Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce >> mail # user >> Problem setting up 0.23 on a Cluster


Copy link to this message
-
Re: Problem setting up 0.23 on a Cluster
Hi,

Here are the steps I had to do to get the `sbin/hadoop-daemons.sh stop
datanode` (note the plural daemons) work.

1. copy the hadoop-env.sh from ./share/hadoop/common/templates/conf to conf
folder.
2. Add the JAVA_HOME in hadoop-env.sh.
3. Comment the below line in hadoop-env.sh.
           export HADOOP_LOG_DIR=${HADOOP_LOG_DIR}/$USER

Still need to figure out the below problem with starting the NodeManager on
the cluster

praveensripati@praveen-laptop:~/Installations/hadoop-0.23.0$
*bin/yarn-daemons.sh
start nodemanager*
cat: ~/home/praveensripati/Installations/hadoop-0.23.0/conf/slaves: No such
file or directory

Regards,
Praveen

On Sat, Jan 7, 2012 at 3:23 PM, Praveen Sripati <[EMAIL PROTECTED]>wrote:

> Ronald,
>
> Here is the output
>
> uid=1000(praveensripati) gid=1000(praveensripati)
> groups=1000(praveensripati),4(adm),20(dialout),24(cdrom),46(plugdev),116(lpadmin),118(admin),124(sambashare)
>
> Regards,
> Praveen
>
> On Sat, Jan 7, 2012 at 12:45 PM, Ronald Petty <[EMAIL PROTECTED]>wrote:
>
>> Praveen,
>>
>> What does 'id' output?
>>
>> Kindest regards.
>>
>> Ron
>>
>>
>> On Fri, Jan 6, 2012 at 9:51 AM, Praveen Sripati <[EMAIL PROTECTED]
>> > wrote:
>>
>>> Hi,
>>>
>>> I am able to run 0.23 on a single node and trying to setup it on a
>>> cluster and getting errors.
>>>
>>> When I try to start the data nodes, I get the below errors. I have also
>>> tried adding `export
>>> HADOOP_LOG_DIR=/home/praveensripati/Installations/hadoop-0.23.0/logs` to
>>> .bashrc and there hadn't been any change.
>>>
>>> praveensripati@praveen-laptop:~/Installations/hadoop-0.23.0$ *sbin/hadoop-daemons.sh
>>> start datanode*
>>> praveen-laptop: mkdir: cannot create directory `/praveensripati':
>>> Permission denied
>>> praveen-laptop: chown: cannot access `/praveensripati/praveensripati':
>>> No such file or directory
>>> praveen-laptop: starting datanode, logging to
>>> /praveensripati/praveensripati/hadoop-praveensripati-datanode-praveen-laptop.out
>>> praveen-laptop:
>>> /home/praveensripati/Installations/hadoop-0.23.0/sbin/hadoop-daemon.sh:
>>> line 144:
>>> /praveensripati/praveensripati/hadoop-praveensripati-datanode-praveen-laptop.out:
>>> No such file or directory
>>> ubuntu-guest: mkdir: cannot create directory `/praveensripati':
>>> Permission denied
>>> ubuntu-guest: chown: cannot access `/praveensripati/praveensripati': No
>>> such file or directory
>>> ubuntu-guest: starting datanode, logging to
>>> /praveensripati/praveensripati/hadoop-praveensripati-datanode-ubuntu-guest.out
>>> ubuntu-guest:
>>> /home/praveensripati/Installations/hadoop-0.23.0/sbin/hadoop-daemon.sh:
>>> line 143:
>>> /praveensripati/praveensripati/hadoop-praveensripati-datanode-ubuntu-guest.out:
>>> No such file or directory
>>> praveen-laptop: head: cannot open
>>> `/praveensripati/praveensripati/hadoop-praveensripati-datanode-praveen-laptop.out'
>>> for reading: No such file or directory
>>> ubuntu-guest: head: cannot open
>>> `/praveensripati/praveensripati/hadoop-praveensripati-datanode-ubuntu-guest.out'
>>> for reading: No such file or directory
>>>
>>> When I try to start the node manager, I get the below errors
>>>
>>> praveensripati@praveen-laptop:~/Installations/hadoop-0.23.0$ *bin/yarn-daemons.sh
>>> start nodemanager*
>>> cat: ~/home/praveensripati/Installations/hadoop-0.23.0/conf/slaves: No
>>> such file or directory
>>>
>>> Is there a bug in the code or am I missing some settings?
>>>
>>> Regards,
>>> Praveen
>>>
>>>
>>
>