Configure the Hadoop cluster service

Configure the Hadoop cluster service

After you deploy your OpenStack environment with the Hadoop cluster service, upload the Sahara image to Glance. You must use a pre-built image for a corresponding Linux distribution. You can build your own image as described in the OpenStack Sahara documentation or use one of the pre-built images available at docs.openstack.org/developer/Sahara If you use a pre-built image, use the corresponding user name provided on the same page.

Sahara supports the following Hadoop platforms:

  • Vanilla Apache Hadoop
  • Hortonworks Data Platform (HDP)
  • Cloudera Hadoop Distribution (CDH)
  • Apache Spark
  • MapR

Typically, the Sahara images are provided in the qcow2 format that is compatible with the default OpenStack hypervisor KVM. If you install your environment with VMware vSphere, you must convert the image to an appropriate format before you upload the image to Glance.

To configure the Hadoop cluster service:

  1. Download or build an appropriate image for the Hadoop cluster.
  2. Upload the image to Glance.
  3. Log in to Horizon.
  4. Register the image in the Sahara Image Registry:
    1. Click Project ‣ Data processing ‣ Image Registry.
    2. Click Register Image.
    3. Specify the username appropriate to the image you use.
    4. Specify a correpsonding plugin and version.
    5. Click Add plugin tags.
    6. Add appropriate tags.
  5. Verify that you have an adequate pool of floating IP addresses available:
    • If you run Neutron networking with auto_assign_floating_ip parameter set to False, provide a floating IP pool in each Node Group Template.
Creative Commons Attribution 3.0 License

Except where otherwise noted, this document is licensed under Creative Commons Attribution 3.0 License. See all OpenStack Legal Documents.

Contents