MapReduce Forum

NameNode as datanode (error)

  • #55441
    David Hall
    Participant

    I’ve got a situation where the namenode is being used as a datanode and mappers are being run there. Not good. For some reason I HoClient.getActiveTrackerNames() is returning the Namenode as well as all datanodes. What configuration mess up could cause this?

    I’m trying to get Sqoop to run my mappers, exactly one per Datanode. I’ve written my InputFormat class to create splits with exactly one hostname (datanodes only) per split, yet am being ignored. Why is the Jobtracker ignoring my splits? The host names I’m submitting are exactly as getActiveTrackerNames() returns, sans the :nnnnn, as suggested in getActiveServersList() by Boris Lublinsky and Mike Segel. btw: this system works in that other Hadoop environment. This is an Amazon hosted system.

to create new topics or reply. | New User Registration

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.