The Golden Gate replication working as usual, but some fine day the extract process was abended. I later tried to check the report of the extract process abended and found the below error information from the report.
GGSCI> view report INEXT1 2021-06-06 07:54:22 ERROR OGG-02028 Failed to attach to logmining server OGG$INEXT1 error 1,292 - ORA-01292: LogMiner for upstream capture cannot find the log file. 2021-06-06 07:54:22 ERROR OGG-01668 PROCESS ABENDING.
Also, let’s verify the error log information from the alert log and it says the same error information as required sequence 88988 is missing dues to the deletion of archive logs by mistake.
GGSCI (ORA-X2) 2> select distinct originating_timestamp, message_text from V$DIAG_ALERT_EXT where originating_timestamp > sysdate - 1; ORIGINATING_TIMESTAMP --------------------------------------------------------------------------- MESSAGE_TEXT -------------------------------------------------------------------------------- 06-JUN-21 07.54.22.644000000 AM +02:00 LOGMINER: Missing log for upstream capture session# 1, FILE NA, THREAD 1, SEQUEN CE 88988, SCN 0x000000281b876999 06-JUN-21 07.54.22.697000000 AM +02:00 OGG Capture client successfully detached from GoldenGate Capture OGG$CAP_INEXT1 (OS pid=17884). SQL> select name from v$archived_log where sequence#=88988; NAME -------------------------------------------------------------------------------- SQL>
The above query does not show an archive log of sequence 88988.
If the required archive log is missing, then there is no way to establish the replication between the source and target. Since this environment is POC only, hence in such or similar cases, the solution can be fixed by unregistering and registering the extract process.
First, unregister the extract process from the database and delete the extract process.
GGSCI (ORA-X2 as ggadmin@PRODT) 3> unregister extract inext1, database 2021-06-06 08:22:29 INFO OGG-01750 Successfully unregistered EXTRACT INEXT1 from database. GGSCI (ORA-X2 as ggadmin@PRODT) 4> delete extract inext1 Deleted EXTRACT INEXT1. GGSCI (ORA-X2 as ggadmin@PRODT) 5>
Again add the extract process, but you should insist extract to start the reading the tranlog from NOW.
GGSCI (ORA-X2 as ggadmin@PRODT) 5> add extract inext1 integrated tranlog begin now EXTRACT (Integrated) added. GGSCI (ORA-X2 as ggadmin@PRODT) 6> add exttrail ./dirdat/et extract inext1 megabytes 1000 EXTTRAIL added. GGSCI (ORA-X2 as ggadmin@PRODT) 7> register extract inext1, database 2021-06-06 08:24:10 INFO OGG-02003 Extract INEXT1 successfully registered with database at SCN 172565923676.
Now, let’s review the configuration of the extract process.
GGSCI (ORA-X2 as ggadmin@PRODT) 8> view params inext1 extract inext1 userid ggadmin@PRODT, password Welcome1 tranlogoptions integratedparams (max_sga_size 2048) exttrail /dirdat/et logallsupcols tranlogoptions updaterecordformat full warnlongtrans 4h, checkinterval 10m, skipemptytrans br broff tranlogoptions excludetag 00112233445566778899AABBCCDDEEFF tranlogoptions excludeuser ggadmin tranlogoptions includeregionid statoptions resetreportstats report at 00:01 reportrollover at 00:03 reportcount every 1 minutes, rate include dirprm/includefile.inc GGSCI (ORA-X2 as ggadmin@PRODT) 9>
Verify the Extract process and start the process.
GGSCI (ORA-X2 as ggadmin@PRODT) 9> info all Program Status Group Lag at Chkpt Time Since Chkpt MANAGER RUNNING EXTRACT STOPPED INEXT1 00:00:00 00:01:19 EXTRACT STOPPED PMP1 00:00:00 09:32:17 REPLICAT STOPPED INREP1 00:00:00 09:21:54 GGSCI (ORA-X2 as ggadmin@PRODT) 10> start inext1 Sending START request to MANAGER ... EXTRACT INEXT1 starting GGSCI (ORA-X2 as ggadmin@PRODT) 11> info all Program Status Group Lag at Chkpt Time Since Chkpt MANAGER RUNNING EXTRACT RUNNING INEXT1 00:00:00 00:01:24 EXTRACT STOPPED PMP1 00:00:00 09:32:22 REPLICAT STOPPED INREP1 00:00:00 09:21:59 GGSCI (ORA-X2 as ggadmin@PRODT) 12>
Tip: ##IE capturing from an autonomous database abends with OGG-02028/ORA-01284 when extract has been down for more than 36 hours (Doc ID 2783387.1)
