Home Forums HDP on Windows – Installation Installed successfully, but dfsadmin report says zero nodes

This topic contains 3 replies, has 3 voices, and was last updated by  Seth Lyubich 1 year, 1 month ago.

  • Creator
    Topic
  • #29640

    Sudhakar GS
    Member

    I installed HDP for Windows on a total of 4 nodes successfully. However, when i go to the NameNode web interface to check DFS health, it says something as below.

    ********************
    6 files and directories, 0 blocks = 6 total. Heap Size is 41.38 MB / 3.87 GB (1%)

    Configured Capacity : 0 KB
    DFS Used : 0 KB
    Non DFS Used : 0 KB
    DFS Remaining : 0 KB
    DFS Used% : 100 %
    DFS Remaining% : 0 %
    Live Nodes : 0
    Dead Nodes : 0
    Decommissioning Nodes : 0
    Number of Under-Replicated Blocks : 0

    There are no datanodes in the cluster.

    ********************
    I get the same when do ‘hadoop dfsadmin -report’ on the command prompt, i get the same report.

    On another note, the Slaves and Masters file in the conf directory has only one entry ‘localhost’. I tried changing the Slaves file to include the hostnames of the slaves and restarted the services, but still no luck.

    Quite stuck here. Appreciate any help I can get. Thanks.

    Regards,
    Sudhakar

Viewing 3 replies - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.

  • Author
    Replies
  • #29773

    Seth Lyubich
    Keymaster

    Hi Sudhakar,

    Please make sure that your nodes comunicate internally using internal hostnames. You can test by making sure that port 8020 is reachable on NameNode host from Datanode machine. You can try telnet with internal ip to perform the test. Please let us know if this is useful.

    Thanks,
    Seth

    Collapse
    #29691

    Sudhakar GS
    Member

    Hi Ted,

    Thanks for the response. I updated the Slaves file to include the Fully Qualified Domain Name of the slave. I restarted both the NameNode and Job Tracker services. No luck.

    I also updated the Masters file on the slave to include the FQDN of the Master. I then restarted DataNode and Task Tracker services. Still no luck.

    BTW, I am trying this on AWS using their Free tier EC2 instances.

    The hostnames of the Slave and Master are of the format
    ec2-XX-XX-XX-XX.ap-southeast-1.compute.amazonaws.com.

    Please let me know if you need more info. Thanks.

    Regards,
    Sudhakar

    Collapse
    #29648

    tedr
    Moderator

    Hi Sudhakar,

    After making the changes to the slaves file did you restart HDFS and MapReduce?

    Thanks,
    Ted.

    Collapse
Viewing 3 replies - 1 through 3 (of 3 total)