Check out the troubleshooting section of the hadoop support - we ran into the same thing and tried to update that with some info on how to get around it: http://wiki.apache.org/cassandra/HadoopSupport#Troubleshooting
On Feb 24, 2012, at 7:20 AM, Patrik Modesto wrote: > Hi, > > I can see some strange behaviour on my test cluster and in production. > Both running cassandra 0.8.10. Strange is that when I compile my > mapreduce job against cassandra-all 0.8.7 everything is ok, but if I > use higher version I get quite a lots of TimedOutException. > > java.lang.RuntimeException: TimedOutException() > at > org.apache.cassandra.hadoop.ColumnFamilyRecordReader$RowIterator.maybeInit(ColumnFamilyRecordReader.java:319) > at > org.apache.cassandra.hadoop.ColumnFamilyRecordReader$RowIterator.computeNext(ColumnFamilyRecordReader.java:333) > at > org.apache.cassandra.hadoop.ColumnFamilyRecordReader$RowIterator.computeNext(ColumnFamilyRecordReader.java:207) > at > com.google.common.collect.AbstractIterator.tryToComputeNext(AbstractIterator.java:135) > at > com.google.common.collect.AbstractIterator.hasNext(AbstractIterator.java:130) > at > org.apache.cassandra.hadoop.ColumnFamilyRecordReader.nextKeyValue(ColumnFamilyRecordReader.java:163) > at > org.apache.hadoop.mapred.MapTask$NewTrackingRecordReader.nextKeyValue(MapTask.java:456) > 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:647) > at org.apache.hadoop.mapred.MapTask.run(MapTask.java:32 > > There is nothing in the cassandra log, the cluster is idle, no one > else is accessing the cluster. > > There are just few rows, nothing big: > > INFO mapred.JobClient: Reduce input groups=14 > INFO mapred.JobClient: Combine output records=0 > INFO mapred.JobClient: Map input records=544009 > INFO mapred.JobClient: Reduce shuffle bytes=33876 > INFO mapred.JobClient: Reduce output records=0 > INFO mapred.JobClient: Spilled Records=38 > INFO mapred.JobClient: Map output bytes=33656 > INFO mapred.JobClient: Combine input records=0 > INFO mapred.JobClient: Map output records=19 > INFO mapred.JobClient: SPLIT_RAW_BYTES=3937 > INFO mapred.JobClient: Reduce input records=19 > > What could be the problem? > > Regards, > P.