Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce, mail # user - NullPointerException when trying to write mapper output


Copy link to this message
-
NullPointerException when trying to write mapper output
Marcelo Elias Del Valle 2013-10-24, 17:19
I am using hadoop 1.0.3 at 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?