suse linux hadoop,SUSE上搭建Hadoop环境（单机模式+伪分布模式）-CSDN博客

【环境】：

经常遭遇因为依赖软件版本不匹配导致的问题，这次大意了，以为java问题不大，就用本来通过yast安装的java1.6 openjdk去搞了，结果可想而知，问题很多，反复定位，反复谷歌百度，最后一朋友启发下决定换换jdk版本。问题解决了，所以这里贴下我的环境

java环境： 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)

系统： openSUSE 11.2 (x86_64)

hadoop版本：Hadoop-1.1.2.tar.gz

【Step1:】创建hadoop用户及用户组

组：hadoop

用户：hadoop -> /home/hadoop

加权限： vi /etc/sudoers 增加 hadoop ALL=(ALL:ALL) ALL

【Stpe2:】安装hadoop

笔者tar xf 安装完后是这样的目录结构(供参考)：

/home/hadoop/hadoop-home/[bin|conf]

【Step3:】配SSH(避免启动hadoop时需要密码)

略安装ssh

ssh-keygen -t rsa -P "" [一路回车及确认]

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

尝试 ssh localhost [检查下是不是不需要密码啦]

【Step4:】安装java

版本见【环境】部分

【Step5:】配conf/hadoop-env.sh

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

export HADOOP_INSTALL=/home/hadoop/hadoop-home

export PATH=$PATH:$HADOOP_INSTALL/bin #[这里是hadoop脚本所在目录]

【Step6:】使用单机模式

hadoop version

mkdir input

man find > input/test.txt

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

【Step7:】伪分布模式(单机实现namenode,datanode,tackerd等模块)

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

core-site.xml

fs.default.name

hdfs://localhost:9000

hadoop.tmp.dir

/usr/local/hadoop/tmp

hdfs-site.xml

dfs.replication

dfs.name.dir

/usr/local/hadoop/datalog1,/usr/local/hadoop/datalog2

mapred.job.tracker

localhost:9001

erty>

dfs.data.dir

/usr/local/hadoop/data1,/usr/local/hadoop/data2

mapred-site.xml

mapred.job.tracker

localhost:9001

【Step8:】启动

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查看进程是否全启动五个java进程 jobtracker/tasktracker/namenode/datanode/sencondarynamenode

可以通过下面的操作来查看服务是否正常，在Hadoop中用于监控集群健康状态的Web界面：

http://localhost:50030/- Hadoop 管理介面

http://localhost:50060/- Hadoop Task Tracker 状态

http://localhost:50070/- Hadoop DFS 状态

【Step9:】操作dfs数据文件

hadoop dfs -mkdir input

hadoop dfs -copyFromLocal input/test.txt input

hadoop dfs -ls input

【Step10:】运行dfs上的mr

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

hadoop dfs -cat output/*

【Step11:】关闭

stop-all.sh

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