文章目录
前言
直接从官网上下载的hadoop-2.6.0-cdh5.7.0.tar.gz部署后,不支持压缩,生产上是不可接受的,可以用hadoop checknative命令查看hadoop所支持的本地库:
#这个是直接下载的hadoop包部署的不支持压缩
[hadoop@hadoop001 ~]$ hadoop checknative
19/04/27 10:22:33 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Native library checking:
hadoop: false
zlib: false
snappy: false
lz4: false
bzip2: false
openssl: false
19/04/27 10:22:34 INFO util.ExitUtil: Exiting with status 1
[hadoop@hadoop001 ~]$
可以看出hadoop不支持这些压缩。所以需要从官网上下载源码进行编译让它支持压缩。
编译所需要的环境条件
在源码的文件夹里有个BUILDING.txt文件,讲述了该源码包的编译要求。
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ more BUILDING.txt
Build instructions for Hadoop
----------------------------------------------------------------------------------
Requirements:
* Unix System
* JDK 1.7+
* Maven 3.0 or later
* Findbugs 1.3.9 (if running findbugs)
* ProtocolBuffer 2.5.0
* CMake 2.6 or newer (if compiling native code), must be 3.0 or newer on Mac
* Zlib devel (if compiling native code)
* openssl devel ( if compiling native hadoop-pipes )
* Internet connection for first build (to fetch all Maven and Hadoop dependencies)
本文所需要的环境条件如下
组件名称 | 版本 |
---|---|
centos | centos6.6 |
Hadoop | Hadoop-2.6.0-cdh5.7.0-src.tar.gz |
jdk | jdk-7u80-linux-x64.tar.gz |
maven | apache-maven-3.3.9-bin.tar.gz |
protobuf | protobuf-2.5.0.tar.gz |
需要注意的是:编译的JDK版本必须是1.7版本,1.8的JDK会导致编译失败。
jdk版本可以安装好多个,需要哪个版本的时候在环境变量里面把注释去掉,把其它版本的注释掉就可以了。
大致过程:把以上软件下载好之后,需要部署安装jdk、maven、protobuf、以及相关的依赖,然后用maven编译Hadoop源码,编译之后,部署Hadoop,验证是否可行。
安装依赖库
root用户:
[root@hadoop001 ~]# yum install -y svn ncurses-devel
[root@hadoop001 ~]# yum install -y gcc gcc-c++ make cmake
[root@hadoop001 ~]# yum install -y openssl openssl-devel svn ncurses-devel zlib-devel libtool
[root@hadoop001 ~]# yum install -y snappy snappy-devel bzip2 bzip2-devel lzo lzo-devel lzop autoconf automake cmake
上传软件到系统上
以前已经存在hadoop用户,没有的话创建一下
hadoop用户
#创建三个文件夹:软件、maven仓库、源码
[hadoop@hadoop001 ~]$ mkdir soft maven_repo source
[hadoop@hadoop001 ~]$ cd soft/
[hadoop@hadoop001 soft]$ rz
[hadoop@hadoop001 soft]$ ll
total 202192
-rw-r--r-- 1 hadoop hadoop 8491533 Apr 27 09:19 apache-maven-3.3.9-bin.tar.gz
-rw-r--r-- 1 hadoop hadoop 42610549 Apr 27 09:19 hadoop-2.6.0-cdh5.7.0-src.tar.gz
-rw-r--r-- 1 hadoop hadoop 153530841 Apr 27 09:18 jdk-7u80-linux-x64.tar.gz
-rw-r--r-- 1 hadoop hadoop 2401901 Apr 27 09:18 protobuf-2.5.0.tar.gz
安装JDK
root用户:
解压安装包,安装目录必须是/usr/java,安装后记得修改拥有者为root
[root@hadoop001 ~]# tar -zxvf /home/hadoop/soft/jdk-7u80-linux-x64.tar.gz -C /usr/java/
[root@hadoop001 ~]# cd /usr/java
[root@hadoop001 java]# ll
total 169220
drwxr-xr-x 8 uucp 143 4096 Apr 11 2015 jdk1.7.0_80
drwxr-xr-x 8 root root 4096 Apr 11 2015 jdk1.8.0_45
-rw-r--r-- 1 root root 173271626 Jan 31 00:34 jdk-8u45-linux-x64.gz
