需要重新编译的教程:http://blog.csdn.net/ggz631047367/article/details/42460589
在Ubuntu下创建hadoop用户组和用户
hadoop的管理员最好就是以后要登录桌面环境运行eclipse的用户,否则后面会有拒绝读写的问题出现。当然不是也有办法办法解决。
1. 创建hadoop用户组;
- sudo addgroup hadoop
2. 创建hadoop用户;
- sudo adduser -ingroup hadoop hadoop
3. 给hadoop用户添加权限,打开/etc/sudoers文件;
- sudo gedit /etc/sudoers
在root ALL=(ALL:ALL) ALL下添加hadoop ALL=(ALL:ALL) ALL.
在Ubuntu下安装JDK
具体见:http://blog.csdn.net/ggz631047367/article/details/42366687 //JAVA_HOME=/usr/lib/jvm/jdk1.8.0_25
安装ssh服务
- sudo apt-get install ssh openssh-server
建立ssh无密码登录本机
切换到hadoop用户,执行以下命令:
- su - hadoop
ssh生成密钥有rsa和dsa两种生成方式,默认情况下采用rsa方式。
1. 创建ssh-key,,这里我们采用rsa方式;
- ssh-keygen -t rsa -P "" (注:回车后会在~/.ssh/下生成两个文件:id_rsa和id_rsa.pub这两个文件是成对出现的)
2. 进入~/.ssh/目录下,将id_rsa.pub追加到authorized_keys授权文件中,开始是没有authorized_keys文件的;
- cd ~/.ssh
- cat id_rsa.pub >> authorized_keys (完成后就可以无密码登录本机了。)
3. 登录localhost;
- ssh localhost
4. 执行退出命令;
- exit
安装hadoop
下载地址:http://apache.fayea.com/hadoop/common/stable/hadoop-2.6.0.tar.gz
1. 把hadoop解压到/usr/local下:
- sudo tar -zxvf hadoop-2.6.0.tar.gz
- sudo mv hadoop-2.6.0 /usr/local/hadoop
- sudo chmod -R 775 /usr/local/hadoop
- sudo chown -R hadoop:hadoop /usr/local/hadoop //否则ssh会拒绝访问
2.配置
- 修改bashrc的配置:
- sudo gedit ~/.bashrc
在文件末尾添加:
- #HADOOP VARIABLES START
- export JAVA_HOME=/usr/lib/jvm/jdk1.8.0_25
- export HADOOP_INSTALL=/usr/local/hadoop
- export PATH=$PATH:$HADOOP_INSTALL/bin
- export PATH=$PATH:$HADOOP_INSTALL/sbin
- 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_COMMON_LIB_NATIVE_DIR=$HADOOP_INSTALL/lib/native
- export HADOOP_OPTS="-Djava.library.path=$HADOOP_INSTALL/lib"
- #HADOOP VARIABLES END
如果不知道JAVA_HOME可以通过命令获得:目录取到java根目录即可。
- update-alternatives --config java
执行下面命令使改动生效:
- source ~/.bashrc
- 修改hadoop-env.sh的配置:
- sudo gedit /usr/local/hadoop/etc/hadoop/hadoop-env.sh
找到JAVA_HOME改为上面的值。
测试
- 通过执行hadoop自带实例WordCount验证是否安装成功
/usr/local/hadoop路径下创建input文件夹
在hadoop目录下执行WordCount:
- mkdir input
- cp README.txt input
- bin/hadoop jar share/hadoop/mapreduce/sources/hadoop-mapreduce-examples-2.6.0-sources.jar
- org.apache.hadoop.examples.WordCount input output
Hadoop伪分布式配置
接着初始化文件系统HDFS。sudo gedit /usr/local/hadoop/etc/hadoop/core-site.xml
- <configuration>
- <property>
- <name>hadoop.tmp.dir</name>
- <value>/usr/local/hadoop/tmp</value>
- <description>Abase for other temporary directories.</description>
- </property>
- <property>
- <name>fs.defaultFS</name>
- <value>hdfs://localhost:9000</value>
- </property>
- </configuration>
sudo gedit /usr/local/hadoop/etc/hadoop/mapred-site.xml //此项不必要
- <configuration>
- <property>
- <name>mapred.job.tracker</name>
- <value>localhost:9001</value>
- </property>
- </configuration>
sudo gedit /usr/local/hadoop/etc/hadoop/yarn-site.xml
- <configuration>
- <property>
- <name>mapreduce.framework.name</name>
- <value>yarn</value>
- </property>
- <property>
- <name>yarn.nodemanager.aux-services</name>
- <value>mapreduce_shuffle</value>
- </property>
- </configuration>
sudo gedit /usr/local/hadoop/etc/hadoop/hdfs-site.xml
- <configuration>
- <property>
- <name>dfs.replication</name>
- <value>1</value>
- </property>
- <property>
- <name>dfs.namenode.name.dir</name>
- <value>file:/usr/local/hadoop/dfs/name</value>
- </property>
- <property>
- <name>dfs.datanode.data.dir</name>
- <value>file:/usr/local/hadoop/dfs/data</value>
- </property>
- <property> //这个属性节点是为了防止后面eclopse存在拒绝读写设置的
- <name>dfs.permissions</name>
- <value>false</value>
- </property>
- </configuration>
sudo gedit /usr/local/hadoop/etc/hadoop/
masters 添加:localhost
sudo gedit /usr/local/hadoop/etc/hadoop/
slaves
添加:localhost
关于配置的一点说明:上面只要配置 fs.defaultFS 和 dfs.replication 就可以运行,不过有个说法是如没有配置 hadoop.tmp.dir 参数,此时 Hadoop 默认的使用的临时目录为 /tmp/hadoo-hadoop,而这个目录在每次重启后都会被干掉,必须重新执行 format 才行(未验证),所以伪分布式配置中最好还是设置一下。
配置完成后,首先在 Hadoop 目录下创建所需的临时目录:
- cd /usr/local/hadoop
- mkdir tmp dfs dfs/name dfs/data
成功的话,最后的提示如下,
- bin/hdfs namenode -format
Exitting with status 0
表示成功,Exitting with status 1:
则是出错。
- sbin/start-dfs.sh
- sbin/start-yarn.sh
Unable to load native-hadoop library for your platform这个提示,解决方式:
1、重新编译源码后将新的lib/native替换到集群中原来的lib/native
2、修改hadoop-env.sh ,增加
export HADOOP_OPTS="-Djava.library.path=$HADOOP_PREFIX/lib:$HADOOP_PREFIX/lib/native"Namenode information:http://localhost:50070来查看Hadoop的信息。
All Applications:http://http://2xx.81.8x.1xx:8088/,将其中的2xx.81.8x.1xx替换为你的实际IP地址。
运行例子:
1.先在hdfs上建个文件夹 bin/hdfs dfs -mkdir -p /user/ha1/input
bin/hdfs dfs -mkdir -p /user/ha1/output
2.上传一些文件:bin/hdfs dfs -put etc/hadoop/ /user/ha1/input 把etc/hadoop文件上传到hdfs的/user/ha1/input中
3.执行指令
bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.6.0.jar grep /user/ha1/input/hadoop /user/ha1/output/temp 'dfs[a-z.]+'
4.查看结果
bin/hdfs dfs -cat /user/ha1/output/temp/*