cloudera 0.18 -> 0.20

 * hadoop install 
{{{
$ sudo su -
# apt-get install hadoop-0.18 hadoop-0.18-namenode  hadoop-conf-pseudo
# apt-get install hadoop-0.18-datanode hadoop-0.18-jobtracker hadoop-0.18-tasktracker
}}}

 * hadoop namenode format
{{{
# su -s /bin/bash - hadoop -c 'hadoop namenode -format'
}}}
   * use instruction 
{{{
# su -s /bin/bash - hadoop -c " COMMAND "
}}}

 * startup 
{{{
# /etc/init.d/hadoop-namenode start
}}}



 * setup the conf files
...
 (hadoop-env.sh, core-site.xml, hdfs-site.xml, mapred-site.xml)
....

 * upgrade cloudera-hadoop-0.18 ---> official-hadoop 0.20
{{{
# su -s /bin/bash - hadoop -c " /opt/hadoop/bin/hadoop namenode -upgrade "
}}}

 * the successful screen shot :
-----------
 [[Image(2010-01-06-181048_959x723_scrot.png)]]
-----------
 * but terminal console is handle by hadoop process 
-----------
 [[Image(2010-01-06-181350_881x389_scrot.png)]]

 * Right now,  our NameNode is in SAFE mode waiting for the DataNodes to connect.

 * Start the rest of HDFS

When the datanodes start, they will connect to the Name Node and an upgrade will be initiated.

{{{
.... hadoop-dfs start
}}}


Once all the DataNodes have upgraded, you should see a message that Safe mode will be turned off automatically in X seconds.

 * start MapReduce
After you have verified correct operation of HDFS, you are ready to start MapReduce.

 * finalize the hdfs upgrade

Now that you've confirmed your cluster is successfully running on Hadoop 0.20, you can go ahead and finalize the HDFS upgrade:
{{{{
$ sudo -u hadoop hadoop dfsadmin -finalizeUpgrade
}}}