Question

我决定使用hadoop2.5.0设置HADOOP_PREFIX，但当我想查看版本或格式namenode时会发生此错误：

[hdfs@master1 bin]$ ./hadoop version
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 16: 
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 18: 
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 23: 
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 29: 
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 30: 
: command not found.5.0/etc/hadoop/hadoop-env.sh: line 32: 
'usr/local/hadoop-2.5.0/etc/hadoop/hadoop-env.sh: line 34: syntax error near unexpected token `do
'usr/local/hadoop-2.5.0/etc/hadoop/hadoop-env.sh: line 34: `for f in $HADOOP_HOME/contrib/capacity-scheduler/*.jar; do
Error: Could not find or load main class org.apache.hadoop.util.VersionInfo

操作系统：CentOs 6.5。

模式：完全分配4个节点：1个主机+ 3个从机。

Answer 1

  Please verified steps from this :

    Step-1 Create a dedicated user(hduser)for hadoop on three machine from terminal

    Command-1  sudo addgroup hadoop
    Command-2  sudo adduser --ingroup hadoop hduser
    Command-3  sudo adduser hduser sudo



    Step-2 Login to user(hduser) on three machine from terminal

    Command-1 su hduser



    Step-3 Create passwordless ssh between all the machines

    Command-1 ssh-keygen -t rsa -P ""

    Command-2 cat $HOME/.ssh/id_rsa.pub >> $HOME/.ssh/authorized_keys

    Command-3 ssh-copy-id -i $HOME/.ssh/id_rsa.pub hduser@slave
    hduser@slave means hduser@IP of slave (Eg:hduser@192.168.213.25)
    Execute Command-3 from master machine to the two slave machines



    Step-4 Setup JAVA JDK on all the machines

    Command-1 sudo apt-get install sun-java6-jdk

    Command-2 sudo update-java-alternatives -s java-6-sun



    Step-5 Download hadoop-2.x tar and unzip it on all machines

    Command-1  cd /usr/local

    Command-2  sudo tar xzf hadoop-2.x.tar.gz

    Command-3  sudo mv hadoop-2.x hadoop

    Command-4  sudo chown -R hduser:hadoop hadoop


    Step-6 Open $HOME/.bashrc on all machines

    Command-1 vi $HOME/.bashrc

    Step-7 Add the following lines to the end of opened .bashrc file on all machines
    (Find location of JAVA_HOME on all of the machines. It should be set accordingly on each machine)
    export JAVA_HOME=/usr/local/java/jdk1.6.0_20 
    export HADOOP_INSTALL=/usr/local/hadoop 
    export PATH=$JAVA_HOME/bin:$HADOOP_INSTALL/bin:$PATH 
    export HADOOP_MAPRED_HOME=$HADOOP_INSTALL 
    export HADOOP_COMMON_HOME=$HADOOP_INSTALL 
    export HADOOP_HDFS_HOME=$HADOOP_INSTALL 
    export YARN_HOME=$HADOOP_INSTALL 
    export HADOOP_CONF_DIR=$HADOOP_INSTALL/etc/hadoop 
    export YARN_CONF_DIR=$HADOOP_INSTALL/etc/hadoop 
    Press Esc and type :wq! To update the file

    Now execute this command

    Command-1 source .bashrc


    Step-8 Update /etc/hosts files on all machine

    I. add ip and name of machines for all the machines in /etc/hosts file
    eg:-

    192.168.213.25 N337
    192.168.213.94 N336
    192.168.213.47 UBUNTU 
    II. comment all other things



    Step-9 Tweak Config Files on all machines

    I.hadoop-config.sh

    Command-1 cd $HADOOP_INSTALL
    Command-2 vi libexec/hadoop-config.sh

    Now add the following line at the start of hadoop-config.sh(Take appropriate location of JAVA_HOME on each machine)
    export JAVA_HOME=/usr/local/java/jdk1.6.0_20

    II.yarn-env.sh

    Command-1 cd $HADOOP_INSTALL/etc/hadoop
    Command-2 vi yarn-env.sh

    #Now add following lines

    export JAVA_HOME=/usr/local/java/jdk1.6.0_20 
    export HADOOP_INSTALL=/usr/local/hadoop 
    export PATH=$JAVA_HOME/bin:$HADOOP_INSTALL/bin:$PATH 
    export HADOOP_MAPRED_HOME=$HADOOP_INSTALL 
    export HADOOP_COMMON_HOME=$HADOOP_INSTALL 
    export HADOOP_HDFS_HOME=$HADOOP_INSTALL 
    export YARN_HOME=$HADOOP_INSTALL 
    export HADOOP_CONF_DIR=$HADOOP_INSTALL/etc/hadoop 
    export YARN_CONF_DIR=$HADOOP_INSTALL/etc/hadoop 


    #Press Esc and type :wq! To update the file

    III.core-site.xml

    Command-1 vi $HADOOP_INSTALL/etc/hosts/core-site.xml

    Add the following lines

    <configuration> 
    <property> 
       <name>fs.default.name</name> 
       <value>hdfs://N364U:9000</value> 
    </property> 
    </configuration> 

    IV.yarn-site.xml

    Command-1 vi $HADOOP_INSTALL/etc/hosts/yarn-site.xml

    Add the following lines(change machine name according to your machine)


    <configuration>
    <property> 
    <name>yarn.nodemanager.aux-services</name> 
    <value>mapreduce_shuffle</value> 
    </property> 
    <property> 
    <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name> 
    <value>org.apache.hadoop.mapred.ShuffleHandler</value> 
    </property> 
    <property> 
    <name>yarn.resourcemanager.resource-tracker.address</name> 
    <value>N337:8025</value> 
    </property> 
    <property> 
    <name>yarn.resourcemanager.scheduler.address</name> 
    <value>N337:8030</value> 
    </property> 
    <property> 
    <name>yarn.resourcemanager.address</name> 
    <value>N337:8040</value> 
    </property> 
    </configuration>

    V.hdfs-site.xml
    Add the following lines
    <configuration> 
    <property> 
    <name>dfs.replication</name> 
    <value>3</value> 
    </property> 
    </configuration>

    VI.mapred-site.xml
    Add the following lines

    <configuration> 
    <property> 
       <name>mapreduce.framework.name</name> 
       <value>yarn</value> 
    </property> 
    </configuration> 

    Don't forget to save all the configuration files. Cross check.


    Step-10 Add slaves on master machine only

    Command-I vi $HADOOP_INSTALL/etc/hosts/slaves

    Add the IP of the two machines on master machine

    eg: 

    192.168.213.94
    192.168.213.47



    Step-11 Format namenode once

    Command-1 cd $HADOOP_INSTALL

    Command-2 bin/hdfs namenode -format


    Step-12 Now start hadoop

    Command-1 cd $HADOOP_INSTALL
    Command-2 sbin/start-all.sh

hadoop-env.sh文件中的语法错误

1 个答案: