Hadoop cluster - Do I need to replicate my code over all machines before running job?

Question

This is what confuses me, when I use wordcount example, I keep code at master and let him do things with slaves and it runs fine

But when I am running my code, it starts to fail on slaves giving weird errors like

Traceback (most recent call last):
  File "/app/hadoop/tmp/mapred/local/taskTracker/hduser/jobcache/job_201110250901_0005/attempt_201110250901_0005_m_000001_1/work/./mapper.py", line 55, in <module>
    from src.utilities import utilities
ImportError: No module named src.utilities
java.lang.RuntimeException: PipeMapRed.waitOutputThreads(): subprocess failed with code 1
        at org.apache.hadoop.streaming.PipeMapRed.waitOutputThreads(PipeMapRed.java:311)
        at org.apache.hadoop.streaming.PipeMapRed.mapRedFinished(PipeMapRed.java:545)
        at org.apache.hadoop.streaming.PipeMapper.map(PipeMapper.java:121)
        at org.apache.hadoop.mapred.MapRunner.run(MapRunner.java:50)
        at org.apache.hadoop.streaming.PipeMapRunner.run(PipeMapRunner.java:36)
        at org.apache.hadoop.mapred.MapTask.runOldMapper(MapTask.java:436)
        at org.apache.hadoop.mapred.MapTask.run(MapTask.java:372)
        at org.apache.hadoop.mapred.Child$4.run(Child.java:261)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1059)
        at org.apache.hadoop.mapred.Child.main(Child.java:255)
java.lang.RuntimeException: PipeMapRed.waitOutputThreads(): subprocess failed with code 1
        at org.apache.hadoop.streaming.PipeMapRed.waitOutputThreads(PipeMapRed.java:311)
        at org.apache.hadoop.streaming.PipeMapRed.mapRedFinished(PipeMapRed.java:545)
        at org.apache.hadoop.streaming.PipeMapper.close(PipeMapper.java:132)
        at org.apache.hadoop.mapred.MapRunner.run(MapRunner.java:57)
        at org.apache.hadoop.streaming.PipeMapRunner.run(PipeMapRunner.java:36)
        at org.apache.hadoop.mapred.MapTask.runOldMapper(MapTask.java:436)
        at org.apache.hadoop.mapred.MapTask.run(MapTask.java:372)
        at org.apache.hadoop.mapred.Child$4.run(Child.java:261)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1059)
        at org.apache.hadoop.mapred.Child.main(Child.java:255)  

Because I dont have code on the path, is that I am doing anything incorrect?

Thank you

Answer 1

With Hadoop Streaming, the code/dependencies have to be copied with the -file flag, if the code is not there on the target machine. Make sure that the map/reduce files and their dependencies are specified in the Hadoop streaming command.

$HADOOP_HOME/bin/hadoop  jar $HADOOP_HOME/hadoop-streaming.jar \
    -input myInputDirs \
    -output myOutputDir \
    -mapper myPythonScript.py \
    -reducer /bin/wc \
    -file myPythonScript.py \
    -file myDictionary.txt \