SUSE上搭建Hadoop環境（單機模式+僞分佈模式）

【環境】：

常常遭遇由於依賴軟件版本不匹配致使的問題，此次大意了，覺得java問題不大，就用原本經過yast安裝的java1.6 openjdk去搞了，結果可想而知，問題不少，反覆定位，反覆谷歌百度，最後一朋友啓發下決定換換jdk版本。問題解決了，因此這裏貼下個人環境

java環境： java version "1.7.0_51"
               Java(TM) SE Runtime Environment (build 1.7.0_51-b13)
               Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode)

系統：       openSUSE 11.2 (x86_64)

hadoop版本：Hadoop-1.1.2.tar.gz

【Step1:】建立hadoop用戶及用戶組

 組：hadoop 

 用戶：hadoop  -> /home/hadoop

 加權限： vi /etc/sudoers  增長 hadoop  ALL=(ALL:ALL)  ALL

【Stpe2:】安裝hadoop

 筆者tar xf 安裝完後是這樣的目錄結構(供參考)：

 /home/hadoop/hadoop-home/[bin|conf]

【Step3:】配SSH(避免啓動hadoop時須要密碼)

 略安裝ssh

 ssh-keygen -t rsa -P "" [一路回車及確認]

 cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys

 嘗試 ssh localhost [檢查下是否是不須要密碼啦]

【Step4:】安裝java

 版本見【環境】部分

【Step5:】配conf/hadoop-env.sh 

export JAVA_HOME=/usr/java/jdk1.7.0_17xxx        #[jdk目錄]

export HADOOP_INSTALL=/home/hadoop/hadoop-home          

export PATH=$PATH:$HADOOP_INSTALL/bin        #[這裏是hadoop腳本所在目錄]

【Step6:】使用單機模式

 hadoop version

mkdir input

 man find > input/test.txt

 hadoop jar hadoop-examples-1.1.2.jar wordcount input output

【Step7:】僞分佈模式（單機實現namenode,datanode,tackerd等模塊）

 conf/[core-site.xml、hdfs-site.xml、mapred-site.xml]

 core-site.xml

<configuration>
    <property>
        <name>fs.default.name</name>
        <value>hdfs://localhost:9000</value>
    </property>
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/usr/local/hadoop/tmp</value>
    </property>
</configuration>

hdfs-site.xml

<configuration>
    <property>
        <name>dfs.replication</name>
        <value>2</value>
    </property>
    <property>
        <name>dfs.name.dir</name>
        <value>/usr/local/hadoop/datalog1,/usr/local/hadoop/datalog2</value>
    </property>
    <prop<configuration>   
    <property>  
        <name>mapred.job.tracker</name>
        <value>localhost:9001</value>   
    </property>
</configuration>erty>
        <name>dfs.data.dir</name>
        <value>/usr/local/hadoop/data1,/usr/local/hadoop/data2</value>
    </property>
</configuration>

mapred-site.xml

<configuration>   
    <property>  
        <name>mapred.job.tracker</name>
        <value>localhost:9001</value>   
    </property>
</configuration>

【Step8:】啓動

 格式化：hadoop namenode -format

 cd bin

 sh start-all.sh

hadoop@linux-peterguo:~/hadoop-home/bin> sh start-all.sh
starting namenode, logging to /home/hadoop/hadoop-home/libexec/../logs/hadoop-hadoop-namenode-linux-peterguo.out
localhost: starting datanode, logging to /home/hadoop/hadoop-home/libexec/../logs/hadoop-hadoop-datanode-linux-peterguo.out
localhost: starting secondarynamenode, logging to /home/hadoop/hadoop-home/libexec/../logs/hadoop-hadoop-secondarynamenode-linux-peterguo.out
starting jobtracker, logging to /home/hadoop/hadoop-home/libexec/../logs/hadoop-hadoop-jobtracker-linux-peterguo.out
localhost: starting tasktracker, logging to /home/hadoop/hadoop-home/libexec/../logs/hadoop-hadoop-tasktracker-linux-peterguo.out

 jps查看進程是否全啓動 五個java進程 jobtracker/tasktracker/namenode/datanode/sencondarynamenode

 能夠經過下面的操做來查看服務是否正常，在Hadoop中用於監控集羣健康狀態的Web界面：
http://localhost:50030/ - Hadoop 管理介面
http://localhost:50060/ - Hadoop Task Tracker 狀態
http://localhost:50070/ - Hadoop DFS 狀態

【Step9:】操做dfs數據文件

hadoop dfs -mkdir input

hadoop dfs -copyFromLocal input/test.txt input

hadoop dfs -ls input

【Step10:】運行dfs上的mr

hadoop jar hadoop-examples-1.1.2.jar wordcount input output 

hadoop dfs -cat output/*

【Step11:】關閉

stop-all.sh

參考：http://blog.csdn.net/zhaoyl03/article/details/8657104