报错:
ERROR Task oracle-logminer-connector-0 threw an uncaught and unrecoverable exception (org.apache.kafka.connect.runtime.WorkerTask:141)
org.apache.kafka.connect.errors.ConnectException: Error at database tier, Please check : java.sql.SQLException: ORA-01292: no log file has been specified for the current LogMiner session
ORA-06512: at "SYS.DBMS_LOGMNR", line 58
ORA-06512: at line 2
at com.ecer.kafka.connect.oracle.OracleSourceTask.start(OracleSourceTask.java:218)
at org.apache.kafka.connect.runtime.WorkerSourceTask.execute(WorkerSourceTask.java:141)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:139)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:182)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
报错前的操作:
由于归档日志满了,
使用 rman target sys/Huawei12#$ 执行下边两条命令
DELETE ARCHIVELOG ALL COMPLETED BEFORE 'SYSDATE-7'; //删除七天前的归档
DELETE ARCHIVELOG FROM TIME 'SYSDATE-7'; //删除七天到现在的归档
做了删除数据,而后又插入数据,kafka-connection-oracle 还未处理,又执行了遍删除。
再重新启动后就报上面的error
原因:
reset.offset=false 是本来的设置
( 如果为true,则在连接器启动时将偏移值设置为数据库的当前SCN
如果为false,则连接器将从上一个偏移值开始 )
由于是false;而且有新的scn 还未消费就被删除,所以报上面的问题