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

YARN Forum

YARN Resource Manager failure situation

  • #46293
    Jakub Stransky


    I am relatively new to Hadoop 2.x and YARN but I have a question regarding fail-over of RM. What happens if resource manager fails? I understand that Application Masters are screwed as there is no-one who can grant resources for them or facilitate the communication. Does it make sense to run resource manager in a clustered fashion? Can someone point me to the right documentation if that possible? I am using HDP 2.0

    Thanks a lot

  • Author
  • #47614

    ResourceManager failover isn’t available in HDP 2.0, it is going to be soon in the subsequent releases.

    To your specific question, depending on whether RM can preserve work across restart or not, RM fail-over may or may not retain work. Work-preserving RM restart is a work in progress. So today, when RM fails over, all apps are restarted. For apps like MapReduce which implemented recovery, jobs will continue from their last saved progress.

The forum ‘YARN’ 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.