MySQL Bugs: #12854: RBR (ps-prot) slave core in ha_myisam::write

Bug #12854	RBR (ps-prot) slave core in ha_myisam::write_row in some tests
Submitted:	29 Aug 2005 18:27	Modified:	24 Sep 2005 10:42
Reporter:	Jonathan Miller	Email Updates:
Status:	Closed	Impact on me:	None
Category:	MySQL Server: Replication	Severity:	S1 (Critical)
Version:	5.0-wl1012	OS:	Linux (Linux)
Assigned to:	Guilhem Bichot	CPU Architecture:	Any

Description:
The following seven test produce the same core:

rpl000008.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x878b3e0, 
    buf=0x878b4c0 "ý\017") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x878b4c0, 
    buf=0x878b4c0 "ý\017") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x8770690, table=0x0, rli=0x87453ac) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x8770690, rli=0x87453ac) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d308, rli=0x87453ac)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f1c8) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0

rpl_000012.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x872c918, 
    buf=0x872c9f8 "ÿò\003") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x872c9f8, 
    buf=0x872c9f8 "ÿò\003") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x872cde8, table=0x0, rli=0x87452fc) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x872cde8, rli=0x87452fc) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d258, rli=0x87452fc)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f118) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

rpl_alter.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x872c8c8, 
    buf=0x872c9b0 "ù\001") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x872c9b0, 
    buf=0x872c9b0 "ù\001") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x8771d30, table=0x0, rli=0x87452fc) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x8771d30, rli=0x87452fc) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d258, rli=0x87452fc)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f118) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

rpl_create_database.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x4130d1e0, 
    buf=0x4130d2c0 "ý\001") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x4130d2c0, 
    buf=0x4130d2c0 "ý\001") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x4130b238, table=0x0, rli=0x874532c) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x4130b238, rli=0x874532c) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x41300590, rli=0x874532c)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f148) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

rpl_insert_select.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x872c918, 
    buf=0x872c9f8 "ÿ\001") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x872c9f8, 
    buf=0x872c9f8 "ÿ\001") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x874e7d0, table=0x0, rli=0x87452fc) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x874e7d0, rli=0x87452fc) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d258, rli=0x87452fc)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f118) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

rpl_row_charset.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x41306e50, 
    buf=0x41306f30 "ý\n") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x41306f30, 
    buf=0x41306f30 "ý\n") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x41305058, table=0x0, rli=0x87452fc) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x41305058, rli=0x87452fc) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d258, rli=0x87452fc)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f118) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

rpl_row_log.core
#0  0x40250da1 in kill () from /lib/libc.so.6
#1  0x40181f4a in pthread_kill () from /lib/libpthread.so.0
#2  0x0829602d in write_core (sig=11) at stacktrace.c:220
#3  0x0818ef2d in handle_segfault (sig=11) at mysqld.cc:2062
#4  0x401855cd in __pthread_sighandler () from /lib/libpthread.so.0
#5  <signal handler called>
#6  0x0823f508 in ha_myisam::write_row(char*) (this=0x8703730, 
    buf=0x8703810 "ý\005") at ha_myisam.cc:297
#7  0x0823c2d4 in handler::ha_write_row(char*) (this=0x8703810, 
    buf=0x8703810 "ý\005") at handler.cc:2528
#8  0x08211deb in Write_rows_log_event::do_exec_row(st_table*, st_relay_log_info*) (this=0x872d110, table=0x0, rli=0x87452fc) at log_event.cc:5615
#9  0x082111dc in Rows_log_event::exec_event(st_relay_log_info*) (
    this=0x872d110, rli=0x87452fc) at log_event.cc:5106
#10 0x0828bdd9 in exec_relay_log_event (thd=0x877d258, rli=0x87452fc)
    at slave.cc:2960
#11 0x08289af8 in handle_slave_sql (arg=0x873f118) at slave.cc:3508
#12 0x4017f6de in pthread_start_thread () from /lib/libpthread.so.0
(gdb) quit

How to repeat:
see above.

A similar problem seems to happen when master issues INSERT DELAYED.

A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/internals/29844

Turns out that the crash in INSERT DELAYED is a different problem, now filed as BUG#13358. Correcting synopsis, so.

fixed, nothing to document.