running fpg algorithm of mahout on hadoop as cluster mod - hadoop

Question

I install mahout-0.7 and hadoop-1.2.1 on linux (centos).hadoop config as multi_node.
I created a user named hadoop and install mahout and hadoop in path /home/hadoop/opt/
I set MAHOU_HOME and HADOOP_HOME and MAHOUT_LOCAL , .... in .bashrc file in the user's environment hadoop
# .bashrc
# Source global definitions
if [ -f /etc/bashrc ]; then
. /etc/bashrc
fi
# User specific aliases and functions
export JAVA_HOME=/usr/lib/jvm/java-1.7.0-openjdk-1.7.0.71/jre
export HADOOP_HOME=/home/hadoop/opt/hadoop
export PATH=$PATH:$HADOOP_HOME/bin
export HADOOP_CONF_DIR=/opt/hadoop/conf
export MAHOUT_HOME=/home/hadoop/opt/mahout
export MAHOUT_CONF_DIR=$MAHOUT_HOME/conf
export PATH=$PATH:$MAHOUT_HOME/bin
I want to run mahout on hadoop systemfile ,When I run the following command, I get an error
command:
hadoop#master mahout$ bin/mahout fpg -i /home/hadoop/output.dat -o patterns -method mapreduce -k 50 -s 2
error:
MAHOUT_LOCAL is not set; adding HADOOP_CONF_DIR to classpath.
hadoop binary is not in PATH,HADOOP_HOME/bin,HADOOP_PREFIX/bin, running locally
Error occurred during initialization of VM
Could not reserve enough space for object heap
Error: Could not create the Java Virtual Machine.
Error: A fatal exception has occurred. Program will exit.
Please help me. I tried but could not fix the error.

Answer 1

It seems that there are some conflicts in your configurations and usage.
In the fist look you can make sure about these:
To make sure that you've set the Mahout path correctly use this command:
echo $MAHOUT_LOCAL
This should not return an empty string (when you run mahout locally)
Also HADOOP_CONF_DIR should be set to $HADOOP_HOME/conf
Here's a list of popular environment variables for Hadoop:
#HADOOP VARIABLES START
export JAVA_HOME=/path/to/jdk1.8.0/ #your jdk path
export HADOOP_HOME=/usr/local/hadoop #your hadoop path
export HADOOP_INSTALL=/usr/local/hadoop #your hadoop path
export PATH=$PATH:$HADOOP_INSTALL/bin
export PATH=$PATH:$HADOOP_INSTALL/sbin
export HADOOP_MAPRED_HOME=$HADOOP_INSTALL
export HADOOP_COMMON_HOME=$HADOOP_INSTALL
export HADOOP_HDFS_HOME=$HADOOP_INSTALL
export YARN_HOME=$HADOOP_INSTALL
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_INSTALL/lib/native
export HADOOP_OPTS="-Djava.library.path=$HADOOP_INSTALL/lib"
export HADOOP_CLASSPATH=/home/hduser/lib/* #thir party libraries to be loaded with Hadoop
#HADOOP VARIABLES END
Also you get a heap error and you should increase your heap size so JVM will be enable to initialize
Also you may help to solve your error by adding more info about your cluster:
how many machine are you using?
what is the hardware spec of these machines?
what distribution and version of Hadoop?