Wednesday, December 1, 2010

How to start and stop datanodes and tasktrackers without bringing down the cluster

Assuming that Hadoop and Java is set up on the nodes, following are the commands that can be used to start and stop datanodes and tasktrackers.

cd <hadoop_installation>/bin

- To start datanode
hadoop-daemon.sh start datanode

- To start tasktracker
hadoop-daemon.sh start tasktracker



- To stop datanode (remove the datanode from the cluster)
hadoop-daemon.sh stop datanode

- To stop tasktracker (remove tasktracker from the cluster)
hadoop-daemon.sh stop tasktracker

1 comment:

  1. Could you pl. also say

    1. what is the configuration file which shall be used by datanode
    2. Where does the configuration file resides
    3. On are the things which needs to be installed on data node.. is it full hadoop or just some run time support etc i.e subset of hadoop distribution only
    4. What are the configuration changes which should exist in data notes i.e slaves
    5. Best UI tool you have come across to visually make the configuration changes

    Regards,
    Raja Nagendra Kumar,
    C.T.O
    www.tejasoft.com

    ReplyDelete