|
Problem IT00923
|
Status: Closed |
PD_GET_DIAG_HIST or db2diag may hang and consume 100% CPU if parsing db2diag.log record containing event string field |
| product: |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 |
| Problem description: |
Execution of PD_GET_DIAG_HIST or db2diag may hang when
db2diag.log containing a record with event string field.
Such as following record with STOP event, which contains
multiple lines and EVENT key word:
2014-03-03-11.00.21.759236+060 E147178466E643 LEVEL: Event
PID : 8604 TID : 140272021272320PROC :
db2acd
INSTANCE: db2dec NODE : 000
FUNCTION: DB2 UDB, Administrative Task Scheduler, AtsTask::run,
probe:2100
STOP : End execution - Task ID: 14 Status: COMPLETE
Task Name : SAPTOOLS.DBH_LCKEVENTS_COLLECT
Database : DEC
Invocation : 2712
Begin Time : 2014-03-03 10:00:00.000000000 (UTC)
SQLCODE : 0
SQLSTATE :
RC : 0
Next Execution : 2014-03-03 11:00:00.000000000 (UTC)
External monitoring tools like top will show a db2fmp process
which consumes 100% of a CPU.
With db2pd you will get the ID of the routine being executed by
the db2fmp process:
# db2pd -fmpexechistory
FMP Process:
FmpPid Bit Flags ActiveThrd PooledThrd ForcedThrd
Active
14695 64 0x00000002 1 2 0
Yes
Active Threads:
EduPid : 17 ThreadId : 1431500544
Routine ID Timestamp
66049 2014-04-08-12.18.10.743262
Using the routine ID from db2pd you will get the name of the
routine by querying the system catalog:
db2 "select varchar(ROUTINESCHEMA,10) ROUTINESCHEMA,
varchar(ROUTINENAME,25) ROUTINENAME, ROUTINEID from
syscat.routines where ROUTINEID=66049"
ROUTINESCHEMA ROUTINENAME ROUTINEID
------------- ------------------------- -----------
SYSPROC PD_GET_DIAG_HIST 66049 |
| Problem Summary: |
****************************************************************
* USERS AFFECTED: *
* All *
****************************************************************
* PROBLEM DESCRIPTION: *
* See Error Description *
****************************************************************
* RECOMMENDATION: *
* Upgrade to DB2 version 9.7 FixPack 10. *
**************************************************************** |
| Local Fix: |
1. To stop the executing db2fmp process run 'db2fmpterm
<db2fmp_PID>'
2. Remove the record from the db2diag.log file or start a new
db2diag.log ('db2diag -A') file and remove the old file. |
| Solution |
First fixed in DB2 version 9.7 FixPack 10. |
| Workaround |
not known / see Local fix |
| Timestamps |
Date - problem reported :
Date - problem closed :
Date - last modified :
| 09.04.2014
24.11.2014
24.11.2014 |
| Problem solved at the following versions (IBM BugInfos) |
9.7.FP10
|
| Problem solved according to the fixlist(s) of the following version(s) |
| 9.7.0.10
|
 |
