Hadoop 2.7.3 分布式集群安装

1. 集群规划：

192.168.1.252 palo252 Namenode+Datanode
192.168.1.253 palo253 YarnManager+Datanode+SecondaryNameNode
192.168.1.254 palo254 Datanode

2. 设定固定IP地址

vi /etc/sysconfig/network-scripts/ifcfg-eth0

TYPE=Ethernet
BOOTPROTO=static
DEFROUTE=yes
NAME=eth0
UUID=7ac09286-c35b-4f15-a9ba-701c093832bf
DEVICE=eth0
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
IPV6_PEERDNS=yes
IPV6_PEERROUTES=yes
IPV6_PRIVACY=no
ONBOOT=yes
DNS1=192.168.1.1
IPADDR=192.168.1.252 #三台机器都要分别设置
PREFIX=
GATEWAY=192.168.1.1

3. 修改主机名：
192.168.1.252

hostnamectl set-hostname palo252
hostnamectl --static set-hostname palo252

192.168.1.253

hostnamectl set-hostname palo253
hostnamectl --static set-hostname palo253

192.168.1.254

hostnamectl set-hostname palo254
hostnamectl --static set-hostname palo254

4. 修改hosts文件

vi /etc/hosts

127.0.0.1 localhost
:: localhost
 
192.168.1.252 palo252
192.168.1.253 palo253
192.168.1.254 palo254

5. 安装JDK(所有节点）
具体到oracle官网下载

6. SSH免密登录

Precondition: install ssh server if not avalible

#install ssh client and ssh-server
sudo yum install -y openssl openssh-server
#enable ssh server to start at system start up
systemctl enable sshd.service
#start ssh server service
systemctl start sshd.service

A） 每台机器生成访问秘钥，复制到192.168.1.252:/home/workspace目录下
192.168.1.252：

ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
cp ~/.ssh/authorized_keys /home/workspace/authorized_keys252
rm -rf ~/.ssh/authorized_keys #删除公钥文件

192.168.1.253：

ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
scp ~/.ssh/authorized_keys 192.168.1.252:/home/workspace/authorized_keys253
rm -rf ~/.ssh/authorized_keys #删除公钥文件

192.168.1.254：

ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
scp ~/.ssh/authorized_keys 192.168.1.252:/home/workspace/authorized_keys254
rm -rf ~/.ssh/authorized_keys #删除公钥文件

B） 在192.168.1.252上将所有的公钥合并成一个公钥文件

cat /home/workspace/authorized_keys252 >> /home/workspace/authorized_keys
cat /home/workspace/authorized_keys253 >> /home/workspace/authorized_keys
cat /home/workspace/authorized_keys254 >> /home/workspace/authorized_keys

C） 将合并后的公钥文件复制到集群中的各个主机中

scp /home/workspace/authorized_keys 192.168.1.253:~/.ssh/
scp /home/workspace/authorized_keys 192.168.1.254:~/.ssh/
cp /home/workspace/authorized_keys ~/.ssh/ #因为目前在252主机中，所以使用的命令为cp而不是scp

注：也可以借助 ssh-copy-id -i ~/.ssh/id_rsa.pub  {ip or hostname}来往远程机器复制公钥

以本集群的配置为例，以上ABC三步的操作亦可以通过下面的操作来完成，操作方法如下：

192.168.1.252，192.168.1.253，192168.1.254 均做以下操作，就完成了私钥的生成，公钥的分发

ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa   #生成本机公钥和私钥
ssh-copy-id -i ~/.ssh/id_rsa.pub palo252   #复制本机的公钥到palo252机器上，默认会存储在远程机器的~/.ssh/authorized_keys文件中，如果此文件不存在，会创建该文件
ssh-copy-id -i ~/.ssh/id_rsa.pub palo253   #复制本机的公钥到palo252机器上，默认会存储在远程机器的~/.ssh/authorized_keys文件中，如果此文件不存在，会创建该文件
ssh-copy-id -i ~/.ssh/id_rsa.pub palo254   #复制本机的公钥到palo252机器上，默认会存储在远程机器的~/.ssh/authorized_keys文件中，如果此文件不存在，会创建该文件

D） 每台机器：

chmod  ~                      #当前用户根目录访问权限
chmod  ~/.ssh/                #.ssh目录权限
chmod  ~/.ssh/id_rsa          #id_rsa的访问权限
chmod  ~/.ssh/id_rsa.pub      #id_rsa.pub的访问权限
chmod  ~/.ssh/authorized_keys #authorized_keys的访问权限

说明：

如果ssh 登录的时候失败或者需要密码才能登陆，可以查看sshd的日志信息。日志信息目录为，/var/log/secure
你会发现如下字样的日志信息。
Jul 22 14:20:33 v138020.go sshd[4917]: Authentication refused: bad ownership or modes for directory /home/edw

