The Hortonworks Community Connection is now live. A completely rebuilt Q&A forum, Knowledge Base, Code Hub and more, backed by the experts in the industry.

You will be redirected here in 10 seconds. If your are not redirected, click here to visit the new site.

The legacy Hortonworks Forum is now closed. You can view a read-only version of the former site by clicking here. The site will be taken offline on January 31,2016

Oozie Forum

Issues with Oozie and HDP

  • #25459


    We’re trying to set up HDP with Oozie according to the docs at

    The following command fails, saying that it doesn’t support version 2.x

    /usr/lib/oozie/bin/ -hadoop 2.x /usr/lib/hadoop -extjs /usr/share/HDP-oozie/

    We worked around this issue and managed to install with the following command (changed to version to 0.20.200)

    /usr/lib/oozie/bin/ -hadoop 0.20.200 /usr/lib/hadoop -extjs /usr/share/HDP-oozie/

    then created folder /etc/oozie/conf/action-conf, changed the config as described in the docs and added a few more changes (mostly so oozie will work with postgresql as its db) and we managed to bring it up.

    Now when we’re trying to start a job and we get HadoopAccessorException. In the logs we see something weird – it’s trying to access namenodename/ipconfig:8021 (I changed the name and ip):

    Caused by: Call to failed on connection exception: Connection refused

    Any idea?

  • Author
  • #25498

    Hi Saggi,

    correct me if I’m wrong, but from the stack trace snippet it looks like you’re using the EC2 external names as the hostnames in your cluster. If that is the case that may be the root of the problem as it should be the internal name/ip for the cluster. the external name are used only when connecting to the nodes from the outside. the internal names usually begin with ‘ip-‘.



    Hi Tedr,

    We didn’t have any problem executing map-reduce jobs directly with the jobtracker and the external IP.

    I think that the problem begins with HDP’s bundled Oozie not being compatible with Hadoop (in the same distribution)


    found the problem – the documentation and the accompanying config files tell the jobtracker to use port 50300 as the IPC port which is not the default 8021 we were telling oozie to use.


    Hi Saggi,

    It’s good to hear that you’ve found the problem, does that mean that you’ve corrected for it as well, and are now up an running?



    Indeed – everything’s working. thanks!

    Larry Liu

    Thanks, Saggi. The example in the oozie package is still using incorrect port. Please correct it before running any example in the package.


The forum ‘Oozie’ is closed to new topics and replies.

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.