NullPointerException when writing mapper output - hadoop

Question

I am using Amazon EMR. I have a map / reduce job configured like this:
private static final String TEMP_PATH_PREFIX = System.getProperty("java.io.tmpdir") + "/dmp_processor_tmp";
...
private Job setupProcessorJobS3() throws IOException, DataGrinderException {
String inputFiles = System.getProperty(DGConfig.INPUT_FILES);
Job processorJob = new Job(getConf(), PROCESSOR_JOBNAME);
processorJob.setJarByClass(DgRunner.class);
processorJob.setMapperClass(EntityMapperS3.class);
processorJob.setReducerClass(SelectorReducer.class);
processorJob.setOutputKeyClass(Text.class);
processorJob.setOutputValueClass(Text.class);
FileOutputFormat.setOutputPath(processorJob, new Path(TEMP_PATH_PREFIX));
processorJob.setOutputFormatClass(TextOutputFormat.class);
processorJob.setInputFormatClass(NLineInputFormat.class);
FileInputFormat.setInputPaths(processorJob, inputFiles);
NLineInputFormat.setNumLinesPerSplit(processorJob, 10000);
return processorJob;
}
In my mapper class, I have:
private Text outkey = new Text();
private Text outvalue = new Text();
...
outkey.set(entity.getEntityId().toString());
outvalue.set(input.getId().toString());
printLog("context write");
context.write(outkey, outvalue);
This last line (context.write(outkey, outvalue);), causes this exception. Of course both outkey and outvalue are not null.
2013-10-24 05:48:48,422 INFO com.s1mbi0se.grinder.core.mapred.EntityMapperCassandra (main): Current Thread: Thread[main,5,main]Current timestamp: 1382593728422 context write
2013-10-24 05:48:48,422 ERROR com.s1mbi0se.grinder.core.mapred.EntityMapperCassandra (main): Error on entitymapper for input: 03a07858-4196-46dd-8a2c-23dd824d6e6e
java.lang.NullPointerException
at java.lang.System.arraycopy(Native Method)
at org.apache.hadoop.mapred.MapTask$MapOutputBuffer$Buffer.write(MapTask.java:1293)
at org.apache.hadoop.mapred.MapTask$MapOutputBuffer$Buffer.write(MapTask.java:1210)
at java.io.DataOutputStream.writeByte(DataOutputStream.java:153)
at org.apache.hadoop.io.WritableUtils.writeVLong(WritableUtils.java:264)
at org.apache.hadoop.io.WritableUtils.writeVInt(WritableUtils.java:244)
at org.apache.hadoop.io.Text.write(Text.java:281)
at org.apache.hadoop.io.serializer.WritableSerialization$WritableSerializer.serialize(WritableSerialization.java:90)
at org.apache.hadoop.io.serializer.WritableSerialization$WritableSerializer.serialize(WritableSerialization.java:77)
at org.apache.hadoop.mapred.MapTask$MapOutputBuffer.collect(MapTask.java:1077)
at org.apache.hadoop.mapred.MapTask$NewOutputCollector.write(MapTask.java:698)
at org.apache.hadoop.mapreduce.TaskInputOutputContext.write(TaskInputOutputContext.java:80)
at com.s1mbi0se.grinder.core.mapred.EntityMapper.map(EntityMapper.java:78)
at com.s1mbi0se.grinder.core.mapred.EntityMapperS3.map(EntityMapperS3.java:34)
at com.s1mbi0se.grinder.core.mapred.EntityMapperS3.map(EntityMapperS3.java:14)
at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:771)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:375)
at org.apache.hadoop.mapred.Child$4.run(Child.java:255)
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:1132)
at org.apache.hadoop.mapred.Child.main(Child.java:249)
2013-10-24 05:48:48,422 INFO com.s1mbi0se.grinder.core.mapred.EntityMapperS3 (main): Current Thread: Thread[main,5,main]Current timestamp: 1382593728422 Entity Mapper end
The first records on each task are just processed ok. In some point of the task processing, I start to take this exception over and over, and then it doesn't process a single record anymore for that task.
I tried to set TEMP_PATH_PREFIX to "s3://mybucket/dmp_processor_tmp", but same thing happened.
Any idea why is this happening? What could be making hadoop not being able to write on it's output?