc:4980
#6 0x00000000005bfedd in schema_table_store_record (table=<optimized out>, thd=<optimized out>)
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_show.cc:2756
#7 fill_schema_processlist (thd=0x7fff9402d870, tables=<optimized out>, cond=<optimized out>)
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_show.cc:2056
#8 0x00000000005ac671 in get_schema_tables_result (join=0x7fff900196e8, executed_place=PROCESSED_BY_JOIN_EXEC)
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_show.cc:7420
#9 0x00000000005a78e6 in JOIN::exec (this=0x7fff900196e8) at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_select.cc:1926
#10 0x00000000005a9c22 in mysql_select (thd=0x7fff9402d870, rref_pointer_array=0x7fff9402fce8, tables=0x7fff8c03e808, wild_num=0,
    fields=<optimized out>, conds=0x7fff8c03f470, og_num=1, order=0x7fff8c03f668, group=0x0, having=0x0, proc_param=0x0, select_options=2684619520,
    result=0x7fff8c03f730, unit=0x7fff9402f4f8, select_lex=0x7fff9402fb18) at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_select.cc:2600
#11 0x00000000005aa64b in handle_select (thd=0x7fff9402d870, lex=0x7fff9402f438, result=0x7fff8c03f730, setup_tables_done_option=0)
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_select.cc:312
#12 0x00000000005661d0 in execute_sqlcom_select (thd=0x7fff9402d870, all_tables=0x7fff8c03e808)
    at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_parse.cc:4735
#13 0x0000000000568f90 in mysql_execute_command (thd=0x7fff9402d870) at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_parse.cc:2279
#14 0x000000000056e49d in mysql_parse (thd=0x7fff9402d870,
    rawbuf=0x7fff8c03e4c0 "select id from information_schema.processlist where id<>connection_id() and info like 'select%' order by rand() limit 1 into @a", length=<optimized out>, parser_state=0x4a920db0) at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_parse.cc:5805
#15 0x000000000056f87a in dispatch_command (command=COM_QUERY, thd=0x7fff9402d870,
    packet=0x7fff94064e51 "select id from information_schema.processlist where id<>connection_id() and info like 'select%' order by rand() limit 1 into @a;", packet_length=<optimized out>) at /tmp/Percona-Server-5.5.18-rel23.0/sql/sql_parse.cc:1060
#16 0x000000000056fbd6 in do_command (thd=0x7fff9402d870)...

Read more...

Stewart Smith (stewart)
Changed in percona-server:
status: Confirmed → Triaged
importance: Undecided → High
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.