oracel cdc 执行由全量数据同步结束转增量同步时报下述异常

[MonsterChenzhuo]

cdc版本：2.2.1
flink版本：1.3.6
oracle版本：11.2.0.2
报错内容：Caused by: Error : 1403, Position : 0, Sql = BEGIN DBMS_LOGMNR_D.BUILD (options => DBMS_LOGMNR_D.STORE_IN_REDO_LOGS); END;, OriginalSql = BEGIN DBMS_LOGMNR_D.BUILD (options => DBMS_LOGMNR_D.STORE_IN_REDO_LOGS); END;, Error Msg = ORA-01403: 未找到任何数据
ORA-06512: 在 "SYS.DBMS_LOGMNR_INTERNAL", line 5972
ORA-06512: 在 "SYS.DBMS_LOGMNR_INTERNAL", line 6070
ORA-06512: 在 "SYS.DBMS_LOGMNR_D", line 12
ORA-06512: 在 line 1

代码：

public class OracleToKafka{

public static void main(String[] args) throws Exception {
    Properties properties = new Properties();
    properties.setProperty("event.processing.failure.handling.mode", "skip");
    properties.setProperty("database.tablename.case.insensitive", "false");
    properties.setProperty("debezium.log.mining.strategy", "online_catalog");
    properties.setProperty("debezium.log.mining.continuous.mine", "true");


    SourceFunction<String> sourceFunction = OracleSource.<String>builder()
            .hostname("")
            .port(1521)
            .database("")
            .schemaList("")
            .tableList("")
            .username("")
            .password("")
            .debeziumProperties(properties)
            .deserializer(new JsonDebeziumDeserializationSchema())
            .build();
    Configuration config = new Configuration();
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(config);
    DataStreamSource<String> dataStreamSource = env.addSource(sourceFunction).setParallelism(1);
    dataStreamSource.setParallelism(1).print("====>");

    env.execute("FlinkCDCOracle");
}

}

[fuyun2024]

This is an error in the process of build data dictionary by debezium, can you execute this SQL : BEGIN DBMS_LOGMNR_D.BUILD (options => DBMS_LOGMNR_D.STORE_IN_REDO_LOGS); END;

[MonsterChenzhuo]

@fuyun2024 执行过一样报错，字典不写redo，换成写文件可以

[MonsterChenzhuo]

EXECUTE dbms_logmnr_d.build('dictionary.ora','/u01/zrprod/temp',options => dbms_logmnr_d.store_in_flat_file) 这种方式可以；
生成的字典文件1.8G，约需要1分多钟；
而在线redo存放字典直接报错
SYS@ZRPROD> BEGIN DBMS_LOGMNR_D.BUILD (options => DBMS_LOGMNR_D.STORE_IN_REDO_LOGS); END;
2 /
BEGIN DBMS_LOGMNR_D.BUILD (options => DBMS_LOGMNR_D.STORE_IN_REDO_LOGS); END;
*
ERROR at line 1:
ORA-01403: no data found
ORA-06512: at "SYS.DBMS_LOGMNR_INTERNAL", line 5972
ORA-06512: at "SYS.DBMS_LOGMNR_INTERNAL", line 6070
ORA-06512: at "SYS.DBMS_LOGMNR_D", line 12
ORA-06512: at line 1

[fuyun2024]

Have you started archiving logs？

[MonsterChenzhuo]

@fuyun2024 您好，给您添麻烦了，是db的问题logmnr配置的有问题

[shadon178]

应该是oracle的问题，我试过‘在线字典文件’和‘存储在归档日志中的字典文件’都可以的。