Follow these steps:

1. Choose a cluster host to be the gateway host.
2. Following the instructions in Adding a Service or Adding a Role Instance, add the roles and services that you want to be available. For example, you can add the HttpFS role to the HDFS service, or add the Oozie service to your cluster.
3. If you have not already done so, add the Oozie service to your cluster, following the instructions in Adding a Service. Assign the Oozie Server role to your selected gateway host.
4. Configure your firewall to block access to all cluster nodes from outside the cluster, with the exception of the gateway host. Open the ports to the gateway host for the services you want to make available. In general, Cloudera recommends that you only allow access to the gateway host from clients that require access. For more information on the ports used by Cloudera software, see Ports.
5. If you have not already done so, enable security for your cluster, including Kerberos authentication, Apache Sentry authorization, and encryption for data at rest or in transit. For more information, see Cloudera Security.

With the Hadoop client:

All of the standard hadoop fs commands work; just make sure to specify -fs webhdfs://HOSTNAME:14000. For example (where GATEWAYHOST is the hostname of the gateway machine):

$ hadoop fs -fs webhdfs://GATEWAYHOST:14000 -cat /user/me/myfile.txt
Hello World!

Without the Hadoop client:

You can run all of the standard hadoop fs commands by using the WebHDFS REST API and any program that can do GET , PUT, POST, and DELETE requests; for example:

$ curl "http://GATEWAYHOST:14000/webhdfs/v1/user/me/myfile.txt?op=OPEN&user.name=me"
Hello World!

  Important: The user.name parameter is valid only if security is disabled. In a secure cluster, you must a initiate a valid Kerberos session.

In general, the command will look like this:

curl "http://GATEWAYHOST/webhdfs/v1/PATH?[user.name=USER&]op=…"

You can find a full explanation of the commands in the WebHDFS REST API documentation.

The Oozie REST API supports the direct submission of jobs for MapReduce, Pig, and Hive; Oozie automatically creates a workflow with a single action. For any other action types, or to execute anything more complicated than a single job, you must create an actual workflow. Required files (JAR files, input data) must already exist on HDFS; if they do not, you can use HttpFS to upload the files.

With the Oozie client:

All of the standard Oozie commands will work. You can find a full explanation of the commands in the documentation for the command-line utilities.

Without the Oozie client:

You can run all of the standard Oozie commands by using the REST API and any program that can do GET, PUT, and POST requests. You can find a full explanation of the commands in the Oozie Web Services API documentation.