Install Hadoop.sh

# Verifies if Java is installed, and if it's not installed, install it.
function isJavaInstalled {
	echo -ne "=> Verifying Java installation"
	
	# Trying to execute java
	java -version
	
	# Getting the exit code of the execution.
  	EXITCODE="$?"
  	echo -ne "... "
  	
  	# If the code is different from zero, it is not installed
	if [ "$EXITCODE" != "0" ];
	then
		echo -ne "Java is not installed :(.\n"
		
		# Proceed to install java
		installJava
	else
		echo -ne "Java is installed :).\n"
	fi
}

# Java Ubuntu install
function installJava {
	
	# Clone my another gist with Java installation for Ubuntu
	git clone https://gist.github.com/7128552e33c6c4c6ab51.git
	cd 7128552e33c6c4c6ab51 || exit
	
	# Give it permissions and execute it
	chmod +x Install\ Java.sh
	./Install\ Java.sh
}

# Installs Hadoop
function installHadoop {
	# Stand in /opt/ directory
	cd /opt/ || exit
	
	echo -ne "(1/4) => Downloading Hadoop"
	sudo wget http://www.eu.apache.org/dist/hadoop/common/current/hadoop-2.7.1.tar.gz
	echo -ne "... :) \n"
	
	echo -ne "=> Decompressing Hadoop..."
	sudo tar -xvzf hadoop-2.7.1.tar.gz
	echo -ne "... :) \n"
	
	echo -ne "=> Creating a symbolic link to hadoop"
	sudo ln -s hadoop-2.7.1 hadoop
	echo -ne "... :) \n"
	
	echo -ne "=> Adding permissions..."
	sudo chmod -R +rx hadoop
	echo -ne "... :) \n"
}

# Create Hadoop user
function createUser {
	echo -ne "=> Creating Hadoop user..."
	sudo useradd -d /home/hadoop -m hadoop
	echo -ne "... :) \n"
	
	echo "Enter the new password for the Hadoop user. (It will have sudo's permissions)"
	sudo passwd hadoop
	usermod -a -G sudo hadoop
	usermod -s /bin/bash hadoop
	echo "Log with the new Hadoop password here. It's now adding the key and testing the hadoop user."
	sudo -u hadoop exit
	
	echo -ne "=> Adding environment variables to ~/.bashrc"
	{
		echo "export HADOOP_HOME=/usr/local/hadoop"
		echo "export PATH=\$PATH:\$HADOOP_HOME/bin"
		echo "export PATH=\$PATH:\$HADOOP_HOME/sbin"
		echo "export HADOOP_MAPRED_HOME=\${HADOOP_HOME}"
		echo "export HADOOP_COMMON_HOME=\${HADOOP_HOME}"
		echo "export HADOOP_HDFS_HOME=\${HADOOP_HOME}"
		echo "export YARN_HOME=\${HADOOP_HOME}"
	} >> /home/hadoop/.bashrc
	echo -ne "... :) \n"
}

# Adds network settings for Hadoop
function networkSettings {
	
	echo "=> Installing ssh..."
	sudo apt-get install -y ssh
	
	echo "=> Generating public key for hadoop user"
	sudo -u hadoop ssh-keygen -t rsa -f ~/.ssh/id_rsa
	
	echo "=> Adding localhost public key to authorized keys"
	cat ~/.ssh/id_rsa.pub | sudo -u hadoop tee -a ~/.ssh/authorized_keys > /dev/null
	
	echo "=> Updating ssh files permissions..."
	sudo -u hadoop sudo chmod go-w "$HOME" "$HOME/.ssh"
	sudo -u hadoop chmod 600 "$HOME"/.ssh/authorized_keys
	sudo -u hadoop chown "$(whoami)" "$HOME/.ssh/authorized_keys"
	
	echo "=> Disabling IPv6"
	{
		echo "net.ipv6.conf.all.disable_ipv6 = 1"
		echo "net.ipv6.conf.default.disable_ipv6 = 1"
		echo "net.ipv6.conf.lo.disable_ipv6 = 1"
	} >> /etc/sysctl.conf
}

# Edit Hadoop single host configuration
function editMainConfig {
  echo "Editing the config files under /opt/hadoop/etc/hadoop/"
  cd /opt/hadoop/etc/hadoop/ || exit
  echo -ne "[1/] core-site.xml... "
  {
    echo "<?xml version=\"1.0\"?>"
    echo "<?xml-stylesheet type=\"text/xsl\" href=\"configuration.xsl\"?>"
    echo "<configuration>"
    echo "  <property>"
    echo "    <name>fs.default.name</name>"
	  echo "    <value>hdfs://localhost:8020</value>"
    echo "    <description>Nombre del filesystem por defecto.</description>"
    echo "  </property>"
    echo "</configuration>"
  } > core-site.xml
  echo -ne " OK\n"
  
  echo -ne "[2/] hdfs-site.xml... "
  {
    echo "<?xml version=\"1.0\"?>"
    echo "<?xml-stylesheet type=\"text/xsl\" href=\"configuration.xsl\"?>"
    echo "<configuration>"
    echo "  <property>"
    echo "    <name>dfs.namenode.name.dir</name>"
    echo "    <value>file:/home/hadoop/workspace/dfs/name</value>"
    echo "    <description>Path del filesystem donde el namenode almacenará los metadatos.</description>"
    echo "  </property>"
    echo "  <property>"
    echo "    <name>dfs.datanode.data.dir</name>"
    echo "    <value>file:/home/hadoop/workspace/dfs/data</value>"
    echo "    <description>Path del filesystem donde el datanode almacenará los bloques.</description>"
    echo "  </property>"
    echo "  <property>"
    echo "    <name>dfs.replication</name>"
    echo "    <value>1</value>"
    echo "    <description>Factor de replicación. Lo ponemos a 1 porque sólo tenemos 1 máquina.</description>"
    echo "  </property>"
    echo "</configuration>"
  } > hdfs-site.xml
  echo -ne "OK\n"
  
  echo -ne "[3/] Creating hdfs directories /home/hadoop/workspace/dfs/name... "
  sudo -u hadoop mkdir -p /home/hadoop/workspace/dfs/name
  sudo -u hadoop mkdir -p /home/hadoop/workspace/dfs/data
  echo -ne "OK \n"
  
  echo -ne "[4/] Copying the mapred-site.xml template file... "
  sudo cp mapred-site.xml.template mapred-site.xml
  echo -ne "OK \n"
  
  {
    echo "<configuration>"
    echo "<property>"
    echo "  <name>yarn.nodemanager.aux-services</name>"
    echo "  <value>mapreduce_shuffle</value>"
    echo "</property>"
    echo "<property>"
    echo "  <name>yarn.nodemanager.aux-services.mapreduce_shuffle.class</name>"
    echo "  <value>org.apache.hadoop.mapred.ShuffleHandler</value>"
    echo "</property>"
    echo "</configuration>"
  } >> yarn-site.xml
}

# Verify if Java is installed
isJavaInstalled

# Install Hadoop
installHadoop

sudo su 

# User creation
createUser

# Network settings
networkSettings

# Config files
editMainConfig

# Format hdfs
sudo hdfs namenode -format

echo "Installation complete, after you log in with hadoop user, run Hadoop using:"
echo "start-dfs.sh & star-yarn.sh"
echo ""