[ 
https://issues.apache.org/jira/browse/DRILL-493?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13963645#comment-13963645
 ] 

Steven Phillips commented on DRILL-493:
---------------------------------------

This problem seems to be related to how the parquet files were generated. the 
original data in the customer.json file has null values for one of the columns. 
This was converted to a tsv file, with empty strings for the null values, which 
was then converted to Parquet using a slightly modified version of parquet-pig. 
The modified version outputs REQUIRED columns, whereas the original outputs 
OPTIONAL columns. This modified version was a quick hack, and was not really 
tested, and I suspect the parquet file that it generates in this case is not 
valid.

> Mondrian query throws ArrayIndexOutOfBoundException
> ---------------------------------------------------
>
>                 Key: DRILL-493
>                 URL: https://issues.apache.org/jira/browse/DRILL-493
>             Project: Apache Drill
>          Issue Type: Bug
>    Affects Versions: 1.0.0-milestone-2
>            Reporter: Chun Chang
>            Priority: Blocker
>
> If you run pretty much any mondrian query, you will hit this exception:
> java.lang.ArrayIndexOutOfBoundsException
> 0
> at parquet.bytes.BytesUtils.readIntLittleEndian(BytesUtils.java:54) 
> ~[parquet-common-1.2.8.jar:na]
> at 
> org.apache.drill.exec.store.parquet.VarLenBinaryReader.readFields(VarLenBinaryReader.java:121)
>  
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at 
> org.apache.drill.exec.store.parquet.ParquetRecordReader.next(ParquetRecordReader.java:368)
>  
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at org.apache.drill.exec.physical.impl.ScanBatch.next(ScanBatch.java:94) 
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at 
> org.apache.drill.exec.record.AbstractSingleRecordBatch.next(AbstractSingleRecordBatch.java:42)
>  
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at 
> org.apache.drill.exec.physical.impl.ScreenCreator$ScreenRoot.next(ScreenCreator.java:85)
>  
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at 
> org.apache.drill.exec.work.fragment.FragmentExecutor.run(FragmentExecutor.java:83)
>  
> ~[drill-java-exec-1.0.0-m2-incubating-SNAPSHOT-rebuffed.jar:1.0.0-m2-incubating-SNAPSHOT]
> at 
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
>  [na:1.7.0_45]
> at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
>  [na:1.7.0_45]
> at java.lang.Thread.run(Thread.java:744) [na:1.7.0_45]



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to