There is a possible timing issue with two agents that are
simultaneously disconnecting from a database that may lead to a
crash as a result of a double free. The db2diag.log will show:
2018-07-30-12.15.08.917524+480 I7100047A495 LEVEL:Severe
PID : 1245 TID : 2212 PROC
:db2sysc 0
INSTANCE: db2inst1 NODE : 000 DB :SAMPLE
APPHDL : 0-587
APPID:*LOCAL.db2inst1.180730115005
AUTHID : db2inst1 HOSTNAME: myhost
EDUID : 2212 EDUNAME: db2agent (SAMPLE) 0
FUNCTION: DB2 UDB, SQO Memory
Management,sqloDiagnoseFreeBlockFailure, probe:30
DATA #1 : String, 12 bytes
Double free.
With the reported stack trace being similar to:
sqle_panic(int) + 0x8a0
sqloCrashOnCriticalMemoryValidationFailure + 0x4c
sqlofmblkEx.fdpr.chunk.25 + 0x1c
sqm_object::operator delete(void*) + 0x28
sqm_agent_entity::database_gone.fdpr.chunk.1(const
sqeLocalDatabase*&) + 0x74
sqm_agent_entity::database_gone(const sqeLocalDatabase*&) +
0x134
glue_pointer_func@glue8 + 0x8c
sqeLocalDatabase::TermDbConnect.fdpr.chunk.65(sqeAgent*,sqlca*,i
nt) + 0x27c
sqeLocalDatabase::TermDbConnect.fdpr.chunk.63(sqeAgent*,sqlca*,i
nt) + 0x4f4
sqeLocalDatabase::TermDbConnect.fdpr.chunk.27(sqeAgent*,sqlca*,i
nt) + 0x240
sqeLocalDatabase::TermDbConnect.fdpr.chunk.26(sqeAgent*,sqlca*,i
nt) + 0x76c
sqeLocalDatabase::TermDbConnect.fdpr.chunk.100(sqeAgent*,sqlca*,
int) + 0x99c
sqlmon_conn::cleanup_sqlmon_conn.fdpr.chunk.17() + 0xe30
sqlesrspWrp(db2UCinterface*) + 0xbc
sqleUCagentConnectReset + 0xd8 |
