Comment 2 for bug 425115

Revision history for this message
Seppo Jaakola (seppo-jaakola) wrote :

A hand crafted debug build can catch the problem in gdb session, here is the trace:

Breakpoint 1, Diagnostics_area::sql_errno (this=0xa6b2f4c0) at sql_class.h:1180
1180 sql_class.h: No such file or directory.
 in sql_class.h
(gdb) bt
#0 Diagnostics_area::sql_errno (this=0xa6b2f4c0) at sql_class.h:1180
#1 0x08396b6c in ha_rollback_trans (thd=0xa6b2e708, all=true)
    at handler.cc:1312
#2 0x0825f288 in end_trans (thd=0xa6b2e708, completion=ROLLBACK)
    at sql_parse.cc:767
#3 0x0826f7b4 in dispatch_command (command=COM_QUERY, thd=0xa6b2e708,
    packet=0xa6b08c09 "UPDATE comm00 SET x = (x + 1) % 65537, y = ((y*x % 65537) + (2528*y % 65537) + (y*z % 65537)) % 65537, z = ((z*x % 65537) + (z*y % 65537) + (2528*z % 65537)) % 65537, r = 0.027815 WHERE p = 3", packet_length=191)
    at sql_parse.cc:1761
#4 0x082706c8 in do_command (thd=0xa6b2e708) at sql_parse.cc:947
#5 0x08257e1c in handle_one_connection (arg=0xa6b2e708) at sql_connect.cc:1155
#6 0x002fe4d2 in start_thread () from /lib/i686/nosegneg/libpthread.so.0
#7 0x0025448e in clone () from /lib/i686/nosegneg/libc.so.6
(gdb) p m_status
$1 = Diagnostics_area::DA_EMPTY
(gdb) p m_sql_errno
$2 = 1317