HDP on Linux – Installation Forum

HDP 2.1, HDFS dos not start

  • #52303
    Lohit Giri
    Participant

    Hi HWX Team,

    i tried to installed HDP 2.1 yesterday on my dev cluster (Centos 6.4 x 64). Installation of all component went fine.

    But at the end Namenode start up timed out from Ambari web console. When i try to look onto namenode logs, there are no errors. But when i try to look onto Datanode logs. i see below errors repeatedly –

    2014-04-24 17:47:14,958 INFO datanode.DataNode (BPServiceActor.java:register(777)) – Problem connecting to server: dev-cluster-nn1.crl.net/172.28.48.21:8020
    2014-04-24 17:48:15,962 INFO datanode.DataNode (BPServiceActor.java:register(777)) – Problem connecting to server: dev-cluster-nn1.crl.net/172.28.48.21:8020
    2014-04-24 17:49:16,967 INFO datanode.DataNode (BPServiceActor.java:register(777)) – Problem connecting to server: dev-cluster-nn1.crl.net/172.28.48.21:8020

    i tried to check the connection from DNs to NN on the port and everything looks alright. i can do telnet on port 8020 for rpc. that looks ok.

    Can someone please suggest, what might be wrong here?

    Also i already did namenode format once alaready and tried restarting DNs and NN. Also i checked /etc/sudoers, and disabled “Defaults requiretty” already.

    Please help.

to create new topics or reply. | New User Registration

  • Author
    Replies
  • #52304
    Lohit Giri
    Participant

    pls ignore this issue, i found the issue. it was rack topology script behaving bad and namenode stuck trying to get rack awareness.

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.