Starting up HDFS

to create new topics or reply. | New User Registration

This topic contains 2 replies, has 2 voices, and was last updated by  Chris Rivadeneira 1 year, 4 months ago.

  • Creator
    Topic
  • #44033

    I’m currently following the manual installation of HDP. I am trying to make it pass the smoke tests for HDFS an I cannot access my files through the browser. I can access the Hadoop Namenode page where it states the status of it, but if I click the “Browse the filesystem” I get this error:

    HTTP ERROR 500

    Problem accessing /nn_browsedfscontent.jsp. Reason:

    Can’t browse the DFS since there are no live nodes available to redirect to.

    From what I have read it is because the datanodes are started properly, and I’ve tried getting them to start but I don’t get any information back from running hdfs datanode as root. If I run hdfs datanode from a standard hadoop user I get this error:

    13/11/19 15:58:11 WARN common.Util: Path /grid/hadoop/hdfs/dn should be specified as a URI in configuration files. Please update hdfs configuration.
    13/11/19 15:58:11 WARN datanode.DataNode: Invalid dfs.datanode.data.dir /grid/hadoop/hdfs/dn :
    EPERM: Operation not permitted

    I’ve followed the tutorial to the best of my ability, but I must be missing something. Any help would be appreciated! Thanks :)

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

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

  • Author
    Replies
  • #44105

    I added some of the users, but I gave permissions to root so datanode could start up under them. Will hadoop not allow datanode to start under root and I need to add the users to the system? Also, what’s the best way to go about setting up these users and groups? I’m currently using CentOS 6

    Collapse
    #44097

    Dave
    Moderator

    Hi Chris,

    Did you change the directories permissions and ownerships as required in the manual install?
    The DataNode should run as hdfs and not root.

    Thanks

    Dave

    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.