Sqoop导入报错Error: java.io.IOException: SQLException in nextKeyValue

Question

我编写了一个脚本来执行从 oracle 表到 HDFS 目录的数据增量导入。我使用以下 sqoop 命令进行导入：

    sqoop -- import \
   --connect $JDBCconnectionString  \
   --username $dbUserName \
   --password-file $passwordLocal \
   --query 'select * from dmt_sim.dim_product WHERE $CONDITIONS' \
   --split-by "PRODUCT_TITLE" \
   --incremental append \
   --check-column "KEY" \
   --last-value "1" \
   --append \
   --fields-terminated-by '\t' \
   --target-dir /user/ksrinivasan/dmn_product

变量 $JDBCconnectionString、$dbUserName、$passwordLocal 的值在运行期间被替换，因此与 oracle 数据库的连接和获取边界值都成功，但是当作业启动时它会抛出错误。

16/01/25 06:19:29 INFO mapreduce.Job: Job job_1452256584707_106782 running in uber mode : false
06:19:36            16/01/25 06:19:29 INFO mapreduce.Job:  map 0% reduce 0%
06:20:07            16/01/25 06:20:00 INFO mapreduce.Job: Task Id : attempt_1452256584707_106782_m_000000_0, Status : FAILED
06:20:07            Error: java.io.IOException: SQLException in nextKeyValue
06:20:07                at org.apache.sqoop.mapreduce.db.DBRecordReader.nextKeyValue(DBRecordReader.java:277)
06:20:07                at org.apache.hadoop.mapred.MapTask$NewTrackingRecordReader.nextKeyValue(MapTask.java:553)
06:20:07                at org.apache.hadoop.mapreduce.task.MapContextImpl.nextKeyValue(MapContextImpl.java:80)
06:20:07                at org.apache.hadoop.mapreduce.lib.map.WrappedMapper$Context.nextKeyValue(WrappedMapper.java:91)
06:20:07                at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
06:20:07                at org.apache.sqoop.mapreduce.AutoProgressMapper.run(AutoProgressMapper.java:64)
06:20:07                at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:784)
06:20:07                at org.apache.hadoop.mapred.MapTask.run(MapTask.java:341)
06:20:07                at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:163)
06:20:07                at java.security.AccessController.doPrivileged(Native Method)
06:20:07                at javax.security.auth.Subject.doAs(Subject.java:415)
06:20:07                at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1628)
06:20:07                at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:158)
06:20:07            Caused by: java.sql.SQLSyntaxErrorException: ORA-00907: missing right parenthesis
06:20:07            
06:20:07                at oracle.jdbc.driver.T4CTTIoer.processError(T4CTTIoer.java:445)
06:20:07                at oracle.jdbc.driver.T4CTTIoer.processError(T4CTTIoer.java:396)
06:20:07                at oracle.jdbc.driver.T4C8Oall.processError(T4C8Oall.java:879)
06:20:07                at oracle.jdbc.driver.T4CTTIfun.receive(T4CTTIfun.java:450)
06:20:07                at oracle.jdbc.driver.T4CTTIfun.doRPC(T4CTTIfun.java:192)
06:20:07                at oracle.jdbc.driver.T4C8Oall.doOALL(T4C8Oall.java:531)
06:20:07                at oracle.jdbc.driver.T4CPreparedStatement.doOall8(T4CPreparedStatement.java:207)
06:20:07                at oracle.jdbc.driver.T4CPreparedStatement.executeForDescribe(T4CPreparedStatement.java:884)
06:20:07                at oracle.jdbc.driver.OracleStatement.executeMaybeDescribe(OracleStatement.java:1167)
06:20:07                at oracle.jdbc.driver.OracleStatement.doExecuteWithTimeout(OracleStatement.java:1289)
06:20:07                at oracle.jdbc.driver.OraclePreparedStatement.executeInternal(OraclePreparedStatement.java:3584)
06:20:07                at oracle.jdbc.driver.OraclePreparedStatement.executeQuery(OraclePreparedStatement.java:3628)
06:20:07                at oracle.jdbc.driver.OraclePreparedStatementWrapper.executeQuery(OraclePreparedStatementWrapper.java:1493)
06:20:07                at org.apache.sqoop.mapreduce.db.DBRecordReader.executeQuery(DBRecordReader.java:111)
06:20:07                at org.apache.sqoop.mapreduce.db.DBRecordReader.nextKeyValue(DBRecordReader.java:235)
06:20:07                ... 12 more
06:20:07            

如果其他人遇到同样的问题并且他们对如何调试此类问题有一些见解，那就太好了？？？

Answer 1

nextKeyValue 中的 SQLException 这意味着在您执行时，您的源 KEY 列中不包含最后一个值“1”。

Answer 2

检查所有可能的列值是否为空，并特别检查日期可能为“0000-00-00”并且您在 dmt_sim.dim_product 的 select 子句中使用的日期列。

识别列并应用 if condition in select 来替换 null 或者和别的东西约会（你能认出来）。

Answer 3

这是因为用于拆分的列不是数字（“PRODUCT_TITLE”）。

当 Sqoop 翻译查询失败时，间隔为“PRODUCT_TITLE” >= xx <= yy

带有数字字段的特征。

执行语句时include --verbose 2> file_log.log（看分句是否正确）