[root@hadoop001 java]# chown -R root:root jdk1.7.0_80
[root@hadoop001 java]# ll
total 169220
drwxr-xr-x 8 root root 4096 Apr 11 2015 jdk1.7.0_80
drwxr-xr-x 8 root root 4096 Apr 11 2015 jdk1.8.0_45
-rw-r--r-- 1 root root 173271626 Jan 31 00:34 jdk-8u45-linux-x64.gz
[root@hadoop001 java]#
添加环境变量
因为我的环境之前安装了jdk1.8版本,所以在这里要把这1.8的给注释掉就好了。
[root@hadoop001 java]# vi /etc/profile
#添加如下环境变量
#export JAVA_HOME=/usr/java/jdk1.8.0_45
export JAVA_HOME=/usr/java/jdk1.7.0_80
export JRE_HOME=$JAVA_HOME/jre
export CLASSPATH=.:$JAVA_HOME/lib:$JER_HOME/lib:$CLASSPATH
export PATH=$JAVA_HOME/bin:$JER_HOME/bin:$PATH
#测试java是否安装成功
[root@hadoop001 java]# source /etc/profile
[root@hadoop001 java]# java -version
java version "1.7.0_80"
Java(TM) SE Runtime Environment (build 1.7.0_80-b15)
Java HotSpot(TM) 64-Bit Server VM (build 24.80-b11, mixed mode)
[root@hadoop001 java]#
安装maven
hadoop用户:
[root@hadoop001 java]# su - hadoop
[hadoop@hadoop001 ~]$ tar -zxvf ~/soft/apache-maven-3.3.9-bin.tar.gz -C ~/app/
添加环境变量
#修改haoop用户的环境变量
[hadoop@hadoop001 ~]$ vi ~/.bash_profile
#添加或修改如下内容,注意MAVEN_OPTS设置了maven运行的内存,防止内存太小导致编译失败
export MAVEN_HOME=/home/hadoop/app/apache-maven-3.3.9
export MAVEN_OPTS="-Xms1024m -Xmx1024m"
export PATH=$MAVEN_HOME/bin:$PATH
#测试maven是否安装成功
[hadoop@hadoop001 ~]$ source ~/.bash_profile
[hadoop@hadoop001 ~]$ which mvn
~/app/apache-maven-3.3.9/bin/mvn
[hadoop@hadoop001 ~]$
配置maven
[hadoop@hadoop001 protobuf-2.5.0]$ vim ~/app/apache-maven-3.3.9/conf/settings.xml
#配置maven的本地仓库位置
<localRepository>/home/hadoop/maven_repo/repo</localRepository>
#添加阿里云中央仓库地址,注意一定要写在<mirrors></mirrors>之间,否则后面编译会报错
<mirror>
<id>nexus-aliyun</id>
<mirrorOf>central</mirrorOf>
<name>Nexus aliyun</name>
<url>http://maven.aliyun.com/nexus/content/groups/public</url>
</mirror>
小知识点:网络慢可能导致mvn第一次编译时下载需要超长的时间甚至编译失败。如果有别人或者自己已经建立好的本地仓库,可以把maven本地仓库添加到/home/hadoop/maven_repo/中去。
安装protobuf
hadoop用户:
解压
[hadoop@hadoop001 ~]$ tar -zxvf ~/soft/protobuf-2.5.0.tar.gz -C ~/app/
[hadoop@hadoop001 protobuf-2.5.0]$ ls
aclocal.m4 config.sub depcomp install-sh Makefile.am python
autogen.sh configure editors INSTALL.txt Makefile.in README.txt
CHANGES.txt configure.ac examples java missing src
config.guess CONTRIBUTORS.txt generate_descriptor_proto.sh ltmain.sh protobuf-lite.pc.in vsprojects
config.h.in COPYING.txt gtest m4 protobuf.pc.in
编译protobuf
#./configure --prefix是指定编译好的包要存放的路径,软件要安装的路径
#make 进行编译,make install进行安装
[hadoop@hadoop001 protobuf-2.5.0]$ ./configure --prefix=/home/hadoop/app/protobuf-2.5.0
[hadoop@hadoop001 protobuf-2.5.0]$ make
[hadoop@hadoop001 protobuf-2.5.0]$ make install
添加环境变量
[hadoop@hadoop001 protobuf-2.5.0]$ vi ~/.bash_profile
#追加如下两行内容
export PROTOBUF_HOME=/home/hadoop/app/protobuf-2.5.0
export PATH=$PROTOBUF_HOME/bin:$PATH
#测试是否生效,若出现libprotoc 2.5.0则为生效
[hadoop@hadoop001 protobuf-2.5.0]$ source ~/.bash_profile
[hadoop@hadoop001 protobuf-2.5.0]$ protoc --version
libprotoc 2.5.0
[hadoop@hadoop001 protobuf-2.5.0]$
编译hadoop
hadoop用户:
解压
[hadoop@hadoop001 ~]$ tar -zxvf ~/soft/hadoop-2.6.0-cdh5.7.0-src.tar.gz -C ~/source/
#进入hadoop的源码目录
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ ls
BUILDING.txt hadoop-client hadoop-mapreduce1-project hadoop-project LICENSE.txt
cloudera hadoop-common-project hadoop-mapreduce-project hadoop-project-dist NOTICE.txt
dev-support hadoop-dist hadoop-maven-plugins hadoop-tools pom.xml
hadoop-assemblies hadoop-hdfs-project hadoop-minicluster hadoop-yarn-project README.txt
编译hadoop使其支持压缩:mvn clean package -Pdist,native -DskipTests -Dtar
#进入hadoop的源码目录
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ cd ~/source/hadoop-2.6.0-cdh5.7.0/
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ mvn clean package -Pdist,native -DskipTests -Dtar
#进行编译,第一次编译会下载很多依赖的jar包,快慢由网速决定,需耐心等待
#我是把别人的本地仓库拷贝到我的本地仓库,所以很快
若报异常,主要信息如下(无异常跳过)(这个我没有碰到,是其他人碰到,学习一下):
[FATAL] Non-resolvable parent POM for org.apache.hadoop:hadoop-main:2.6.0-cdh5.7.0: Could not transfer artifact com.cloudera.cdh:cdh-root:pom:5.7.0 from/to cdh.repo (https://repository.cloudera.com/artifactory/cloudera-repos): Remote host closed connectio
#分析:是https://repository.cloudera.com/artifactory/cloudera-repos/com/cloudera/cdh/cdh-root/5.7.0/cdh-root-5.7.0.pom文件下载不了,但是虚拟机确实是ping通远程的仓库,很是费解为什么。
#解决方案:前往本地仓库到目标文件目录,然后 通过wget 文件,来成功获取该文件,重新执行编译命令,或者执行4.5的可选步骤,将需要的jar直接放到本地仓库
编译成功后会有提示:
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
查看编译后的包:hadoop-2.6.0-cdh5.7.0.tar.gz
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ ll /home/hadoop/source/hadoop-2.6.0-cdh5.7.0/hadoop-dist/target/
total 563956
drwxrwxr-x 2 hadoop hadoop 4096 Apr 27 12:42 antrun
drwxrwxr-x 3 hadoop hadoop 4096 Apr 27 12:42 classes
-rw-rw-r-- 1 hadoop hadoop 1998 Apr 27 12:42 dist-layout-stitching.sh
-rw-rw-r-- 1 hadoop hadoop 690 Apr 27 12:42 dist-tar-stitching.sh
drwxrwxr-x 9 hadoop hadoop 4096 Apr 27 12:42 hadoop-2.6.0-cdh5.7.0
-rw-rw-r-- 1 hadoop hadoop 191842632 Apr 27 12:42 hadoop-2.6.0-cdh5.7.0.tar.gz
-rw-rw-r-- 1 hadoop hadoop 7314 Apr 27 12:42 hadoop-dist-2.6.0-cdh5.7.0.jar
-rw-rw-r-- 1 hadoop hadoop 385574240 Apr 27 12:43 hadoop-dist-2.6.0-cdh5.7.0-javadoc.jar
-rw-rw-r-- 1 hadoop hadoop 4855 Apr 27 12:42 hadoop-dist-2.6.0-cdh5.7.0-sources.jar
-rw-rw-r-- 1 hadoop hadoop 4855 Apr 27 12:42 hadoop-dist-2.6.0-cdh5.7.0-test-sources.jar
drwxrwxr-x 2 hadoop hadoop 4096 Apr 27 12:42 javadoc-bundle-options
drwxrwxr-x 2 hadoop hadoop 4096 Apr 27 12:42 maven-archiver
drwxrwxr-x 3 hadoop hadoop 4096 Apr 27 12:42 maven-shared-archive-resources
drwxrwxr-x 3 hadoop hadoop 4096 Apr 27 12:42 test-classes
drwxrwxr-x 2 hadoop hadoop 4096 Apr 27 12:42 test-dir
伪分布式部署
由于我之前已经部署过hadoop-2.6.0-cdh5.7.0,所以这里先把它名字修改一下
hadoop@hadoop001 app]$ mv hadoop-2.6.0-cdh5.7.0 hadoop-2.6.0-cdh5.7.0-20190427bak
解压安装包
[hadoop@hadoop001 target]$ pwd
/home/hadoop/source/hadoop-2.6.0-cdh5.7.0/hadoop-dist/target
[hadoop@hadoop001 target]$ tar -zxvf hadoop-2.6.0-cdh5.7.0.tar.gz -C ~/app
配置环境变量
(因为我之前已经配置过环境变量,所以这里不需要配置,没有配的话,可以这样配一下)
[hadoop@hadoop001 ~]$ vi ~/.bash_profile
export HADOOP_HOME=/home/hadoop/app/hadoop-2.6.0-cdh5.7.0
export PATH=$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$PATH
[hadoop@hadoop001 ~]$ source ~/.bash_profile
[hadoop@hadoop001 ~]$ which hadoop
~/app/hadoop-2.6.0-cdh5.7.0/bin/hadoop
配置ssh
为什么要配置ssh,因为hadoop在启动的时候要用到ssh。
[hadoop@hadoop001 ~]$ rm -rf ~/.ssh
[hadoop@hadoop001 ~]$ ssh-keygen
[hadoop@hadoop001 ~]$ cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
[hadoop@hadoop001 ~]$ chmod 600 ~/.ssh/authorized_keys
#测试ssh是否成功,用户第一ssh会提示输入是否连接,yes。成功显示时间
[hadoop@hadoop001 ~]$ ssh hadoop001 date
修改配置文件
编辑hadoop-env.sh文件相关配置
[hadoop@hadoop001 hadoop]$ pwd
/home/hadoop/app/hadoop-2.6.0-cdh5.7.0/etc/hadoop
[hadoop@hadoop001 hadoop]$ vi hadoop-env.sh
export JAVA_HOME=/usr/java/jdk1.7.0_80
编辑core-site.xml文件相关配置
#这里我创建了hadoop_tmp目录
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ mkdir hadoop_tmp
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ chmod 777 hadoop_tmp
[hadoop@hadoop001 hadoop]$ vi core-site.xml
<configuration>
<property>
<name>fs.defaultFS</name>
<value>hdfs://hadoop001:9000</value>
</property>
#这里我配置了它的临时目录设定,默认是在/tmp/hadoop目录下的,linux会定期清理/tmp目录
<property>
<name>hadoop.tmp.dir</name>
<value>/home/hadoop/app/hadoop-2.6.0-cdh5.7.0/hadoop_tmp</value>
</property>
</configuration>
编辑hdfs-site.xml文件相关配置
[hadoop@hadoop001 hadoop]$ vi hdfs-site.xml
<configuration>
<property>
<name>dfs.replication</name>
<value>1</value>
</property>
<property>
<name>dfs.namenode.secondary.http-address</name>
<value>hadoop001:50090</value>
</property>
<property>
<name>dfs.namenode.secondary.https-address</name>
<value>hadoop001:50091</value>
<property>
</configuration>
编辑mapred-site.xml文件相关配置
[hadoop@hadoop001 hadoop]$ cp mapred-site.xml.template mapred-site.xml
[hadoop@hadoop001 hadoop]$ vi mapred-site.xml
<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration>
编辑yarn-site.xml文件相关配置
[hadoop@hadoop001 hadoop]$ vi yarn-site.xml
<configuration>
<!-- Site specific YARN configuration properties -->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
</configuration>
添加slaves节点
[hadoop@hadoop001 ~]$ vim ~/app/hadoop-2.6.0-cdh5.7.0/etc/hadoop/slaves
#把默认的localhost修改成hadoop001
hadoop001
格式化namenode
[hadoop@hadoop001 ~]$ hdfs namenode -format
......
