Cisco Cisco Workload Automation 6.3 User Guide

      www.cisco.com

Configuring the Hive Adapter

Overview

The Hive Adapter software is installed as part of a standard installation of Cisco Workload Automation. However, before 
the Hive Adapter can be used, the following configuration procedures must be completed: 



 – Install the necessary Hadoop client libraries for Hive.



 – Add optional configuration properties to the service.props file.



 – Apply the license to the Hive Adapter. You cannot define a Hive connection until you 

have applied the Hive license. 



 – Define Hive users that the Adapter can use to establish authenticated sessions with 

the Hive server and permit requests to be made on behalf of the authenticated account. 



 – Define a Hive connection so the master can communicate with the Hive server.

See 

 for details about configuring service.props to control such things as polling, output, and 

log gathering.

Installing the Hadoop Client Libraries

Hadoop client libraries are required for processing the Hadoop-related DataMover, Hive, MapReduce, and Sqoop jobs. 
As of CWA 6.3, Hadoop libraries are not included with CWA. Instead, we provide a Maven script (POM.xml) to install the 
required libraries.

If you do not already have Maven, you must download and install it. Then you need to run the POM.xml script to download 
the required Hadoop client libraries. Instructions for obtaining Maven and downloading the Hadoop libraries are included 
in these sections:





 The instructions here are for Windows.

Installing Maven

If you do not have Maven installed, follow the instructions below. 

Maven Prerequisites


JDK must be installed.



The JAVA_HOME environment variable must be set and point to your JDK.