unable to run map reduce using python in Hadoop?

问题

I have written mapper and reducer in python for word count program that works fine. Here is a sample:

echo "hello hello world here hello here world here hello" | wordmapper.py | sort -k1,1 | wordreducer.py 
hello   4
here    3
world   2

Now when i try to submit a hadoop job for a large file, I get errors

hadoop jar share/hadoop/tools/sources/hadoop-*streaming*.jar -file wordmapper.py -mapper wordmapper.py  -file wordreducer.py -reducer wordreducer.py -input /data/1jrl.pdb -output /output/py_jrl
Exception in thread "main" java.lang.ClassNotFoundException: share.hadoop.tools.sources.hadoop-streaming-2.2.0-test-sources.jar
    at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
    at java.security.AccessController.doPrivileged(Native Method)
    at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
    at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
    at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
    at java.lang.Class.forName0(Native Method)
    at java.lang.Class.forName(Class.java:249)
    at org.apache.hadoop.util.RunJar.main(RunJar.java:205)

I removed changed the commandline to the following (removed wild card from above);

hadoop jar share/hadoop/tools/sources/hadoop-streaming-2.2.0-sources.jar -file wordmapper.py -mapper wordmapper.py  -file wordreducer.py -reducer wordreducer.py -input /data/1jrl.pdb -output /output/py_jrl
Exception in thread "main" java.lang.ClassNotFoundException: -file
    at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
    at java.security.AccessController.doPrivileged(Native Method)
    at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
    at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
    at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
    at java.lang.Class.forName0(Native Method)
    at java.lang.Class.forName(Class.java:249)
    at org.apache.hadoop.util.RunJar.main(RunJar.java:205)

why I get these errors and how to fix this? I use hadoop2. Thanks!

回答1:

Well at least one of your issues is that you are using the -sources.jar which is just .java files and can't be executed.

Try using this instead...

share/hadoop/tools/lib/hadoop-streaming-2.2.0.jar

And if that doesn't exist, look for a hadoop-streaming*.jar that doesn't have -sources in the file name.

来源：https://stackoverflow.com/questions/24661653/unable-to-run-map-reduce-using-python-in-hadoop

标签

python

Hadoop

MapReduce

hadoop2