Role Instances

Adding a Role Instance

Minimum Required Role: Cluster Administrator (also provided by Full Administrator)

After creating services, you can add role instances to the services. For example, after initial installation in which you created the HDFS service, you can add a DataNode role instance to a host where one was not previously running. Upon upgrading a cluster to a new version of CDH you might want to create a role instance for a role added in the new version. For example, in CDH 5 Impala has the Impala Llama ApplicationMaster role, which must be added after you upgrade a CDH 4 cluster to CDH 5.

  1. Go to the service for which you want to add a role instance. For example, to add a DataNode role instance, go to the HDFS service.
  2. Click the Instances tab.
  3. Click the Add Role Instances button.
  4. Customize the assignment of role instances to hosts. The wizard evaluates the hardware configurations of the hosts to determine the best hosts for each role. The wizard assigns all worker roles to the same set of hosts to which the HDFS DataNode role is assigned. These assignments are typically acceptable, but you can reassign them if necessary.

    Click a field below a role to display a dialog containing a list of hosts. If you click a field containing multiple hosts, you can also select All Hosts to assign the role to all hosts or Custom to display the pageable hosts dialog.

    The following shortcuts for specifying hostname patterns are supported:
    • Range of hostnames (without the domain portion)
      Range Definition Matching Hosts
      10.1.1.[1-4] 10.1.1.1, 10.1.1.2, 10.1.1.3, 10.1.1.4
      host[1-3].company.com host1.company.com, host2.company.com, host3.company.com
      host[07-10].company.com host07.company.com, host08.company.com, host09.company.com, host10.company.com
    • IP addresses
    • Rack name

    Click the View By Host button for an overview of the role assignment by hostname ranges.

  5. Click Continue.
  6. In the Review Changes page, review the configuration changes to be applied. Confirm the settings entered for file system paths. The file paths required vary based on the services to be installed. For example, you might confirm the NameNode Data Directory and the DataNode Data Directory for HDFS. Click Continue. The wizard finishes by performing any actions necessary to prepare the cluster for the new role instances. For example, new DataNodes are added to the NameNode dfs_hosts_allow.txt file. The new role instance is configured with the default role group for its role type, even if there are multiple role groups for the role type. If you want to use a different role group, follow the instructions in Managing Role Groups for moving role instances to a different role group. The new role instances are not started automatically.

Starting, Stopping, and Restarting Role Instances

Minimum Required Role: Operator (also provided by Configurator, Cluster Administrator, Full Administrator)

If the host for the role instance is currently decommissioned, you will not be able to start the role until the host has been recommissioned.
  1. Go to the service that contains the role instances to start, stop, or restart.
  2. Click the Instances tab.
  3. Check the checkboxes next to the role instances to start, stop, or restart (such as a DataNode instance).
  4. Select Actions for Selected > Start, Stop, or Restart, and then click Start, Stop, or Restart again to start the process. When you see a Finished status, the process has finished.
Also see Rolling Restart.

Decommissioning Role Instances

Minimum Required Role: Operator (also provided by Configurator, Cluster Administrator, Full Administrator)

You can remove a role instance such as a DataNode from a cluster while the cluster is running by decommissioning the role instance. When you decommission a role instance, Cloudera Manager performs a procedure so that you can safely retire a host without losing data. Role decommissioning applies to HDFS DataNode, MapReduce TaskTracker, YARN NodeManager, and HBase RegionServer roles.

You cannot decommission a DataNode or a host with a DataNode if the number of DataNodes equals the replication factor (which by default is three) of any file stored in HDFS. For example, if the replication factor of any file is three, and you have three DataNodes, you cannot decommission a DataNode or a host with a DataNode.

A role will be decommissioned if its host is decommissioned. See Decommissioning and Recommissioning Hosts for more details.

To decommission role instances:
  1. If you are decommissioning DataNodes, perform the steps in Tuning HDFS Prior to Decommissioning DataNodes.
  2. Click the service instance that contains the role instance you want to decommission.
  3. Click the Instances tab.
  4. Check the checkboxes next to the role instances to decommission.
  5. Select Actions for Selected > Decommission, and then click Decommission again to start the process. While decommissioning is in progress, the role instance displays the icon. If one role instance of a service is decommissioned, the DECOMMISSIONED facet displays in the Filters on the Instances page and the icon displays on the role instance's page. When you see a Finished status, the decommissioning process has finished.

Recommissioning Role Instances

Minimum Required Role: Operator (also provided by Configurator, Cluster Administrator, Full Administrator)

  1. Click the service instance that contains the role instance you want to recommission.
  2. Click the Instances tab.
  3. Check the checkboxes next to the decommissioned role instances to recommission.
  4. Select Actions for Selected > Recommission, and then click Recommission again to start the process. When you see a Finished status, the recommissioning process has finished.

Deleting Role Instances

Minimum Required Role: Cluster Administrator (also provided by Full Administrator)

  1. Click the service instance that contains the role instance you want to delete. For example, if you want to delete a DataNode role instance, click an HDFS service instance.
  2. Click the Instances tab.
  3. Check the checkboxes next to the role instances you want to delete.
  4. If the role instance is running, select Actions for Selected > Stop and click Stop to confirm the action.
  5. Select Actions for Selected > Delete. Click Delete to confirm the deletion.

Configuring Roles to Use a Custom Garbage Collection Parameter

Every Java-based role in Cloudera Manager has a configuration setting called Java Configuration Options for role where you can enter command line options. Commonly, garbage collection flags or extra debugging flags would be passed here. To find the appropriate configuration setting, select the service you want to modify in the Cloudera Manager Admin Console, then use the Search box to search for Java Configuration Options.

You can add configuration options for all instances of a given role by making this configuration change at the service level. For example, to modify the setting for all DataNodes, select the HDFS service, then modify the Java Configuration Options for DataNode setting.

To modify a configuration option for a given instance of a role, select the service, then select the particular role instance (for example, a specific DataNode). The configuration settings you modify will apply to the selected role instance only.

For detailed instructions see Modifying Configuration Properties.