Skip to content

Hadoop: installation and configuration#

This guide proposes an example of installation and configuration of Apache Hadoop.

Installation#

  • Download hadoop distribution from https://hadoop.apache.org/releases.html and uncompress.
cd /tmp
wget https://downloads.apache.org/hadoop/common/hadoop-3.1.3/hadoop-3.1.3.tar.gz
cd /opt
tar zxf /tmp/hadoop-3.1.3.tar.gz
ln -s hadoop-3.1.3 hadoop
  • Create a user and update permissions
useradd hadoop
chown hadoop:root -R /opt/hadoop*
  • Create a datastore and set permissions
mkdir /opt/thp/thehive/hdfs
chown hadoop:root -R /opt/thp/thehive/hdfs
  • Create ssh keys for hadoop user:
su - hadoop
ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
chmod 0600 ~/.ssh/authorized_keys
  • Update .bashrcfile for hadoop user. Add following lines:
export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64
export HADOOP_HOME=/opt/hadoop
export PATH=$PATH:$HADOOP_HOME/bin
export PATH=$PATH:$HADOOP_HOME/sbin
export HADOOP_MAPRED_HOME=$HADOOP_HOME
export HADOOP_COMMON_HOME=$HADOOP_HOME
export HADOOP_HDFS_HOME=$HADOOP_HOME
export YARN_HOME=$HADOOP_HOME

Note: Apache has a well detailed documentation for more advanced configuration with Hadoop.

Configuration the Hadoop Master#

Configuration files are located in etc/hadoop (/opt/hadoop/etc/hadoop). They must be identical in all nodes.

Notes:

  • The configuration described there is for a single node server. This node is the master node, namenode and datanode (refer to Hadoop documentation for more information). After validating this node is running successfully, refer to the related guide to add nodes;
  • Ensure you update the port value to something different than 9000 as it is already reserved for TheHive application service;

  • Edit the file core-site.xml:

<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
  <property>
    <name>fs.defaultFS</name>
    <value>hdfs://thehive1:10000</value>
  </property>
  <property>
    <name>hadoop.tmp.dir</name>
    <value>/opt/thp/thehive/hdfs/temp</value>
  </property>
  <property>
    <name>dfs.client.block.write.replace-datanode-on-failure.best-effort</name>
    <value>true</value>
  </property>
</configuration>
  • Edit the file hdfs-site.xml
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
  <property>
    <name>dfs.replication</name>
    <value>2</value>
  </property>
  <property>
    <name>dfs.namenode.name.dir</name>
    <value>/opt/thp/thehive/hdfs/namenode/data</value>
  </property>
  <property>
    <name>dfs.datanode.name.dir</name>
    <value>/opt/thp/thehive/hdfs/datanode/data</value>
  </property>
  <property>
    <name>dfs.namenode.checkpoint.dir</name>
    <value>/opt/thp/thehive/hdfs/checkpoint</value>
  </property>
  <property>
    <name>dfs.namenode.http-address</name>
    <value>0.0.0.0:9870</value>
  </property>
  <!--
  <property>
    <name>dfs.client.block.write.replace-datanode-on-failure.best-effort</name>
    <value>true</value>
  </property>
-->
  <property>
    <name>dfs.client.block.write.replace-datanode-on-failure.policy</name>
    <value>NEVER</value>
  </property>
</configuration>

Format the volume and start services#

  • Format the volume
su - hadoop
cd /opt/hadoop
bin/hdfs namenode -format

Run it as a service#


Create the /etc/systemd/system/hadoop.service file with the following content:

[Unit]
Description=Hadoop
Documentation=https://hadoop.apache.org/docs/current/index.html
Wants=network-online.target
After=network-online.target

[Service]
WorkingDirectory=/opt/hadoop
Type=forking

User=hadoop
Group=hadoop
Environment=JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64
Environment=HADOOP_HOME=/opt/hadoop
Environment=YARN_HOME=/opt/hadoop
Environment=HADOOP_COMMON_HOME=/opt/hadoop
Environment=HADOOP_HDFS_HOME=/opt/hadoop
Environment=HADOOP_MAPRED_HOME=/opt/hadoop
Restart=on-failure

TimeoutStartSec=2min


ExecStart=/opt/hadoop/sbin/start-all.sh
ExecStop=/opt/hadoop/sbin/stop-all.sh

StandardOutput=null
StandardError=null

# Specifies the maximum file descriptor number that can be opened by this process
LimitNOFILE=65536

# Disable timeout logic and wait until process is stopped
TimeoutStopSec=0

# SIGTERM signal is used to stop the Java process
KillSignal=SIGTERM

# Java process is never killed
SendSIGKILL=no

[Install]
WantedBy=multi-user.target

Start the service#

service hadoop start

You can check cluster status in http://thehive1:9870

Add nodes#

To add Hadoop nodes, refer the the related guide.


Last update: March 3, 2021 13:33:59