WebHCat Installation

WebHCat Installed with Hive

...

1. Ensure that the required related installations are in place, and place required files into the Hadoop distributed cache.
2. Download and unpack the HCatalog distribution.
3. Set the TEMPLETON_HOME environment variable to the base of the HCatalog REST server installation. This will usually be same as HCATALOG_HOME. This is used to find the WebHCat (Templeton) configuration.
4. Set JAVA_HOME, HADOOP_PREFIX, and HIVE_HOME environment variables.
5. Review the configuration and update or create webhcat-site.xml as required. Ensure that site-specific component installation locations are accurate, especially the Hadoop configuration path. Configuration variables that use a filesystem path try to have reasonable defaults, but it's always safe to specify a full and complete path.
6. Verify that HCatalog is installed and that the hcat executable is in the PATH.
7. Build HCatalog using the command ant jar from the top level HCatalog directory.
8. Start the REST server with the command "hcatalog/sbin/webhcat_server.sh start" for Hive 0.11.0 releases and later, or "sbin/webhcat_server.sh start" for installations prior to HCatalog merging with Hive.

9. Check that your local install works. Assuming that the server is running on port 50111, the following command would give output similar to that shown.

   No Format
   % curl -i http://localhost:50111/templeton/v1/status HTTP/1.1 200 OK Content-Type: application/json Transfer-Encoding: chunked Server: Jetty(7.6.0.v20120127) {"status":"ok","version":"v1"} %

   
   

...

The server requires some files be accessible on the Hadoop distributed cache. For example, to avoid the installation of Pig and Hive everywhere on the cluster, the server gathers a version of Pig or Hive from the Hadoop distributed cache whenever those resources are invoked. After placing the following components into HDFS please update the site configuration as required for each.

• Hive: Download the Hive tar.gz file and place it in HDFS. For example, for Hive version 0.11.0:

  No Format
  hadoop fs -put /tmp/hive-0.11.0.tar.gz /apps/templeton/hive-0.11.0.tar.gz

  
  

• Pig: Download the Pig tar.gz file and place it into HDFS. For example, for Pig version 0.11.1:

  No Format
  hadoop fs -put /tmp/pig-0.11.1.tar.gz /apps/templeton/pig-0.11.1.tar.gz

  
  

• Hadoop Streaming: Place hadoop-streaming-*.jar into HDFS. Use the following command:

  No Format
  hadoop fs -put <hadoop streaming jar> \ <templeton.streaming.jar>/hadoop-streaming-*.jar

  where <templeton.streaming.jar> is a property value defined in webhcat-default.xml which can be overridden in the webhcat-site.xml file, and <hadoop streaming jar> is the Hadoop streaming jar in your Hadoop version:

  • hadoop-1.*/contrib/streaming/hadoop-streaming-*.jar in the Hadoop 1.x tar

  • hadoop-2.*/share/hadoop/tools/lib/hadoop-streaming-*.jar in the Hadoop 2.x tar
    For example,

    No Format
    hadoop fs -put hadoop-2.1.0/share/hadoop/tools/lib/hadoop-streaming-2.1.0.jar \ /apps/templeton/hadoop-streaming.jar

    
    

• Override Jars: Place override jars required (if any) into HDFS. Note: Hadoop versions prior to 1.0.3 required a patch (HADOOP-7987) to properly run WebHCat. This patch is distributed with WebHCat (located at templeton/src/hadoop_temp_fix/ugi.jar) and should be placed into HDFS, as reflected in the current default configuration.

  No Format
  hadoop fs -put ugi.jar /apps/templeton/ugi.jar

  
  

...

Permissions

Permission must given be given for the user running the WebHCat executable to run jobs for other users. That is, the WebHCat server will impersonate users on the Hadoop cluster.

...

Also, set the WebHCat configuration variables templeton.kerberos.principal and templeton.kerberos.keytab.

Proxy User Support

Proxy User Support in WebHCat allows the caller of WebHCat to instruct WebHCat to run commands on the Hadoop cluster as a particular user.

The canonical example is Joe using Hue to submit a MapReduce job through WebHCat. For the following description, assume Joe has the Unix name 'joe', Hue is 'hue' and WebHCat is 'hcat'. If Hue specifies 'doAs=joe' when calling WebHCat, WebHCat submits the MR job as 'joe' so that the Hadoop cluster can perform securitiy checks with respect to 'joe'. If the doAs value is not specified, the MR job will be submitted as user 'hue'.

To set up Proxy User Support, make the following edits in configuration files.

In hive-site.xml, set:

In webhcat-site.xml, set:

In core-site.xml, make sure the following are also set: