《OD大数据实战》Hadoop伪分布式环境搭建

一、安装并配置Linux

8. 使用当前root用户创建文件夹，并给/opt/下的所有文件夹及文件赋予775权限，修改用户组为当前用户

mkdir -p /opt/modules
mkdir -p /opt/software
mkdir -p /opt/datas
mkdir -p /opt/tools
chmod  /opt/*
chown beifeng:beifeng /opt/*

最终效果如下：

[beifeng@beifeng-hadoop- opt]$ pwd
/opt
[beifeng@beifeng-hadoop- opt]$ ll
total
drwxrwxr-x.   beifeng beifeng  Jul  : clusterapps
drwxr-xr-x.  beifeng beifeng  Jul  : datas
drwxr-xr-x.   beifeng beifeng  Jul  : modules
drwxr-xr-x.   beifeng beifeng  Jul  : software
drwxr-xr-x.   beifeng beifeng  Jul  : tools

二、安装并配置JDK

1. 安装文件

jdk-7u67-linux-x64.tar.gz

2. 解压

tar -zxvf jdk-7u67-linux-x64.tar.gz -C /opt/modules

3. 配置jdk

1）使用sudo配置/etc/profile，在文件尾加上以下配置

#JAVA_HOME
export JAVA_HOME=/opt/modules/jdk1..0_67
export PATH=$PATH:$JAVA_HOME/bin

2）配置完成后，使用su - root 切换到root用户，使用source命令生效配置。

source /etc/profile

3）验证jdk是否安装成功

[root@beifeng-hadoop- ~]# java -version
java version "1.7.0_67"
Java(TM) SE Runtime Environment (build 1.7.0_67-b01)
Java HotSpot(TM) -Bit Server VM (build 24.65-b04, mixed mode)
[root@beifeng-hadoop- ~]# javac -version
javac 1.7.0_67

三、安装并配置hadoop

1. 安装文件

下载地址：http://archive.cloudera.com/cdh5/cdh/5/

下载： hadoop-2.5.0-cdh5.3.6.tar.gz

2. 解压

tar -zxvf hadoop-2.5.-cdh5.3.6.tar.gz -C /opt/modules/cdh/

3. 配置伪分布式环境

参考文档： http://hadoop.apache.org/docs/r2.5.2/hadoop-project-dist/hadoop-common/ClusterSetup.html

cd /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/etc/hadoop

修改/etc/profile，在文件尾增加以下配置：

#HADOOP_HOME
export HADOOP_HOME=/opt/modules/cdh/hadoop-2.5.-cdh5.3.6
export PATH=$PATH:$HADOOP_HOME/bin
export PATH=$PATH:$HADOOP_HOME/sbin
export HADOOP_MAPRED_HOME=$HADOOP_HOME
export HADOOP_COMMON_HOME=$HADOOP_HOME
export HADOOP_HDFS_HOME=$HADOOP_HOME
export YARN_HOME=$HADOOP_HOME
export HADOOP_COMMON_LIB_NATIE_DIR=$HADOOP_HOME/lib/native
export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib"

建议使用远程sftp编辑工具，windows上可以使用notepad++,mac上推荐使用skEdit。

1）修改hadoop-evn.sh

export JAVA_HOME=/opt/modules/jdk1.7.0_67

2）修改yarn-env.sh

export JAVA_HOME=/opt/modules/jdk1.7.0_67

3）修改mapred-env.sh

export JAVA_HOME=/opt/modules/jdk1.7.0_67

4）修改core-site.xml

<configuration>
    <property>
        <name>fs.defaultFS</name>
        <value>hdfs://beifeng-hadoop-02:9000</value>
    </property>
     <property>
         <name>hadoop.tmp.dir</name>
         <value>/opt/modules/cdh/hadoop-2.5.0-cdh5.3.6/data/tmp</value>
     </property>
     <property>
          <name>hadoop.http.staticuser.user</name>
          <value>beifeng</value>
     </property>
</configuration>

5）修改hdfs-site.xml

<configuration>

        <!-- 数据副本数，副本数等于所有datanode的总和 -->
        <property>
                <name>dfs.replication</name>
                <value>1</value>
        </property>

        <property>
                <name>dfs.namenode.secondary.http-address</name>
                <value>beifeng-hadoop-02:50090</value>
        </property>

        <property>
                <name>dfs.permissions.enabled</name>
                <value>false</value>
        </property>

</configuration>

6）修改slaves

beifeng-hadoop-

7）修改yarn-site.xml

<configuration>

<!-- Site specific YARN configuration properties -->
        <property>
                <name>yarn.nodemanager.aux-services</name>
                <value>mapreduce_shuffle</value>
        </property>

        <property>
                <name>yarn.resourcemanager.hostname</name>
                <value>beifeng-hadoop-02</value>
        </property>

        <!-- 是否启用日志聚集功能 -->
        <property>
                <name>yarn.log-aggregation-enable</name>
                <value>true</value>
        </property>

        <!-- 日志保留时间(单位为秒) -->
        <property>
                <name>yarn.log-aggregation.retain-seconds</name>
                <value>106800</value>
        </property>
</configuration>

