Hadoop YARN ResourceManager Start

Start the ResourceManager server. You can also start the server manually with the following two commands:

service hadoop-yarn-resourcemanager start
su -l yarn -c "export HADOOP_LIBEXEC_DIR=/usr/hdp/current/hadoop-client/libexec && /usr/lib/hadoop-yarn/sbin/yarn-daemon.sh --config /etc/hadoop/conf start resourcemanager"

module.exports = header: 'YARN RM Start', retry: 3, handler: ({options}) ->

Wait

Wait for Kerberos, Zookeeper, HDFS, YARN and the MapReduce History Server. The History Server must be started because the ResourceManager will try to load the history of MR jobs from there.

  @call 'masson/core/krb5_client/wait', once: true, options.wait_krb5_client
  @call 'ryba/zookeeper/server/wait', once: true, options.wait_zookeeper_server
  @call 'ryba/hadoop/hdfs_dn/wait', once: true, options.wait_hdfs_dn
  if options.wait_yarn_ts
    @call 'ryba/hadoop/yarn_ts/wait', once: true, options.wait_yarn_ts
  @call 'ryba/hadoop/mapred_jhs/wait', once: true, options.wait_mapred_jhs

Cleanup

Ensure the service pid is removed on retry.

  @system.remove
    target: "#{options.pid_dir}/yarn-#{options.user.name}-resourcemanager.pid"
    if: options.attempt > 0

Run

Start the service.

  @service.start
    name: 'hadoop-yarn-resourcemanager'