There is possibility of a trap when trying to unlock a latch
that has been already unlocked.
The top function in the stack should show at least
'sqlrlc_auth_find_insert()' and 'sqle_panic()'.
Here is a stack example:
0x090000000055BED4 pthread_kill + 0xD4
0x090000000055B764 _p_raise + 0x44
0x09000000000399E8 raise + 0x48
0x0900000000054F64 abort + 0xC4
0x090000000B21E950 sqloExitEDU + 0x320
0x09000000083DA30C sqle_panic + 0x4DC
0x090000000AAE6414 sqloSpinLockReleaseConflict + 0x2AC
0x090000000A9A46CC sqloSpinLockReleaseConflict + 0x154
0x09000000098502D8 sqlrlc_auth_find_insert + 0x17A8
0x090000000DE0F810 sqlrlc_sysuserauth_get_authPDs + 0x134
0x090000000DE7F69C sqlrlgua + 0x6C4
From db2diag.log, you should find error message(s) as below:
2014-10-14-11.20.04.373998+480 I414087385A1895 LEVEL:
Severe
PID : 4980906 TID : 20820 PROC :
db2sysc 0
INSTANCE: db2inst1 NODE : 000 DB :
SAMPLE
APPHDL : 0-13124 APPID:
10.31.84.61.50654.141014030223
AUTHID : db2inst1 HOSTNAME: HOSTNAME1
EDUID : 20820 EDUNAME: db2agent (SAMPLE) 0
FUNCTION: DB2 UDB, SQO Latch Tracing,
sqlo_xlatch::releaseConflict, probe:10
DATA #1 : String, 27 bytes
unlocking an unlatched lock
DATA #2 : Pointer, 8 bytes
0x0a000200b1ecfcd0
DATA #3 : String, 109 bytes
{
lock = { 0x00000000 [ unlocked ] }
identity = sqlrlc_entry_common::usage_latch (575)
}
DATA #4 : Hexdump, 8 bytes
0x0A000200B1ECFCD0 : 0000 0000 023F 0000
.....?..
CALLSTCK: (Static functions may not be resolved correctly, as
they are resolved to the nearest symbol)
[0] 0x090000000AAE640C sqloSpinLockReleaseConflict + 0x2A4
[1] 0x090000000A9A46CC sqloSpinLockReleaseConflict + 0x154
[2] 0x09000000098502D8 sqlrlc_auth_find_insert..
[3] 0x090000000DE0F810 sqlrlc_sysuserauth_get_authPD..
[4] 0x090000000DE7F69C sqlrlgua__FP8...
[5] 0x090000000DE0F138 sqlrlgua__FP8...
[6] 0x090000000C3D1B64 sqlra_check_each_privs__FP8...
[7] 0x090000000CC54CE4 sqlra_check_privs__FP8...
[8] 0x090000000B1DA7A4 sqlra_check_stmt_privs__FP8... |
