Problem configuring job tracker history (unable to access web ui

Tagged: 

This topic contains 2 replies, has 2 voices, and was last updated by  Michiel Van Oosterhout 1 year, 4 months ago.

  • Creator
    Topic
  • #42256

    I succesfully installed HDP for Windows on Windows Server 2012 R2 running in a Hyper-V virtual machine. The machine name of the guest virtual machine is “Hadoop”, and I used “hadoop” as the hostname in clusterproperties.txt. I also added “hadoop” to the hosts file on the host OS. I added firewall rules in the guest OS to access the Hadoop web UI from the host OS

    Almost everything seems to work well, e.g. I can browse http://hadoop:50070 on the host machine, and I see the web UI of the Hadoop service. Except for the job tracker history web UI (http://hadoop:51111/jobhistoryhome.jsp).

    When I change the configuration value “mapred.job.tracker.http.address” from “hadoop:51111″ to “0.0.0.0:51111″ in mapred-site.xml, and then restart the jobtracker service, I can access it. But after a restart I can no longer access it. But if I then change the configuration value back, and restart the service, I can access it again. And then again, after a system restart, I can no longer access it.

    How should I configure my single node system to be able to access the job history web UI.

Viewing 2 replies - 1 through 2 (of 2 total)

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

  • Author
    Replies
  • #42659

    Hello Seth,

    Thank you for your reply. I can access the Job Tracker UI (port 50030) in my browser. When I click the history link at the bottom of the screen, I can now access the history UI as well… So, it seems to work now, although I didn’t change anything since I last tried.

    Thanks again!

    Collapse
    #42641

    Seth Lyubich
    Keymaster

    Hi Michiel,

    Can you please try telneting to “hadoop” host on port 51111 once History Server UI is not working? This test should show you if there are any networking/name resolution issues.

    Also, can you try accessing History server from Job Tracker UI (http://hadoop:50030/jobtracker.jsp) ?

    Hope this helps,

    Thanks,
    Seth

    Collapse
Viewing 2 replies - 1 through 2 (of 2 total)
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.