Install, setup Cloudera Hadoop on Linux

Install Cloudera Hadoop on Linux

This article talks about installing Hadoop on a single host machine.

Hadoop is the framework for a large amount of data processing paralleled

Hadoop implementation is provided by different vendors like hortionworks and Cloudera.

This article talks about installing Cloudera Hadoop on a single machine.

To set up Cloudera Hadoop, java is required.
if java is not already installed, install JDK 1.6, at least update 8

Please download Cloudera-testing. repo from http://archive.cloudera.com/redhat/cdh/ and copy it to /etc/yum.repos.d/ and make sure you update the yum command.

Please run the below commands to install hadoop, hive, and pig

yum install hadoop-0.20 -y  
  
yum install hadoop-hive -y  
  
yum install hadoop-pig -y  
  

The above commands installs hadoop to /usr/lib/hadoop folder, hive installs to /usr/lib/hive, pig to /usr/lib/pig

please set up the environment variables as described below in the .bash_rc file

$ \\vi ~/.bashrc  
  
  
export HADOOP\_HOME=/usr/lib/hadoop  
  
export HIVE\_HOME=/usr/lib/hive  
  
export PIG\_HOME=/usr/lib/pig  
  
export PATH=$HADOOP\_HOME/bin:$PATH:$PIG\_HOME/bin:$HIVE\_HOME/bin  
  

```save it to .bashrc file  
  
$ source ~/.bashrc  
  
Open $HADOOP\_HOME/conf/hadoop-env.sh. Add JAVA\_HOME path. Ex:  
export JAVA\_HOME=/usr/java/jdk1.6.0\_18  
  
\* Open $HADOOP\_HOME/conf/core-site.XML. Add the Namenode server name or localhost and port for fs.default.name. Ex:  
  
fs.default.name  
hdfs://localhost:9000