本文共 2658 字，大约阅读时间需要 8 分钟。

2018-06-22 14:58:04 WARNING OGG-01003 Oracle GoldenGate Delivery for Oracle, repsl.prm: Repositioning to rba 67512607 in seqno 320.

2018-06-22 14:58:04 WARNING OGG-01154 Oracle GoldenGate Delivery for Oracle, repsl.prm: SQL error 14402 mapping SE.T_TEST_SET to TS.T_TEST_SET OCI Error ORA-14402: updating partition key column would cause a partition change. UPDATE "SSP"."T_TEST_SET" SET "STR_DT" = :a1 WHERE "SET_ONO" = :b0.2018-06-22 14:58:04 WARNING OGG-01003 Oracle GoldenGate Delivery for Oracle, repsl.prm: Repositioning to rba 67512607 in seqno 320.2018-06-22 14:58:04 ERROR OGG-01296 Oracle GoldenGate Delivery for Oracle, repsl.prm: Error mapping from SE.T_TEST_SET to TS.T_TEST_SET.2018-06-22 14:58:04 ERROR OGG-01668 Oracle GoldenGate Delivery for Oracle, repsl.prm: PROCESS ABENDING.

The replication process for the REPSL group has encountered an error during the synchronization of data. The error log indicates that an SQL statement is failing to map correctly, resulting in an ORA-14402 error. This error occurs when updating a partition key column, which would cause a partition change. The failed SQL statement is:

UPDATE "SSP"."T_TEST_SET" SET "STR_DT" = :a1 WHERE "SET_ONO" = :b0

Upon examining the GoldenGate logs and performing a logdump, it was identified that the replication process had paused due to an ABENDED status. The process was stuck in a state where it could not continue without resolving the mapping error.

To resolve this, the following steps were taken:

Identify the problematic sequence number and RBA (Redo Branch Advancement): The logdump revealed that the process was attempting to reposition to RBA 67512607 in sequence number 320. This indicated that the replication was trying to access an incorrect or invalid position in the log.

Adjust the sequence number and RBA: The alter REPSL ,extseqno 320 extrba 67512782 command was executed to update the sequence number and the RBA. This effectively reset the replication process to a valid state.

Restart the replication process: After confirming that the REPSL group was ABENDED, the start repsl command was issued to resume the replication process.

Following these steps, the replication process was successfully restarted, and the error condition was resolved. The replication group now operates normally, with no further ABENDED states or ORA-14402 errors being reported.

This issue highlights the importance of monitoring GoldenGate replication processes and being able to respond promptly to errors that prevent the replication from continuing. Regular log analysis and adjustments are crucial to maintaining the integrity and availability of replicated data.

转载地址：http://ivpfk.baihongyu.com/