|
Problem IC94964
|
Status: Closed |
DROPPING ALL INDEXES FROM INDEX OBJECT FOLLOWING REORG INDEX RECLAIM EXTENTS MAY RETURN SQL1034C AND CAUSE INSTANCE ABEND |
| product: |
DB2 FOR LUW / DB2FORLUW / A50 - DB2 |
| Problem description: |
SQL1034C may occur when doing extent reclaim on index object and
dropping all indexes within it, such as:
REORG INDEXES ALL FOR TABLE T1 CLEANUP RECLAIM EXTENTS
DROP INDEX INDX1
...
DROP INDEX INDXN
This may result in:
SQL1034C The database is damaged. All applications processing
the database
have been stopped. SQLSTATE=58031
The database then must undergo crash recovery.
The following conditions must be met for the problem to occur:
1) Extent reclaim of an index object.
2) The index must be located in an old-type DMS tablespace, i.e.
a DMS table space created before 9.7 and migrated to the current
release.
From db2diag.log, the error starts like this:
2013-08-15-14.57.27.937959-240 I167322E4722 LEVEL:
Severe
PID : 16643 TID : 46913369925952 KTID :
16664
PROC : db2sysc
INSTANCE: db2inst1 NODE : 000 DB :
SAMPLE
APPHDL : 0-44 APPID:
*LOCAL.db2inst1.130815185723
AUTHID : DB2INST1 HOSTNAME: myhostname
EDUID : 18 EDUNAME: db2agent (SAMPLE)
FUNCTION: DB2 UDB, buffer pool services,
sqlbPurgeOrFlushAllPagesInSmallRange, probe:8555
MESSAGE : ZRC=0x8402001A=-2080243686=SQLB_EMP_MAP_INFO_END "EMP
MAP INFO END"
DATA #1 : String, 38 bytes
Error mapping object page to pool page
The top stack trace:
sqlbPurgeOrFlushAllPagesInSmallRange
sqlbPurgeObjectLocal
sqlbPurgeObject
sqldmpnd
sqlptppl
sqlpxcm1
sqlrrcom_dps
sqlrrcom
sqlrr_commit |
| Problem Summary: |
****************************************************************
* USERS AFFECTED: *
* See Error Description *
****************************************************************
* PROBLEM DESCRIPTION: *
* See Error Description *
****************************************************************
* RECOMMENDATION: *
* Upgrade to DB2 10.5 for Linux, UNIX, and Windows Fix Pack 3 *
**************************************************************** |
| Local Fix: |
1. Avoid extent reclaim, OR
2. db2set DB2_HASH_PURGE_THRESH=0. A side effect is a
potentially slower performance when dropping small objects,
accompanied with a possible latch contention. However, if small
objects are not being dropped frequently, this workaround will
be acceptable. Setting DB2_HASH_PURGE_THRESH=0 reverts the
performance of dropping small objects to the same level as
pre-9.7 releases. |
| available fix packs: |
DB2 Cancun Release 10.5.0.4 (also known as Fix Pack 4) for Linux, UNIX, and Windows
DB2 Version 10.5 Fix Pack 9 for Linux, UNIX, and Windows
|
| Solution |
Problem first fixed in DB2 10.5 for Linux, UNIX, and Windows Fix
Pack 3 |
| Workaround |
not known / see Local fix |
| BUG-Tracking |
forerunner : APAR is sysrouted TO one or more of the following: IC98005 IT09645
follow-up : |
| Timestamps |
Date - problem reported :
Date - problem closed :
Date - last modified :
| 16.08.2013
19.09.2014
19.09.2014 |
| Problem solved at the following versions (IBM BugInfos) |
|
| Problem solved according to the fixlist(s) of the following version(s) |
| 10.5.0.4
|
 |
