HDP on Linux – Installation Forum

Automatically starting services

  • #27050
    finance turd

    I am running Horonworks/Ambari on a single VM for development purposes. We have an automated install for our multi-node production environment.

    I have added to /etc/rc.local:
    ambari-server restart
    ambari-agent restart

    And they both start up after system reboot… however…

    None of the Hadoop services (HDFS, Namenode, Oozie, etc.) are started.

    I have to go into Ambari and click on each one, and that’s really a pain.

    I tried to look to see what the command is to start these myself, but it’s not self-evident, at least not to me.

    Could you please tell me what commands to add to execute on system start-up to bring up not only ambari but also the Hadoop services that Ambari manages?

    Thank you.


to create new topics or reply. | New User Registration

  • Author
  • #27104

    Hi Finance,

    You can add these commands to your start/stop scripts:

    Stop Services

    curl –user admin:$PASSWORD -i -X PUT -d ‘{“ServiceInfo”: {“state” : “INSTALLED”}}’ http://$HOST:8080/api/v1/clusters/$CLUSTER_NAME/services/$SERVICE

    Start Services

    curl –user admin:$PASSWORD -i -X PUT -d ‘{“ServiceInfo”: {“state” : “STARTED”}}’ http://$HOST:8080/api/v1/clusters/$CLUSTER_NAME/services/$SERVICE

    You will need to set the following environment variables in your script:
    PASSWORD = the password for the Ambari admin user
    CLUSTER_NAME = the name you gave to your cluster
    SERVICE = one of (HDFS, MAPREDUCE, WEBHCAT, ZOOKEEPER, OOZIE, HBASE, HCATALOG, HIVE, GANGLIA, NAGIOS) – this is the order that is best to start the services, reverse it for stopping them.


You must be to reply to this topic. | Create Account

Support from the Experts

A HDP Support Subscription connects you experts with deep experience running Apache Hadoop in production, at-scale on the most demanding workloads.

Enterprise Support »

Become HDP Certified

Real world training designed by the core architects of Hadoop. Scenario-based training courses are available in-classroom or online from anywhere in the world

Training »

Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.