#若出现 has been successfully formatted 则表示格式化成功
Re-format filesystem in Storage Directory /tmp/hadoop-hadoop/dfs/name ? (Y or N) Y
19/04/27 15:00:42 INFO namenode.FSImage: Allocated new BlockPoolId: BP-835614160-127.0.0.1-1556348442807
19/04/27 15:00:43 INFO common.Storage: Storage directory /tmp/hadoop-hadoop/dfs/name has been successfully formatted.
19/04/27 15:00:43 INFO namenode.NNStorageRetentionManager: Going to retain 1 images with txid >= 0
19/04/27 15:00:43 INFO util.ExitUtil: Exiting with status 0
19/04/27 15:00:43 INFO namenode.NameNode: SHUTDOWN_MSG:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop001/127.0.0.1
************************************************************/
[hadoop@hadoop001 hadoop]$
启动hadoop
[hadoop@hadoop001 sbin]$ ./start-all.sh
This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh
Starting namenodes on [hadoop001]
hadoop001: starting namenode, logging to /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/logs/hadoop-hadoop-namenode-hadoop001.out
hadoop001: starting datanode, logging to /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/logs/hadoop-hadoop-datanode-hadoop001.out
Starting secondary namenodes [hadoop001]
hadoop001: starting secondarynamenode, logging to /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/logs/hadoop-hadoop-secondarynamenode-hadoop001.out
starting yarn daemons
starting resourcemanager, logging to /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/logs/yarn-hadoop-resourcemanager-hadoop001.out
hadoop001: starting nodemanager, logging to /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/logs/yarn-hadoop-nodemanager-hadoop001.out
#查询出五个守护进程
[hadoop@hadoop001 sbin]$ jps
868 ResourceManager
456 NameNode
1353 DataNode
1419 Jps
720 SecondaryNameNode
961 NodeManager
[hadoop@hadoop001 sbin]$
验证hadoop
hdfs验证:
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ hdfs dfs -ls /
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ hdfs dfs -put README.txt hdfs://hadoop001:9000/
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ hdfs dfs -ls /
Found 1 items
-rw-r--r-- 1 hadoop supergroup 1366 2019-04-27 15:39 /README.txt
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$
yarn验证:
#用这个案例验证一下,看一下结果
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ hadoop jar ./share/hadoop/mapreduce/hadoop-mapreduce-examples-2.6.0-cdh5.7.0.jar pi 4 10
检测是否支持压缩
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$ hadoop checknative
19/04/27 16:04:13 INFO bzip2.Bzip2Factory: Successfully loaded & initialized native-bzip2 library system-native
19/04/27 16:04:13 INFO zlib.ZlibFactory: Successfully loaded & initialized native-zlib library
Native library checking:
hadoop: true /home/hadoop/app/hadoop-2.6.0-cdh5.7.0/lib/native/libhadoop.so.1.0.0
zlib: true /lib64/libz.so.1
snappy: true /usr/lib64/libsnappy.so.1
lz4: true revision:99
bzip2: true /lib64/libbz2.so.1
openssl: true /usr/lib64/libcrypto.so
[hadoop@hadoop001 hadoop-2.6.0-cdh5.7.0]$
本文参考:
①主要参考博客:https://blog.csdn.net/qq_32641659/article/details/89074365#commentBox
②关于Hadoop的本地库参考:
http://blog.sina.com.cn/s/blog_3d9e90ad0102wqrp.html
③Protocol Buffer介绍
https://blog.csdn.net/carson_ho/article/details/70568606
④./configure、make、make install 命令详解:
https://www.cnblogs.com/tinywan/p/7230039.html
https://www.cnblogs.com/qiaopei/p/5550168.html