Kumar, Your config seems alright. That post described it for 0.21/trunk scripts I believe. On 0.20.x based release, like CDH3, you can also simply use the hadoop-daemon.sh to do it. Just have to mess with some PID files.
Here's how I do it on my Mac to start 3 DNs: $ ls conf* conf conf.1 conf.2 $ hadoop-daemon.sh start datanode # Default $ rm pids/hadoop-harsh-datanode.pid $ hadoop-daemon.sh --config conf.1 start datanode # conf.1 DN $ rm pids/hadoop-harsh-datanode.pid $ hadoop-daemon.sh --config conf.2 start datanode # conf.2 DN To kill any DN, jps/ps and find out which one it is you want to and kill the PID displayed. On Sat, Jun 11, 2011 at 5:34 AM, Kumar Kandasami <kumaravel.kandas...@gmail.com> wrote: > Thank you Harsh. > > I have been following the documentation in the mailing list, and have an > issue starting the second data node (because of port conflict). > > - First, I don't see the bin/hdfs in the directory ( I am using Mac m/c and > installed hadoop using CDH3 tarball) > - I am using the following command instead of the one mentioned in the step > #3 in the mailing list. > > ./bin/hadoop-daemon.sh --config ../conf2 start datanode > > Error: datanode running as process 5981. Stop it first. > > - Port configuration in the hdfs-site.xml below. > > Data Node #1: Conf file > > <property> > <name>dfs.replication</name> > <value>1</value> > </property> > <property> > <name>dfs.permissions</name> > <value>false</value> > </property> > > <property> > <!-- specify this so that running 'hadoop namenode -format' formats the > right dir --> > <name>dfs.name.dir</name> > <value>/var/lib/hadoop-0.20/cache/hadoop/dfs/name</value> > </property> > > <property> > <name>dfs.data.dir</name> > <value>/var/lib/hadoop-0.20/cache/hadoop/dfs/data</value> > </property> > <property> > <name>dfs.datanode.address</name> > <value>0.0.0.0:50010</value> > </property> > > <property> > <name>dfs.datanode.ipc.address</name> > <value>0.0.0.0:50020</value> > <description> > The datanode ipc server address and port. > If the port is 0 then the server will start on a free port. > </description> > </property> > > <property> > <name>dfs.datanode.http.address</name> > <value>0.0.0.0:50075</value> > </property> > > <property> > <name>dfs.datanode.https.address</name> > <value>0.0.0.0:50475</value> > </property> > </configuration> > > Data Node #2: Conf (2) file > > <property> > <name>dfs.replication</name> > <value>1</value> > </property> > <property> > <name>dfs.permissions</name> > <value>false</value> > </property> > > <property> > <!-- specify this so that running 'hadoop namenode -format' formats the > right dir --> > <name>dfs.name.dir</name> > <value>/var/lib/hadoop-0.20/cache/hadoop/dfs/name</value> > </property> > > <property> > <name>dfs.data.dir</name> > <value>/var/lib/hadoop-0.20/cache/hadoop/dfs/data2</value> > </property> > > <property> > <name>dfs.datanode.address</name> > <value>0.0.0.0:50012</value> > </property> > > <property> > <name>dfs.datanode.ipc.address</name> > <value>0.0.0.0:50022</value> > <description> > The datanode ipc server address and port. > If the port is 0 then the server will start on a free port. > </description> > </property> > > <property> > <name>dfs.datanode.http.address</name> > <value>0.0.0.0:50077</value> > </property> > > <property> > <name>dfs.datanode.https.address</name> > <value>0.0.0.0:50477</value> > </property> > > > > Kumar _/|\_ > www.saisk.com > ku...@saisk.com > "making a profound difference with knowledge and creativity..." > > > On Fri, Jun 10, 2011 at 12:20 AM, Harsh J <ha...@cloudera.com> wrote: > >> Try using search-hadoop.com, its pretty kick-ass. >> >> Here's what you're seeking (Matt's reply in particular): >> >> http://search-hadoop.com/m/sApJY1zWgQV/multiple+datanodes&subj=Multiple+DataNodes+on+a+single+machine >> >> On Fri, Jun 10, 2011 at 9:04 AM, Kumar Kandasami >> <kumaravel.kandas...@gmail.com> wrote: >> > Environment: Mac 10.6.x. Hadoop version: hadoop-0.20.2-cdh3u0 >> > >> > Is there any good reference/link that provides configuration of >> additional >> > data-nodes on a single machine (in pseudo distributed mode). >> > >> > >> > Thanks for the support. >> > >> > >> > Kumar _/|\_ >> > www.saisk.com >> > ku...@saisk.com >> > "making a profound difference with knowledge and creativity..." >> > >> >> >> >> -- >> Harsh J >> > -- Harsh J