There is a timing issue where there is a brief moment in time
that the state of a sibling entry in package cache workspace can
be set to a particular state outside of latch control, so the
entry might be reused or freed up from the lru lists in the
cache by someone else, as a result, panic may happen due to
memory corruption.
The stack can be vary:
Stack 1:
sqle_panic
sqloCrashOnCriticalMemoryValidationFailure
SQLO_MEM_POOL::diagnoseMemoryCorruptionAndCrash
sqloDiagnoseFreeBlockFailure
sqlofmblkEx
sqlra_sqlW_mem_free_sibling_list
sqlra_sqlW_mem_freespace
sqlra_sqlW_check_for_cleanup
sqlra_sqlW_get_dynamic_section_new
sqlra_load_var
sqlra_get_var
sqlrr_execimmd
sqlrr_execimmd
Stack 2:
PID : 1704972 TID : 79974 PROC :
db2sysc 0
INSTANCE: db2inst1 NODE : 000 DB :
SAMPLE
APPHDL : 0-360 APPID:
*LOCAL.db2inst1.180203112927
AUTHID : DB2INST1 HOSTNAME: MYHOST
EDUID : 79974 EDUNAME: db2agent (SAMPLE) 0
FUNCTION: DB2 UDB, SQO Latch Tracing,
SQLO_SLATCH_CAS64::getConflictComplex, probe:210
MESSAGE :
ZRC=0x870F00FD=-2029059843=SQLO_SLATCH_ERROR_HELDS_WITH_SHARED_W
AITER
"A latch with X starvation policy should StarveX latch
should not be held shared and have a shared waiter."
CALLED : OS, -, unspecified_system_function
DATA #1 : String, 13 bytes
invalid latch
DATA #2 : File name, 16 bytes
sqloLatchCAS64.C
DATA #3 : Source file line number, 8 bytes
592
DATA #4 : Codepath, 8 bytes
0
DATA #5 : String, 136 bytes
0xFAB0A00020002EA0: {
held X: 0
reserved for X: 0
shared holders: 11936
firstSharIndex: 0x200
firstExclIndex: 0x1D5850
}
DATA #6 : LatchMode, PD_TYPE_LATCH_MODE, 8 bytes
0x10000 (SQLO_LATCH_MODE_EXCLUSIVE)
DATA #7 : String, 397 bytes
{
state = 0xFAB0A00020002EA0
= {
held X: 0
reserved for X: 0
shared holders: 11936
firstSharIndex: 0x200
firstExclIndex: 0x1D5850
}
starve X mode = true
xWaitCount = 0
requestCount = 0
identity = NO_IDENTITY (0)
}
DATA #8 : Pointer, 8 bytes
0x0a00020002ede448
DATA #9 : Hexdump, 16 bytes
0x0A00020002EDE448 : FAB0 A000 2000 2EA0 0A00 0200 02ED E108
.... ...........
CALLSTCK: (Static functions may not be resolved correctly, as
they are resolved to the nearest symbol)
[0] 0x090000000DC8088C
dumpDiagInfoAndPanic__17SQLO_SLATCH_CAS64CFCPCcCUiCUlT3ClT3CiT1T
3T7 + 0x2FC
[1] 0x090000000CA0C814
getConflictComplex__17SQLO_SLATCH_CAS64FCUl + 0xE8C
[2] 0x090000000CE80D54 getConflict__17SQLO_SLATCH_CAS64FCUl +
0x118
[3] 0x090000000B8D9F2C
sqlra_sqlW_mem_freespace__FP8sqlrr_cbUlPUli + 0x1C4
[4] 0x090000000B8A561C
sqlra_sqlW_check_for_cleanup__FP8sqlrr_cbUlP7OSSTick + 0x264
[5] 0x090000000B899D7C
sqlra_sqlW_get_dynamic_section_new__FP8sqlrr_cbP16sqlra_cached_v
ar + 0x14B8
[6] 0x090000000B89ADF8
sqlra_load_var__FP8sqlrr_cbP16sqlra_cached_varP14SQLP_LOCK_INFOP
b + 0xAC
[7] 0x090000000B89AA8C sqlra_get_var__FP8sqlrr_cbiT2bPbT5 +
0x884
[8] 0x090000000C587808
sqlrr_execute__FP14db2UCinterfaceP9UCstpInfo + 0x114
[9] 0x090000000C587B10
sqlrr_execute__FP14db2UCinterfaceP9UCstpInfo + 0x41C |