8） 修改mapred-site.xml

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

9）启动服务

（1）格式化hdfs

bin/hdfs namenode -format

（2）启动namenode和datanode

sbin/hadoop-daemon.sh start namenode
sbin/hadoop-daemon.sh start datanode

使用jps命令，或者web UI界面查看namenode是否已启动成功。

[beifeng@beifeng-hadoop- hadoop-2.5.-cdh5.3.6]$ jps
 DataNode
 Jps
 NameNode

hdfs可视化界面： http://beifeng-hadoop-02:50070/dfshealth.html#tab-overview

（2）启动resourcemanager和nodemanager

sbin/yarn-daemon.sh start resourcemanager
sbin/yarn-daemon.sh start nodemanager

使用jps命令，或者web UI界面查看resourcemanager和nodemanager是否已成功启动

[beifeng@beifeng-hadoop- hadoop-2.5.-cdh5.3.6]$ jps
 DataNode
 NodeManager
 Jps
 NameNode
 ResourceManager

yarn可视化界面： http://beifeng-hadoop-02:8088/cluster

（3）启动job历史服务器

sbin/mr-jobhistory-daemon.sh start historyserver

查看是否已成功启动：

历史服务器可视化界面：http://beifeng-hadoop-02:19888/

（4）启动secondarynamenode

sbin/hadoop-daemon.sh start secondarynamenode

查看是否已成功启动：

secondarynamenode可视化界面 http://beifeng-hadoop-02:50090/status.html

（5）所有相关服务停止命令

sbin/hadoop-daemon.sh stop namenode
sbin/hadoop-daemon.sh stop datanode
sbin/yarn-daemon.sh stop resourcemanager
sbin/yarn-daemon.sh stop nodemanager
sbin/mr-jobhistory-daemon.sh stop historyserver
sbin/hadoop-daemon.sh stop secondarynamenode

10）跑一个wordcount 验证环境搭建结果

文件系统shell：http://archive.cloudera.com/cdh5/cdh/5/hadoop-2.5.0-cdh5.3.6/hadoop-project-dist/hadoop-common/FileSystemShell.html

hdfs dfs -mkdir -p /user/beifeng/input

hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.5.-cdh5.3.6.jar wordcount /user/beifeng/input /user/beifeng/output 

hdfs dfs -cat /user/beifeng/output/part-r-

四、给Hadoop2.x添加Snappy解压缩库

1. 修改配置

1）修改core-site.xml

     <!-- SNAPPY compress -->
     <property>
         <name>io.compression.codecs</name>
         <value>org.apache.hadoop.io.compress.GzipCodec,
                 org.apache.hadoop.io.compress.DefaultCodec,
                 org.apache.hadoop.io.compress.BZip2Codec,
                 org.apache.hadoop.io.compress.SnappyCodec
        </value>
        <description>A comma-separated list of the compression codec classes that can
            be used for compression/decompression. In addition to any classes
            specified with this property (which take precedence), codec classes on the classpath are discovered
            using a Java ServiceLoader.
        </description>
    </property>

2）修改mapred-site.xml

    <!-- 开启 MapReduce map 输出结果压缩功能 -->
    <property>
        <name>mapreduce.map.output.compress</name>
        <value>true</value>
    </property>
    <property>
        <name>mapreduce.map.output.compress.codec</name>
        <value>org.apache.hadoop.io.compress.SnappyCodec</value>
    </property>

2. 安装snappy

1）解压

tar -zxvf snappy-1.1..tar.gz -C /opt/modules/cdh/

cd /opt/modules/cdh/snappy-1.1.

2）预编译

./configure

3）编译安装

sudo make && sudo make install

4）编译成功后,查看安装目录

cd /usr/local/lib && ls

3. 安装hadoop-snappy

1）解压

tar -zxvf hadoop-snappy.tar.gz -C /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/

2）打包编译

cd /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/hadoop-snappy

mvn package -Dsnappy.prefix=/usr/local

ubuntu安装hadoop常见错误与解决方法

sudo ln -s /opt/modules/jdk1.7.0_67/jre/lib/amd64/server/libjvm.so /usr/local/lib

3）copy 编译好的jar包到hadoop lib下

cp /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/hadoop-snappy/target/hadoop-snappy-0.0.-SNAPSHOT.jar /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/lib

4）修改hadoop-env.sh

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/modules/cdh/hadoop-2.5.-cdh5.3.6/native/Linux-amd64-/

5）编译生成后的动态库 copy 到 $HADOOP_HOME/lib/native/ 目录下

cd /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/hadoop-snappy/target/hadoop-snappy-0.0.-SNAPSHOT-tar/hadoop-snappy-0.0.-SNAPSHOT/lib
cp -r native/Linux-amd64- /opt/modules/cdh/hadoop-2.5.-cdh5.3.6/lib/native/

6）copy Linux-amd64-64 目录下的文件，到/opt/modules/cdh/hadoop-2.5.0-cdh5.3.6/lib/native/

cd Linux-amd64-/

cp -r ./* ../

4. 编译hadoop-2.5.0-cdh5.3.6-src源码

注意.m2/settings.xml文件，使用maven原生的配置，否则无法加载pom

mvn package -Pdist,native -DskipTests -Dtar -Drequire.snappy

执行了一半，磁盘空间不够

http://os.51cto.com/art/201012/240726_all.htm

http://www.cnblogs.com/chenmh/p/5096592.html

http://www.linuxfly.org/post/243/

1）替换 hadoop 安装目录下的 lib/native 目录下的本地库文件

/opt/modules/hadoop-2.5.0-src/hadoop-dist/target/hadoop-2.5.0/lib/native

cp ./* /opt/modules/cdh/hadoop-2.5.0-cdh5.3.6/lib/native/

5. 验证

hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.5.-cdh5.3.6.jar pi  

hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.5.-cdh5.3.6.jar wordcount /user/beifeng/input /user/beifeng/output03 

hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.5.-cdh5.3.6.jar wordcount -Dmapreduce.map.output.compress=true -Dmapreduce.map.output.codec=org.apache.hadoop.io.compress.SnappyCodec /user/beifeng/input /user/beifeng/output02