Installation
------------

HDFS FDW Installation from the source code:

To compile the Hadoop foreign data wrapper, we need Java Compiler,
JDK Include files and the JVM Library.
We have used jdk-8u111-linux-x64.tar.gz.

- Steps to compile hiveclient

For C/C++ part:

    export JDK_INCLUDE:=$JAVA_HOME/include
    export INSTALL_DIR:=/path/to/ppas/install/dir/lib
    export PATH:=/path/to/ppas/install/dir/bin:$PATH

    Make sure that pg_config is now in path.

    cd /path/to/hdfs_fdw/libhive
    make
    make install

For Java part:

    cd /path/to/hdfs_fdw/libhive/jdbc
    javac MsgBuf.java
    javac HiveJdbcClient.java
    jar cf HiveJdbcClient-1.0.jar *.class
    cp HiveJdbcClient-1.0.jar /path/to/install/folder/lib/postgresql/
  
Now that all the required dependencies are built we can build and
install the fdw.


- Steps to compile the HDFS FDW source code:

To build on POSIX-compliant systems you need to ensure the `pg_config`
executable is in your path when you run `make`. This executable is
typically in your PostgreSQL installation's `bin` directory. For
example:

    export PATH=/usr/local/pgsql/bin/:$PATH

    make USE_PGXS=1
    make USE_PGXS=1 install

Please note that the HDFS_FDW extension has only been tested on Ubuntu
and CentOS systems but it should work on other *UNIX's systems without
any problems.


* How To Start Hadoop.
The detail installation instruction of Hadoop can be found on this
[site][1]. Here are the steps to start and stop the Hadoop:

* Stop and start Hdfs on Single Node
    # $HADOOP_HOME/sbin/stop-dfs.sh
    # $HADOOP_HOME/sbin/start-dfs.sh

* YARN on Single Node
    # $HADOOP_HOME/sbin/stop-yarn.sh
    # $HADOOP_HOME/sbin/start-yarn.sh

* Starting HiveServer2
  $HIVE_HOME/bin/hiveserver2
  or
  $HIVE_HOME/bin/hive --service hiveserver2


- Steps to execute the regression:

1. Start the Hive server if not already running.

2. Load sample data for the test cases by using the following command.

    `hdfs_fdw/test/insert_hive.sh`

3. In the terminal where regression will be executed, set following
   environment variables using export command.

    HIVE_SERVER
    HIVE_PORT
    HIVE_USER  (LDAP User when HiveServer running with LDAP Mode, empty when HiveServer running with NOSASL Mode)
    HIVE_PASSWORD
    AUTH_TYPE (LDAP when HiveServer running with LDAP Mode, NOSASL when HiveServer running with NOSASL Mode)
    CLIENT_TYPE (hiveserver2 or spark)

    e.g

    export HIVE_SERVER='127.0.0.1'
    export HIVE_PORT='10000'
    export HIVE_USER='hive_user'
    export HIVE_PASSWORD='hive_password'
    export AUTH_TYPE='LDAP'
    export CLIENT_TYPE='hiveserver2'

4. Execute the regression using the following command.
  
    `hdfs_fdw/make installcheck`


[1]: http://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-common/SingleCluster.html