Three ways to start hadoop

 

• the first way

  Start: Start HDFS and MapReduce respectively

  The command is as follows: start-dfs.sh  start-mapreted.sh

  The commands are as follows: stop-dfs.sh   stop-mapreted.sh

• the second way

  start all or stop all

  start up:

  Command: start-all.sh

  Startup order: NameNode , DateNode , SecondaryNameNode , JobTracker , TaskTracker

   

  stop:

  Command: stop-all.sh

  Turn off sequentiality: JobTracker , TaskTracker , NameNode , DateNode , SecondaryNameNode

   

• The third way to start

  Each daemon thread is started one by one, and the startup sequence is as follows:

  NameNode DateNode SecondaryNameNode，JobTracker，TaskTracker _ _

  The command is as follows:

  start up:

        hadoop-daemon.shdaemon ( daemon )

   

  hadoop-daemon.sh start purpose

  hadoop-daemon.sh start datenode

   

  hadoop-daemon.sh start secondarynamenode

  hadoop-daemon.sh start jobtracker

  hadoop-daemon.sh start tasktracker

   

  shutdown command:

  hadoop-daemon.sh stop tasktracker

   

  hadoop-daemons.sh  starts multiple processes

  The datanode and tasktracker will not be divided into multiple machines. Start from the node and use the

           Big data blockchain group: 518093343