AnsweredAssumed Answered

File system to be used by Hadoop for data storage

Question asked by raj on Feb 14, 2013
Latest reply on Feb 15, 2013 by Ted Dunning
IS THE BELOW ASSUMPTION CORRECT ??

or How to make sure a particular file system is used by HDFS for data storage on DATANODES.

for example : I have three node cluster (BD_LX1,BD_LX2,BD_LX3) BD_LX1 used for namenode and BD_LX2 and DB_LX3 for datanode and DB_LX2 and DB_LX3 (datanodes) has a file system called /data which is 500GB on each node and is to be used by Hadoop for HDFS purpose (data storage in blocks).
 
so DO WE PUT THE BELOW in hdfs-site.xml on both BD_LX2 and BD_LX3 and then FORMAT for HDFS use using "# bin/hadoop namenode -format"

== hdfs-site.xml
<property>
    <name>dfs.name.dir</name>
    <value>/data/</value>
</property>

Outcomes