[DRILL-211] index out of bounds exception when reading parquet file

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.1.0-m1
• Fix Version/s: 0.1.0-m1
• Assignee: Jason Altekruse
• Reporter: Steven Phillips
• Created: 04/Sep/13 22:08
• Updated: 09/Oct/13 18:09
• Resolved: 06/Sep/13 06:04

Description

I see this when reading a table composed of 4 parquet files, each about 70MB.

java.lang.ArrayIndexOutOfBoundsException: 1048748
parquet.bytes.BytesUtils.readIntLittleEndian(BytesUtils.java:54) ~[parquet-common-1.0.1.jar:na]
org.apache.drill.exec.store.parquet.VarLenBinaryReader.readFields(VarLenBinaryReader.java:116) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.store.parquet.ParquetRecordReader.next(ParquetRecordReader.java:387) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.physical.impl.ScanBatch.next(ScanBatch.java:95) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.record.AbstractSingleRecordBatch.next(AbstractSingleRecordBatch.java:25) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.physical.impl.sort.SortBatch.next(SortBatch.java:95) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.physical.impl.aggregate.AggBatch.next(AggBatch.java:68) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.physical.impl.ScreenCreator$ScreenRoot.next(ScreenCreator.java:77) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
org.apache.drill.exec.work.FragmentRunner.run(FragmentRunner.java:79) ~[java-exec-1.0.0-m1-rebuffed.jar:1.0.0-m1]
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) [na:1.7.0_21]
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_21]
java.lang.Thread.run(Thread.java:722) [na:1.7.0_21]