Hi All,
I would like to ask you for opinions regarding troubleshooting steps. I have reproduced Golden Gate rep issue based on real scenario that happened on production. Let me describe the situation first.
1. environment:
db1 oracle 11g, gg11, linux os VM
db2 oracle 11g, gg11, linux os VM
2. topology
unidirectional
3. config
EXTRACT
EXTRACT ext1
USERID ggate, PASSWORD qwerty
EXTTRAIL /GG_INSTALL/dirdat/ex
DDL INCLUDE ALL
TABLE TEST_SCHEMA.*;
PUMP
EXTRACT dp1
USERID ggate, PASSWORD qwerty
RMTHOST db2.local, MGRPORT 7809
RMTTRAIL /GG_INSTALL/dirdat/rt
DDL INCLUDE ALL
TABLE TEST_SCHEMA.*;
REPLICAT
REPLICAT rep1
USERID ggate, PASSWORD qwerty
ASSUMETARGETDEFS
DISCARDFILE /GG_INSTALL/discards, PURGE
DDL INCLUDE ALL
DDLERROR DEFAULT IGNORE
MAP TEST_SCHEMA.*, TARGET TEST_SCHEMA.*;
4. error description, disaster scenario
table1 on source database consists 25k rows. same on target, replication is working just fine, ddl as well. I deleted one row on target database for table1 ID=10, then updated same row (ID=10) on source and watched how will golden gate behave.
error log shows as follows:
2017-09-18 11:03:26 WARNING OGG-01003 Oracle GoldenGate Delivery for Oracle, rep1.prm: Repositioning to rba 48230 in seqno 19.
2017-09-18 11:03:26 ERROR OGG-01296 Oracle GoldenGate Delivery for Oracle, rep1.prm: Error mapping from TEST_SCHEMA.TEST_TABLE6 to TEST_SCHEMA.TEST_TABLE6.
2017-09-18 11:03:26 ERROR OGG-01668 Oracle GoldenGate Delivery for Oracle, rep1.prm: PROCESS ABENDING.
GGSCI (db2.local) 81> info all
Program Status Group Lag at Chkpt Time Since Chkpt
MANAGER RUNNING
REPLICAT ABENDED REP1 00:39:32 01:08:14
So GG cannot udpate the row because it does not exist on target and replicat gets abended. when i try to restart it it stays in this error. now, is there any solution for this? i know that it is possible to import whole table to target for example with data pump and re-initiate replication.
Kindly advise, is there another solution for this kind of situation? thank you in advance for answers.
Greg