Detailed steps to install Hadoop cluster under Linux

#hadoop
export HADOOP_HOME=/usr/hadoop/hadoop-2.6.0
export CLASSPATH=$CLASSPATH:$HADOOP_HOME/lib
export PATH=$PATH:$HADOOP_HOME/bin 

source /etc/profile 

cd /usr/hadoop/hadoop-2.6.0/etc/hadoop 

export JAVA_HOME=/usr/java/jdk1.8.0_181 

<configuration>
<property>
        <name>hadoop.tmp.dir</name>
        <value>file:/root/hadoop/tmp</value>
</property>
<!--Port number 9000-->
<property>
        <name>fs.default.name</name>
        <value>hdfs://z1:9000</value>
</property>
<!--Turn on the trash can mechanism in minutes-->
<property>
    <name>fs.trash.insterval</name>
    <value>10080</value>
</property>
<!--Buffer size, actual work depends on server performance-->
<property>
    <name>io.file.buffer.sizei</name>
    <value>4096</value>
</property>
</configuration>
                                                                                                                                                                  39,9 bottom 

cp mapred-site.xml.template mapred-site.xml
vim mapred-site.xml

<configuration>
<property>
<!--Specify Mapreduce to run on yarn-->
   <name>mapreduce.framework.name</name>
   <value>yarn</value>
 </property>
<!--Start MapReduce's small task mode-->
<property>
      <name>mapred.job.ubertask.enable</name>
      <value>true</value>
</property>
<property>
      <name>mapred.job.tracker</name>
      <value>z1:9001</value>
</property>
 
<property>
<name>mapreduce.jobhistory.address</name>
<value>CMaster:10020</value>
</property>
</configuration> 

vim yarn-site.xml

<configuration>
 
<!-- Site specific YARN configuration properties -->
 
<!--Configure the location of the yarn master node-->
<property>
        <name>yarn.resourcemanager.hostname</name>
        <value>z1</value>
</property>
<property>
<!-- mapreduce, the way to get data when executing shuff1e.-->
<description>The address of the applications manager interface in the RM.</description>
     <name>yarn.resourcemanager.address</name>
     <value>z1:8032</value>
</property>
<property>
  <name>yarn.resourcemanager.scheduler.address</name>
  <value>z1:8030</value>
</property>
 
<property>
  <name>yarn.resourcemanager.webapp.address</name>
  <value>z1:8088</value>
</property>
 
<property>
  <name>yarn.resourcemanager.webapp.https.address</name>
  <value>z1:8090</value>
</property>
<property>
  <name>yarn.resourcemanager.resource-tracker.address</name>
  <value>z1:8031</value>
</property>
<property>
  <name>yarn.resourcemanager.admin.address</name>
  <value>z1:8033</value>
</property>
<property><!--The way to get data when mapreduce executes shuff1e, -->
  <name>yarn.nodemanager.aux-services</name>
  <value>mapreduce_shuffle</value>
</property>
<property>
<!--Set memory, memory allocation of yarn-->
  <name>yarn.scheduler.maximum-a11ocation-mb</name>
  <value>2024</value>
  <discription>Available memory per node, unit: M, default: 8182MB</discription>
</property>
<property>
  <name>yarn.nodemanager.vmem-pmem-ratio</name>
  <value>2.1</value>
</property>
<property>
  <name>yarn.nodemanager.resource.memory-mb</name>
  <value>1024</value>
</property>
<property>
  <name>yarn.nodemanager.vmem-check-enabled</name>
  <value>false</value>
</property>
 
<property>
<name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
<value>org.apache.hadoop.mapred.ShuffleHandler</value>
</property>
 
 
</configuration>
                                                    

vim hdfs-site.xml 

<configuration>
<property>
        <name>dfs.namenode.name.dir</name>
        <value>file:/usr/hadoop/hadoop-2.6.0/hadoopDesk/namenodeDatas</value>
</property>
 <property>
        <name>dfs.datanode.data.dir</name>
        <value>file:/usr/hadoop/hadoop-2.6.0/hadoopDatas/namenodeDatas</value>
    </property>
<property>
<!--Number of copies-->
<name>dfs.replication</name>
<value>3</value>
</property>
<!--Set hdfs file permissions-->
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
<!--Set the size of a file slice: 128m-->
<property>
<name>dfs.bloksize</name>
<value>134217728</value>
</property>
</configuration>

vim slaves

scp -r /etc/profile root@z2:/etc/profile #Distribute the environment variable profile file to the z2 node scp -r /etc/profile root@z3:/etc/profile #Distribute the environment variable profile file to the z3 node scp -r /usr/hadoop root@z2:/usr/ #Distribute the hadoop file to the z2 node scp -r /usr/hadoop root@z3:/usr/ #Distribute the hadoop file to the z3 node

source /etc/profile

hadoop namenode -format

cd /usr/hadoop/hadoop-2.6.0
sbin/start-all.sh starts Hadoop and operates only on the master node