hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From A Df <abbey_dragonfor...@yahoo.com>
Subject Hadoop Cluster setup - no datanode
Date Fri, 12 Aug 2011 23:19:02 GMT
Hello Mates:

Thanks to everyone for their help so far. I have learnt a lot and have now done single and
pseudo mode. I have a hadoop cluster but I ran jps on the master node and slave node but
not all process are started

22160 NameNode
22716 Jps
22458 JobTracker

32195 Jps

I also checked the logs and I see files for all the datanodes, jobtracker, namenode, secondarynamenode,
and tasktracker. The tasktracker has one slave node log missing. The namenode formatted correctly.
I set the values for below so I'm not sure if I need more. My cluster is 11 nodes (1 master,
10 slaves). I do not have permission to access root only my directory so hadoop is installed
in there. I can ssh to the slaves properly.
	* fs.default.name, dfs.name.dir, dfs.data.dir, mapred.job.tracker, mapred.system.dir

It also gave errors regarding:
	* it cannot find the hadoop-daemon.sh file but I can see it

/home/my-user/hadoop-0.20.2_cluster/bin/hadoop-daemon.sh: line 40: cd: /home/my-user/hadoop-0.20.2_cluster/bin:
No such file or directory

	* it has the wrong path for the hadoop-config.sh so which parameter sets this field??

/home/my-user/hadoop-0.20.2_cluster/bin/hadoop-daemon.sh: line 42: /home/my-user/hadoop-0.20.2_cluster/hadoop-config.sh:
No such file or directory

	* not being able to create the log directory on the same slave node that doesn't have its
tasktracker, which parameters should be used to set the log directory?

The same slave node which is giving problems also has:
 Usage: hadoop-daemon.sh [--config <conf-dir>] [--hosts hostlistfile] (start|stop)
<hadoop-command> <args...>

Thanks for your help.


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message