DB2 - Problem description
| Problem IC78090 | Status: Closed |
SHORTLY AFTER HADR STANDBY CONNECTS TO PRIMARY, STANDBY COMES DOWN WITH MESSAGE "HADR STANDBY FOUND BAD LOG" IN THE DIAGLOG. | |
| product: | |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 | |
| Problem description: | |
A newly connected HADR standby will be brought down in the
following scenario (when the following situation exists).
a) On Primary: Table <tab1> has uncommitted changes.
b) On Primary: Reader reads <tab1> with currently committed
isolation.
c) On Primary: Primary has logs that are archived, no longer
exist in the active log path, and which have not
shipped to standby yet.
d) Standby connects to primary, requests log files for remote
catchup.
Normally, primary should retrieve the log file from archive
if it's not in active log path any more. However, when the
mentioned situation takes place, primary skips retrieving the
log files , primary's diaglog will contain messages similar to
these,
2011-08-02-05.45.52.552545-240 I2122290A481 LEVEL: Error
PID : 11734 TID : 55 PROC : db2sysc
INSTANCE: svtdbm9 NODE : 000
EDUID : 55 EDUNAME: db2hadrp (SB27069)
FUNCTION: DB2 UDB, High Availability Disaster Recovery, hdrEduP,
probe:20590
MESSAGE : HADR primary database failed to read log pages for
remote catchup.
sqlplfrScanNext returned rc = 0x860f000a, scanPages
= 0, scanFlagsOut
= 0x2
On standby, the following messages will appear in db2diag.log
afterwards:
2011-08-02-05.45.42.235773-240 I328422A429 LEVEL: Warning
PID : 11255 TID : 49 PROC : db2sysc
INSTANCE: svtdbm9 NODE : 000
EDUID : 49 EDUNAME: db2hadrs (SB27069)
FUNCTION: DB2 UDB, High Availability Disaster Recovery,
hdrSPrepareLogWrite, probe:10325
MESSAGE : Incomplete log record at end, RCUStartLSO 608499820376
recoveryLastPageLso 608499844480
2011-08-02-05.45.53.748220-240 I328852A368 LEVEL: Error
PID : 11255 TID : 49 PROC :
db2sysc
INSTANCE: svtdbm9 NODE : 000
EDUID : 49 EDUNAME: db2hadrs (SB27069)
FUNCTION: DB2 UDB, High Availability Disaster Recovery,
hdrEduS, probe:21210
MESSAGE : HADR: Missing log on primary. Standby shutting down
2011-08-02-05.45.53.748857-240 I329221A398 LEVEL: Error
PID : 11255 TID : 49 PROC :
db2sysc
INSTANCE: svtdbm9 NODE : 000
EDUID : 49 EDUNAME: db2hadrs (SB27069)
FUNCTION: DB2 UDB, High Availability Disaster Recovery,
hdrEduS, probe:21210
RETCODE : ZRC=0x87800148=-2021654200=HDR_ZRC_BAD_LOG
"HADR standby found bad log" | |
| Problem Summary: | |
**************************************************************** * USERS AFFECTED: * * HADR users on all platforms * **************************************************************** * PROBLEM DESCRIPTION: * * Without the fix, customer could be exposed to the problem * * described in the APAR error description * **************************************************************** * RECOMMENDATION: * * Upgrade to product v97fp5 * **************************************************************** | |
| Local Fix: | |
Try start hadr on standby again with no readers on Primary | |
| available fix packs: | |
DB2 Version 9.7 Fix Pack 6 for Linux, UNIX, and Windows | |
| Solution | |
The fix is delivered db2_v97fp5. Customers will not see the standby going down when the situation described in the APAR error description happens. | |
| Workaround | |
not known / see Local fix | |
| Timestamps | |
Date - problem reported : Date - problem closed : Date - last modified : | 11.08.2011 12.12.2011 14.02.2012 |
| Problem solved at the following versions (IBM BugInfos) | |
9.7.FP5 | |
| Problem solved according to the fixlist(s) of the following version(s) | |
| 9.7.0.6 |
|