1. Download the latest stable release of Hive from the following location, using the command:

wget http://www-us.apache.org/dist/hive/hive-2.1.1/apache-hive-2.1.1-bin.tar.gz

2. Change to a user directory and extract the contents of the tar using the following command:

tar -xzvf ${DOWNLOAD_DIR}/apache-hive-2.1.1-bin.tar.gz

3. Configure and export the environment variable ${HIVE_HOME} pointing to the extracted directory and append its binaries to the path. Append the same to ~/.bashrc file

export HIVE_HOME=<Hive directory>
export PATH=$PATH:$HIVE_HOME/bin

4. Install latest SASL (Simple Authentication and Security Layer) packages for your operating system as Hive has dependency on this. For CentOS, these can be installed with following command:

sudo yum install *sasl*

5. Hive provides services which needs a metadata store for managing metadata information. Let us configure PostgreSQL server as the metadata server using the following commands using psql (interactive terminal for working with PostgreSQL) client in shell/Command Prompt.:
   • Create postgresql user as hiveuser and the database as metastore with the following commands:

sudo -u postgres psql

The previous command initializes and starts the psql client (the shell would show postgres=#) for running queries. Now let us create the user and database for Hive metastore with following queries:

 postgres=# CREATE USER hiveuser WITH PASSWORD 'mypassword';
 postgres=# CREATE DATABASE metastore;

postgres=# c metastore
metastore=# pset tuples_only on
metastore=# o /tmp/grant-privs
metastore=#   SELECT 'GRANT SELECT,INSERT,UPDATE,DELETE ON "'  || schemaname || '". "' ||tablename ||'" TO hiveuser ;'
metastore-#   FROM pg_tables
metastore-#   WHERE tableowner = CURRENT_USER and schemaname = 'public';
metastore=# o
metastore=# pset tuples_only off
metastore=# i /tmp/grant-privs

cp ${HIVE_HOME}/conf/hive-default.xml.template ${HIVE_HOME}/conf/hive-site.xml

wget https://jdbc.postgresql.org/download/postgresql-9.4.1212.jre6.jar

6. Configure the ${HADOOP_HOME}/etc/hadoop/core-site.xml with the following entries. Change hadoop.proxyuser.centos to your user account, instead of centos, in the following sample configurations:

<property>
   <name>hadoop.proxyuser.centos.hosts</name>
   <value>*</value>
</property>
<property>
   <name>hadoop.proxyuser.centos.groups</name>
   <value>*</value>
</property>

7. Restart the DFS service using the following commands:

stop-dfs.sh
start-dfs.sh

8. Configure ${HUE_HOME}/desktop/conf/hue.ini to make Hue work with the Hive service with following properties. Search for this property in hue.ini and change the CentOS to your user account name:

hive_conf_dir=/home/centos/apache-hive-2.1.1-bin/conf

9. Restart the hue service by gracefully stopping (find the supervisor process and then kill it) and starting it with following command:

${HUE_HOME}/build/env/bin/supervisor -d

10. Use the schematool to generate schema with the following command:

${HIVE_HOME}/bin/schematool -dbType postgres -initSchema --verbose

Go to pgAdmin and you should see a new database namely metastore containing the various tables generated by the schematool

11. Verify the install by following steps:
    1. Check for hive shell by running the given command:

${HIVE_HOME}/bin/hive

If things go well you should see the hive shell. You can run various Hive queries using this shell if needed.

${HIVE_HOME}/bin/hive --service hiveserver2 -hiveconf hive.root.logging=console

Now that we have all the required components installed and working, we will look at a few examples. The initial example would cover loading aspect of the data and as we proceed through these examples we will also see the processing aspects of Hadoop layer.