DB2 - Problem description
Problem IC83191 | Status: Closed |
FAST WRITER EDU MAY NOT START UP WHICH CAN LEAD TO A HANG | |
product: | |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 | |
Problem description: | |
Fast Writer EDUs are integral to the recording of LOCKING, ACTIVITIES, UNIT OF WORK and PACKAGE CACHE event monitors. They start up automatically when a database activates. It is possible for one or more Fast Writers to fail to activate and fail with SQLCODE: SQL1040N The maximum number of applications is already connected to the database. and the following db2diag.log probe is recorded: 2012-03-13-08.06.22.099088-240 I207559262E1549 LEVEL: Error PID : 24223 TID : 47103166376256PROC : db2sysc INSTANCE: svtdbm8 NODE : 000 DB : LNX3 APPHDL : 0-7821 APPID: *LOCAL.svtdbm8.120315115333 AUTHID : SVTDBM8 EDUID : 10205 EDUNAME: db2agent (LNX3) FUNCTION: DB2 UDB, database monitor, sqmFastWriterManager::startupIcoord, probe:20 MESSAGE : ZRC=0x8005006D=-2147155859=SQLE_CA_BUILT "SQLCA has been built and saved in component specific control block." CALLED : DB2 UDB, base sys utilities, sqeIcoordCB::DispatchIndependentDBAgent RETCODE : ZRC=0x8005006D=-2147155859=SQLE_CA_BUILT "SQLCA has been built and saved in component specific control block." DATA #1 : signed integer, 4 bytes -1040 DATA #2 : Hexdump, 136 bytes 0x00000002011B5D20 : 5351 4C43 4120 2020 8800 0000 F0FB FFFF SQLCA ........ 0x00000002011B5D30 : 0000 2020 2020 2020 2020 2020 2020 2020 .. 0x00000002011B5D40 : 2020 2020 2020 2020 2020 2020 2020 2020 0x00000002011B5D50 : 2020 2020 2020 2020 2020 2020 2020 2020 0x00000002011B5D60 : 2020 2020 2020 2020 2020 2020 2020 2020 0x00000002011B5D70 : 2020 2020 2020 2020 5351 4C45 5342 434E SQLESBCN 0x00000002011B5D80 : 0000 0000 0000 0000 0000 0000 0000 0000 ................ 0x00000002011B5D90 : 0000 0000 0000 0000 2020 2020 2020 2020 ........ 0x00000002011B5DA0 : 2020 2020 2020 2020 If one or more fast writers fail to start up and an aforementioned event monitor is activated and event collection occurs, the instance may hang when all fast writer records end up on the queue of a failed fast writer (all records waiting to be processed) and agents then block waiting for records to be freed up. One possible symptom of this is that the Queue Manager Information displayed in the "db2pd -gfw" output may show zero free records available and have agents blocked waiting for records to be made available. For example: Queue Manager Info Current Number of Records: 1585 Free Records Available : 0 Maximum Number of Records: 85000 Maximum Cached Records : 30000 Minimum Cached Records : 1000 Number of Agents Blocked : 1 | |
Problem Summary: | |
**************************************************************** * USERS AFFECTED: * * ALL * **************************************************************** * PROBLEM DESCRIPTION: * * It is * * possible for one or more Fast Writers to fail to activate * * and * * fail with SQLCODE: * * * * SQL1040N The maximum number of applications is already * * connected to the database. * **************************************************************** * RECOMMENDATION: * * Upgrade to DB2 V9.7 Fix Pack 7. * **************************************************************** | |
Local Fix: | |
available fix packs: | |
DB2 Version 9.7 Fix Pack 7 for Linux, UNIX, and Windows | |
Solution | |
This will be fixed in DB2 V9.7 Fix Pack 7. | |
Workaround | |
not known / see Local fix | |
Timestamps | |
Date - problem reported : Date - problem closed : Date - last modified : | 08.05.2012 20.10.2012 20.10.2012 |
Problem solved at the following versions (IBM BugInfos) | |
9.7.FP7 | |
Problem solved according to the fixlist(s) of the following version(s) | |
9.7.0.7 |