Oracle / LogMiner Issue ORA-01287

14 views
Skip to first unread message

Ajay Bhatnagar

unread,
Feb 12, 2026, 5:05:52 PM (10 days ago) Feb 12
to debezium
Hello,

We are using Oracle with LogMiner (until we get XStream working), and are running into this error. This specific DB instance gets restored periodically from another env, and is likely the root cause here. Has anyone found a good workaround?

org.apache.kafka.connect.errors.ConnectException: An exception occurred in the change event producer. This connector will be stopped.
 at io.debezium.pipeline.ErrorHandler.setProducerThrowable(ErrorHandler.java:67)
 at io.debezium.connector.oracle.logminer.LogMinerStreamingChangeEventSource.execute(LogMinerStreamingChangeEventSource.java:254)
 at io.debezium.connector.oracle.logminer.LogMinerStreamingChangeEventSource.execute(LogMinerStreamingChangeEventSource.java:62)
 at io.debezium.pipeline.ChangeEventSourceCoordinator.streamEvents(ChangeEventSourceCoordinator.java:205)
 at io.debezium.pipeline.ChangeEventSourceCoordinator.executeChangeEventSources(ChangeEventSourceCoordinator.java:172)
 at io.debezium.pipeline.ChangeEventSourceCoordinator.lambda$start$0(ChangeEventSourceCoordinator.java:118)
 at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539)
 at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
 at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
 at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
 at java.base/java.lang.Thread.run(Thread.java:833)
Caused by: java.sql.SQLException: ORA-01287: file +RECOC8/LMPERF_OCI/ARCHIVELOG/2026_02_09/thread_1_seq_731.6096.1224784579 is from a different database incarnation
ORA-06512: at "SYS.DBMS_LOGMNR", line 82
ORA-06512: at line 1

Thanks,

Ajay

Chris Cranford

unread,
Feb 12, 2026, 7:58:23 PM (10 days ago) Feb 12
to debe...@googlegroups.com
Hi Ajay -

This most often happens when a database administrator restores the database from another environment, as a post restore step you roll the system incarnation forward. This should have been fixed in Debezium 3.1.2.Final and 3.2.0.Final as part of DBZ-8744 [1]. Are you using a version prior to those builds? If so, can you please update.

[1]: https://issues.redhat.com/browse/DBZ-8744
--
You received this message because you are subscribed to the Google Groups "debezium" group.
To unsubscribe from this group and stop receiving emails from it, send an email to debezium+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/debezium/7411c270-0b83-4440-beb7-46e5f5931a24n%40googlegroups.com.

Reply all
Reply to author
Forward
0 new messages