You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Anton Slutsky (JIRA)" <ji...@apache.org> on 2013/01/31 22:11:13 UTC
[jira] [Comment Edited] (CASSANDRA-5209) For small datasets/single
hadoop/single cassandra clusters, ColumnFamilyRecordReader throws
InvalidRequestException(why:Start token sorts after end token)
[ https://issues.apache.org/jira/browse/CASSANDRA-5209?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13568103#comment-13568103 ]
Anton Slutsky edited comment on CASSANDRA-5209 at 1/31/13 9:10 PM:
-------------------------------------------------------------------
diff --git a/src/java/org/apache/cassandra/dht/Range.java b/src/java/org/apache/cassandra/dht/Range.java
index 99286fe..402fe6c 100644
--- a/src/java/org/apache/cassandra/dht/Range.java
+++ b/src/java/org/apache/cassandra/dht/Range.java
@@ -229,12 +229,20 @@ public class Range<T extends RingPosition> extends AbstractBounds<T> implements
public List<Range<T>> unwrap()
{
- T minValue = (T) partitioner.minValue(right.getClass());
+ T minValue = (T) partitioner.minValue(right.getClass());
if (!isWrapAround() || right.equals(minValue))
return Arrays.asList(this);
+
+ List<T> boundaries = new ArrayList<T>(3);
+ boundaries.add(minValue);
+ boundaries.add(left);
+ boundaries.add(right);
+
+ Collections.<T>sort(boundaries);
+
List<Range<T>> unwrapped = new ArrayList<Range<T>>(2);
- unwrapped.add(new Range<T>(left, minValue, partitioner));
- unwrapped.add(new Range<T>(minValue, right, partitioner));
+ unwrapped.add(new Range<T>(boundaries.get(0), boundaries.get(1), partitioner));
+ unwrapped.add(new Range<T>(boundaries.get(1), boundaries.get(2), partitioner));
return unwrapped;
}
was (Author: anton.slutsky):
Attached is the fix that worked for me if anybody cares
> For small datasets/single hadoop/single cassandra clusters, ColumnFamilyRecordReader throws InvalidRequestException(why:Start token sorts after end token)
> ----------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-5209
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5209
> Project: Cassandra
> Issue Type: Bug
> Components: Hadoop
> Affects Versions: 1.2.1
> Environment: Ubuntu 12.04
> Reporter: Anton Slutsky
> Fix For: 1.2.2
>
>
> When running the word count example, the following exception is thrown in most cases, but not all the time. I have debugged through and it appears that the reason is that Partitioner implementations can get into a strange state producing splits with ranges with left value being less than minValue. This confuses org.apache.cassandra.dht.Range.unwrap() which returns bogus ranges that start later than they finish.
> java.lang.RuntimeException: InvalidRequestException(why:Start token sorts after end token)
> at org.apache.cassandra.hadoop.ColumnFamilyRecordReader$WideRowIterator.maybeInit(ColumnFamilyRecordReader.java:475)
> at org.apache.cassandra.hadoop.ColumnFamilyRecordReader$WideRowIterator.computeNext(ColumnFamilyRecordReader.java:481)
> at org.apache.cassandra.hadoop.ColumnFamilyRecordReader$WideRowIterator.computeNext(ColumnFamilyRecordReader.java:428)
> at com.google.common.collect.AbstractIterator.tryToComputeNext(AbstractIterator.java:143)
> at com.google.common.collect.AbstractIterator.hasNext(AbstractIterator.java:138)
> at org.apache.cassandra.hadoop.ColumnFamilyRecordReader.nextKeyValue(ColumnFamilyRecordReader.java:199)
> at org.apache.hadoop.mapred.MapTask$NewTrackingRecordReader.nextKeyValue(MapTask.java:532)
> at org.apache.hadoop.mapreduce.MapContext.nextKeyValue(MapContext.java:67)
> at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:143)
> at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:764)
> at org.apache.hadoop.mapred.MapTask.run(MapTask.java:370)
> at org.apache.hadoop.mapred.LocalJobRunner$Job.run(LocalJobRunner.java:212)
> Caused by: InvalidRequestException(why:Start token sorts after end token)
> at org.apache.cassandra.thrift.Cassandra$get_paged_slice_result.read(Cassandra.java:14168)
> at org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:78)
> at org.apache.cassandra.thrift.Cassandra$Client.recv_get_paged_slice(Cassandra.java:769)
> at org.apache.cassandra.thrift.Cassandra$Client.get_paged_slice(Cassandra.java:753)
> at org.apache.cassandra.hadoop.ColumnFamilyRecordReader$WideRowIterator.maybeInit(ColumnFamilyRecordReader.java:460)
> ... 11 more
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira