DB2 - Problembeschreibung
| Problem IC80600 | Status: Geschlossen |
DB2 HADR STANDBY MARKED BAD DUE TO "LOG PAGE CHECKSUM MISMATCH" | |
| Produkt: | |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 | |
| Problembeschreibung: | |
A HADR standby using superasync synchronous mode could hit log
page checksum mismatch issue bringing down the standby database.
Please look into the DB2 diaglog.log file for the relevant
messages similar to the below ones.
2011-11-26-04.02.07.346200-300 I398177E457 LEVEL: Error
PID : 1593 TID : 140118371329792PROC :
db2sysc
INSTANCE: svtdbm NODE : 000 DB : HDRLNX
APPHDL : 0-8 APPID: *LOCAL.DB2.111126055301
EDUID : 41 EDUNAME: db2shred (HDRLNX)
FUNCTION: DB2 UDB, High Availability Disaster Recovery,
hdrGetBlock, probe:40100
MESSAGE : Log page checksum mismatch. pageLsn 0001DBC66FFB
2011-11-26-04.02.07.349893-300 I398635E474 LEVEL: Error
PID : 1593 TID : 140118379718400PROC :
db2sysc
INSTANCE: svtdbm NODE : 000 DB : HDRLNX
APPHDL : 0-8 APPID: *LOCAL.DB2.111126055301
EDUID : 39 EDUNAME: db2redom (HDRLNX)
FUNCTION: DB2 UDB, recovery manager, sqlpPRecReadLog, probe:1275
RETCODE : ZRC=0x87800148=-2021654200=HDR_ZRC_BAD_LOG
"HADR standby found bad log" | |
| Problem-Zusammenfassung: | |
**************************************************************** * USERS AFFECTED: * * HADR users using SUPERASYNC sync mode on all platforms. * **************************************************************** * PROBLEM DESCRIPTION: * * The problem is that in superasyn mode, it is possible for * * lfr to read a little ahead of logw causing a bad page read. * * The fix is to do a checksome verification on Primary before * * the page is shipped to standby. If the checksome fails then * * the page will be reread. * **************************************************************** * RECOMMENDATION: * * Recommend customers to upgrade to the product DB2 version * * v97 fixpack 5 or later. * **************************************************************** | |
| Local-Fix: | |
Manually copy the log file for which the log page checksum was reported from Primary to Standby and restart the standby. | |
| Lösung | |
The problem is first fixed in DB2 version v97 fixpack 5. | |
| Workaround | |
keiner bekannt / siehe Local-Fix | |
| Weitere Daten | |
Datum - Problem gemeldet : Datum - Problem geschlossen : Datum - der letzten Änderung: | 27.12.2011 12.03.2012 12.03.2012 |
| Problem behoben ab folgender Versionen (IBM BugInfos) | |
9.7.FP5 | |
| Problem behoben lt. FixList in der Version | |