piccolbo
commented
11 years ago Please see the debugging guide in the RHadoop wiki. Unfortunately if you run your first program in distributed mode and just send me the console output, there isn't a lot for me to work on there. A screenshot has the only advantage of being harder to read, I hope people stop doing that. There's nothing I can do about it, there are multiple logs that are local to the machines that are executing the different processes. That's the way Hadoop is and there's probably good reasons why but whether or not, we need to work with it. Which means you either switch to standalone mode (see Cloudera documentation for how to do that) or go and fetch the stderr logs from the web UI. And while in this case there is nothing wrong with your program (works for me), these are skills that you need anyway to debug your own programs, so learning your way around the different modes and logs is a necessary investment.
raghavi222
i've posted the screen shot of my error. i'm using jdk 1.6
HADOOP_HOME=/usr/lib/hadoop HADOOP_CMD=/usr/lib/hadoop/bin/hadoop HADOOP_STREAMING=/usr/lib/hadoop-0.20-mapreduce/contrib/streaming/hadoop-streaming-2.0.0-mr1-cdh4.2.0.jar JAVA_HOME=/usr/java/jdk1.7.0_21 i've set the above environment variables in Renviron and .bashrc file.. i'm using cdh4 hadoop..