[
https://issues.apache.org/jira/browse/SQOOP-2564?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mirek Szymanski updated SQOOP-2564:
-----------------------------------
Environment: Teradata 13, Oracle 11g
> Problem with Teradata input table with INTEGER and FLOAT columns
> ----------------------------------------------------------------
>
> Key: SQOOP-2564
> URL: https://issues.apache.org/jira/browse/SQOOP-2564
> Project: Sqoop
> Issue Type: Bug
> Affects Versions: 1.99.6
> Environment: Teradata 13, Oracle 11g
> Reporter: Mirek Szymanski
>
> I'm trying to transfer data between Teradata and Oracle databases.
> If the input table in Teradata has column of type INTEGER (columntype 'I' in
> dbc.columnsV) or FLOAT (columntype 'F') to Oracle table - exception is thrown
> "org.apache.sqoop.common.SqoopException: MAPRED_EXEC_0017:Error occurs during
> extractor run"
> Exception in case of FLOAT column:
> 2015-09-14 10:59:09,200 INFO [AsyncDispatcher event handler]
> org.apache.hadoop.mapreduce.v2.app.job.impl.TaskAttemptImpl: Diagnostics
> report from attempt_1441804561641_0239_m_000001_0: Error:
> org.apache.sqoop.common.SqoopException: MAPRED_EXEC_0017:Error occurs during
> extractor run
> at org.apache.sqoop.job.mr.SqoopMapper.run(SqoopMapper.java:99)
> at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:784)
> at org.apache.hadoop.mapred.MapTask.run(MapTask.java:341)
> at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:163)
> at java.security.AccessController.doPrivileged(Native Method)
> at javax.security.auth.Subject.doAs(Subject.java:415)
> at
> org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1628)
> at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:158)
> Caused by: java.lang.ClassCastException: java.lang.Double cannot be cast to
> java.lang.Float
> at
> org.apache.sqoop.connector.common.SqoopIDFUtils.toCSVFloatingPoint(SqoopIDFUtils.java:164)
> at
> org.apache.sqoop.connector.common.SqoopIDFUtils.toCSV(SqoopIDFUtils.java:606)
> at
> org.apache.sqoop.connector.idf.CSVIntermediateDataFormat.toCSV(CSVIntermediateDataFormat.java:116)
> at
> org.apache.sqoop.connector.idf.CSVIntermediateDataFormat.setObjectData(CSVIntermediateDataFormat.java:87)
> at
> org.apache.sqoop.job.mr.SqoopMapper$SqoopMapDataWriter.writeArrayRecord(SqoopMapper.java:125)
> at
> org.apache.sqoop.connector.jdbc.GenericJdbcExtractor.extract(GenericJdbcExtractor.java:91)
> at
> org.apache.sqoop.connector.jdbc.GenericJdbcExtractor.extract(GenericJdbcExtractor.java:38)
> at org.apache.sqoop.job.mr.SqoopMapper.run(SqoopMapper.java:95)
> ... 7 more
> In case of INTEGER column:
> ...
> java.lang.ClassCastException: java.lang.Integer cannot be cast to
> java.math.BigDecimal
> at
> org.apache.sqoop.connector.common.SqoopIDFUtils.toCSVDecimal(SqoopIDFUtils.java:176)
> at
> org.apache.sqoop.connector.common.SqoopIDFUtils.toCSV(SqoopIDFUtils.java:602)
> at
> org.apache.sqoop.connector.idf.CSVIntermediateDataFormat.toCSV(CSVIntermediateDataFormat.java:116)
> at
> org.apache.sqoop.connector.idf.CSVIntermediateDataFormat.setObjectData(CSVIntermediateDataFormat.java:87)
> at
> org.apache.sqoop.job.mr.SqoopMapper$SqoopMapDataWriter.writeContent(SqoopMapper.java:149)
> ... 11 more
> The fix (or maybe a workaround) is to remove the cast to BigDecimal from
> SqoopIDFUtils.toCSVDecimal(). When the cast is removed, my scenario works
> fine.
> I wonder if the casts are necessary in toCSVFloatingPoint and toCSVDecimal
> methods. What is the real purpose of them - some kind of type checking?
> If this is type checking - why don't use something more explicit, like
> instanceof()?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
