0

I am trying to set up a single node cluster in hadoop, but after giving the command start-all.sh , jps doesn't show namenode and when I stop all processes, I get the result as no namenode to stop. Also, I saw at few places that clearing the tmp folder data solves the problem, but I am not able to find it in my hadoop folder. Following is the result of jps command:

root@neel:~/sbin# jps
4166 ResourceManager
4793 Jps
4299 NodeManager
3995 SecondaryNameNode
3804 DataNode
27852 -- process information unavailable
root@neel:~/sbin#
Neel Shah
  • 319
  • 7
  • 16
  • Go to the `/var/log/hadoop` (or where ever your logs are). Inside there you will find a log file for `NameNode`. Post that log content here. It is hard to say what is issue with just `jps` showing no `NameNode` daemon. – ViKiG Jun 21 '16 at 15:46

1 Answers1

0
1 Clear the namenode and datanode folder 
2 format the namnode and datanode
   $HADOOP_HOME/bin/hadoop namenode -format
   $HADOOP_HOME/bin/hdfs namenode -format
   $HADOOP_HOME/bin/hadoop datanode -format
   $HADOOP_HOME/bin/hdfs datanode -format
3 Then start hadoop
Jibi Gopi
  • 61
  • 2
  • 5042 Main 31077 ResourceManager 31210 NodeManager 30906 SecondaryNameNode 31532 Jps 30718 DataNode I get the following output from jps command. Still the namenode has not been started. – Neel Shah Jun 21 '16 at 04:59
  • have you clear your namenode folder? after clearing please stop and start hadoop – Jibi Gopi Jun 21 '16 at 05:31