DB2 - Problem description
Problem IC75916 | Status: Closed |
MONITOR HEAP LEAKING IN SIZE 16 AND 24 BYTES MAY OCCUR IF THERE ARE MANY TASKS RUN LIKE RESET MONITOR AND INSTANCE DETACH. | |
product: | |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 | |
Problem description: | |
If there are many tasks like "db2 reset monitor all" and "db2 detach" run in the background, from the the db2pd -memblock output, it could be perceived that db2 is leaking 16 and 24 bytes monitor memory at a stretch, like below: Address PoolID PoolName BlockAge Size I LOC File 0x0780000003694AC8 11 monh 7195 24 1 146 3254716357 0x0780000003694B08 11 monh 43116 16 1 146 3254716357 0x0780000003694B48 11 monh 43117 24 1 146 3254716357 0x078000000369D8C8 11 monh 63753 16 1 146 3254716357 0x078000000369D908 11 monh 63754 24 1 146 3254716357 0x078000000369DA28 11 monh 27486 16 1 146 3254716357 0x078000000369DA68 11 monh 27487 24 1 146 3254716357 0x078000000369DDA8 11 monh 19330 16 1 146 3254716357 0x078000000369DE68 11 monh 30457 16 1 146 3254716357 0x078000000369DF68 11 monh 55378 16 1 146 3254716357 ... Under such condition, the monitor memory heap will NOT be released all the while and the error DIA8300C will be come up with in the db2diag.log at the end as below: 2011-04-11-13.52.10.204659+480 I1287415A500 LEVEL: Severe PID : 2150452 TID : 1 PROC : db2agent(instance) 0 INSTANCE: db2inst1 NODE : 000 APPHDL : 0-298 APPID:*LOCAL.db2inst1.2007070720080708 AUTHID : hujinpei FUNCTION: DB2 UDB, database monitor, sqm_object::new, probe:100 RETCODE : ZRC=0x8B0F000B=-1961951221=SQLO_NOMEM_MONH "No memory available in 'Database Monitor Heap'" DIA8300C A memory heap error has occurred. | |
Problem Summary: | |
**************************************************************** * USERS AFFECTED: * * ALL * **************************************************************** * PROBLEM DESCRIPTION: * * If there are many tasks like "db2 reset monitor all" and * * "db2 detach" run in the background, from the the db2pd * * -memblock output, it could be perceived that db2 is leaking * * 16 and 24 bytes monitor memory at a stretch, like below: * * * * Address PoolID PoolName BlockAge * * Size(Bytes) I LOC File * * 0x0780000003694AC8 11 monh 7195 24 * * 1 146 3254716357 * * 0x0780000003694B08 11 monh 43116 16 * * 1 146 3254716357 * * 0x0780000003694B48 11 monh 43117 24 * * 1 146 3254716357 * * 0x078000000369D8C8 11 monh 63753 16 * * 1 146 3254716357 * * 0x078000000369D908 11 monh 63754 24 * * 1 146 3254716357 * * 0x078000000369DA28 11 monh 27486 16 * * 1 146 3254716357 * * 0x078000000369DA68 11 monh 27487 24 * * 1 146 3254716357 * * 0x078000000369DDA8 11 monh 19330 16 * * 1 146 3254716357 * * 0x078000000369DE68 11 monh 30457 16 * * 1 146 3254716357 * * 0x078000000369DF68 11 monh 55378 16 * * 1 146 3254716357 * * ... * * * * Under such condition, the monitor memory heap will NOT be * * released all the while and the error DIA8300C will be come * * up with in the db2diag.log at the end as below: * * * * 2011-04-11-13.52.10.204659+480 I1287415A500 LEVEL: * * Severe * * PID : 2150452 TID : 1 PROC : * * db2agent * * (instance) 0 * * * * INSTANCE: db2inst1 NODE : 000 * * * * APPHDL : 0-298 APPID: * * *LOCAL.db2inst1.2007070720080708 * * AUTHID : hujinpei * * * * FUNCTION: DB2 UDB, database monitor, sqm_object::new, * * probe:100 * * RETCODE : ZRC=0x8B0F000B=-1961951221=SQLO_NOMEM_MONH * * * * "No memory available in 'Database Monitor Heap'" * * * * DIA8300C A memory heap error has occurred. * **************************************************************** * RECOMMENDATION: * * Upgrade to v9.7 FP5 * **************************************************************** | |
Local Fix: | |
The memory lack only happen when detaching the instance attachment. Keep using the same instance attachment to do the "reset monitor" and snapshot could avoid the memory leak. | |
available fix packs: | |
DB2 Version 9.7 Fix Pack 5 for Linux, UNIX, and Windows | |
Solution | |
The memory lack only happen when detaching the instance attachment. Keep using the same instance attachment to do the "reset monitor" and snapshot could avoid the memory leak. This defect will be fixed in v9.7 FP5 | |
Workaround | |
not known / see Local fix | |
Timestamps | |
Date - problem reported : Date - problem closed : Date - last modified : | 20.04.2011 10.12.2011 01.06.2012 |
Problem solved at the following versions (IBM BugInfos) | |
9.7.FP5 | |
Problem solved according to the fixlist(s) of the following version(s) | |
9.7.0.5 |