This is the documentation for Cloudera Manager 5.0.x. Documentation for other versions is available at Cloudera Documentation.

Decommissioning a Host

Decommissioning a host lets you decommission all roles on a single host without having to go to each service and decommission the roles individually. Once all roles on the host have been decommissioned and stopped, the host can be removed from service.

Decommissioning applies to only to HDFS DataNode, MapReduce TaskTracker, YARN NodeManager, and HBase RegionServer roles. If the host you select has other roles running on it, those roles will simply be stopped.

  Note: You cannot decommission a host if it results in fewer than three DataNode role instances.

Host decommissioning supports decommissioning multiple hosts in parallel.

Decommissioning One or More Hosts

If you are planning to decommission a large number of hosts, see the tuning recommendations.

  1. Click the Hosts tab.
  2. Select the host(s) you want to decommission.
  3. From the Actions for Selected menu, click Decommission.

A confirmation pop-up informs you of the roles that will be decommissioned or stopped on the hosts you have selected. To proceed with the decommissioning, click Confirm.

A Command Details window appears that will show each stop or decommission command as it is run, service by service. You can click one of the decommission links to see the subcommands that are run for decommissioning a given role. Depending on the role, the steps may include adding the host to an "exclusions list" and refreshing the NameNode, JobTracker, or NodeManager, stopping the Balancer (if it is running), and moving data blocks or regions. Roles that do not have specific decommission actions are just stopped.

While decommissioning is in progress, the host is marked Decommissioning in the list under the Hosts tab. Once all roles have been decommissioned or stopped, the host is marked Decommissioned.

Roles on a decommissioned host cannot be restarted until the host is recommissioned.

Tuning Prior to Decommissioning a Large Number of Hosts

If you are going to decommission a large number of hosts in batches, there is some tuning you should do first to ensure the decommissioning will go smoothly, since replicating a can be a very time-consuming process.
  1. Raise the heap size of the DataNodes. DataNodes should be configured with at least 4 GB heap to allow for the increase in iterations and max streams.
    1. Go to the HDFS service page (select the HDFS service from the Clusters tab).
    2. Select Configuration > View and Edit.
    3. Under each DataNode role group (The DataNode Default Group and any additional DataNode role groups) go to the Resource Management category, and set the Java Heap Size of DataNode in Bytes property as recommended.
  2. Increase the replication work multiplier to a larger number (the default is 2, however 10 is recommended):
    1. Go to the NameNode Default Group > Advanced category.
    2. Search for "dfs.namenode.replication" and increase the dfs.namenode.replication.work.multiplier.per.iteration property to a value such as 10.
    3. Click Save Changes.
  3. Set the replication maximum streams to a value such as 100.
    1. Go to the Service-Wide > Advanced category.
    2. In the HDFS Service Configuration Advanced Configuration Snippet for hdfs-site.xml, paste in the following:
      <property>
        <name>dfs.namenode.replication.max-streams</name>
        <value>100</value>
      </property
    3. Click Save Changes.
  4. Restart the HDFS service.

Recommissioning a Host

Only hosts that are decommissioned using Cloudera Manager can be recommissioned.

To recommission one or more hosts:

  1. Click the Hosts tab.
  2. Select the host(s) you want to recommission.
  3. From the Actions for Selected menu, click recommission.

This will recommission the host (that is, remove it from the exclusion lists and run the appropriate refresh) so that the roles that reside on it can be restarted. The Decommissioned indicator is removed from the host. It also removes the Decommissioned indicator from the roles that reside on the host. However, the roles themselves are not restarted by the recommission command.

You can restart all the roles on a recommissioned host in a single command from the Hosts page:

  1. Select the host(s) on which you want to start the decommissioned roles.
  2. From the Actions for Selected menu, click Start All Roles.

This will start all the roles on the selected host.

Page generated September 3, 2015.