Database Reference
In-Depth Information
Now I can install Hadoop. For demonstration purposes, I do only a basic installation to get the name node and
data nodes running. From that point, the installation process is the same as was shown in Chapter 2, but it is sourced
from Bigtop. On the Name Node machine (hc2nn), I install the Name Node and Resource Manager components as the
root user:
yum install hadoop-hdfs-namenode hadoop-yarn-resourcemanager
On the Data Node machines (hc2r1m1 to hc2r1m4), I install the data node and node manager components as the
root user:
yum install hadoop-hdfs-datanode hadoop-yarn-nodemanager
Now, I set up the configuration on each machine in the cluster under /etc/hadoop/conf. I present the necessary
entries and changes briefly here, but you can find full details in Chapter 2.
I add the following to the core-site.xml file:
<property>
<name>fs.default.name</name>
<value>hdfs://hc2nn:8020</value>
</property>
I then modify the hdfs-site.xml file as follows:
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<property>
<name>dfs.permissions.superusergroup</name>
<value>hadoop</value>
</property>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
Next, I add the following changes to the yarn-site.xml file:
<property>
<name>yarn.resourcemanager.address</name>
<value>hc2nn:8032</value>
</property>
<property>
<name>yarn.resourcemanager.scheduler.address</name>
<value>hc2nn:8030</value>
</property>
 
Search WWH ::




Custom Search