则需要设置权限：sshd为了安全，对属主的目录和文件权限有所要求。如果权限不对，则ssh的免密码登陆不生效。

用户目录权限为 755 或者 700，就是不能是77x。
.ssh目录权限一般为755或者700。
rsa_id.pub 及authorized_keys权限一般为644
rsa_id权限必须为600

可通过来查看ssh过程中的日志.

cat /var/log/secure

7. 配置hadoop
7-1) 解压
下载地址：https://archive.apache.org/dist/hadoop/common/hadoop-2.7.3/hadoop-2.7.3.tar.gz

tar xzvf hadoop-2.7..tar.gz -C /opt/

7-2) 创建存放数据的目录（必须事先创建好，否则会报错)

mkdir -p /opt/hadoop-2.7./data/full/tmp/
mkdir -p /opt/hadoop-2.7./data/full/tmp/dfs/name
mkdir -p /opt/hadoop-2.7./data/full/tmp/dfs/data

7-3) 配置/opt/hadoop-2.7.3/etc/hadoop下面的配置文件

cd opt/hadoop-2.7./etc/hadoop #定位到配置文件目录

7-3-1) core-site.xml

<configuration>
    <!-- 指定HDFS中NameNode的地址 -->
    <property>
        <name>fs.defaultFS</name>
        <value>hdfs://127.0.0.1:9000</value>
        <description>hdfs://127.0.0.1:9000</description>
    </property>
    <!-- 指定hadoop运行时产生文件的存储目录 -->
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/home/lenmom/workspace/software/hadoop-2.7.3/data/tmp</value>
        <description>是hadoop文件系统依赖的基础配置，很多路径都依赖它。如果hdfs-site.xml中不配 置namenode和datanode的存放位置，默认就放在这个路径中</description>
    </property>
    <!--启用 webhdfs-->
    <property>
        <name>dfs.webhdfs.enabled</name>
        <value>true</value>
        <description>启用 webhdfs</description>
    </property>
 
    <!--use hadoop native library-->
    <property>
        <name>hadoop.native.lib</name>
        <value>true</value>
        <description>Should native hadoop libraries, if present, be used.</description>
    </property>
</configuration>

7-3-2) yarn-site.xml

<configuration>
    <property>
         <!-- reducer获取数据的方式 -->
        <name>yarn.nodemanager.aux-services</name>
        <value>mapreduce_shuffle</value>
    </property>
    <property>
         <!-- 指定YARN的ResourceManager的地址 -->
        <name>yarn.resourcemanager.hostname</name>
        <value>palo253</value>
    </property>
    <property>
        <name>yarn.resourcemanager.address</name>
        <value>palo253:8032</value>
    </property>
    <property>
        <name>yarn.resourcemanager.scheduler.address</name>
        <value>palo253:8030</value>
    </property>
    <property>
        <name>yarn.resourcemanager.resource-tracker.address</name>
        <value>palo253:8031</value>
    </property>
    <property>
        <name>yarn.nodemanager.resource.memory-mb</name>
        <value>10240</value>
    </property>
    <property>
        <name>yarn.scheduler.minimum-allocation-mb</name>
        <value>1024</value>
    </property>
    <property>
        <name>yarn.nodemanager.vmem-pmem-ratio</name>
        <value>2.1</value>
    </property>
</configuration>

7-3-3) slaves

palo252
palo253
palo254

7-3-4) mapred-site.xml

<configuration>
<!-- 指定mr运行在yarn上 -->
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
<property>
<name>mapreduce.jobhistory.address</name>
<value>palo252:</value>
</property>
<property>
<name>mapreduce.jobhistory.webapp.address</name>
<value>palo252:</value>
</property>
</configuration>

7-3-5) hdfs-site.xml

<configuration>
     <property>
          <name>dfs.replication</name>
          <value>1</value>
          <description>不能大于datanode的数量,默认为3</description>
     </property>
     <!-- 设置secondname的端口 -->
     <property>
          <name>dfs.namenode.secondary.http-address</name>
          <value>palo253:50090</value>
     </property>
     <property>
          <name>dfs.data.dir</name>
          <value>file:/opt/hadoop-2.7.3/data/full/tmp/dfs/data</value>
          <description>用于确定将HDFS文件系统的数据保存在什么目录下,可以将这个参数设置为多个分区上目录，即可将HDFS建立在不同分区上。</description>
     </property>
     <property>
          <name>dfs.name.dir</name>
          <value>file:/opt/hadoop-2.7.3/data/full/tmp/dfs/name</value>
          <description>这个参数用于确定将HDFS文件系统的元信息保存在什么目录下,如果这个参数设置为多个目录，那么这些目录下都保存着元信息的多个备份.</description>
     </property>
 
     <!--设置 hadoop的代理用户-->
     <property>
          <name>hadoop.proxyuser.hadoop.hosts</name>
          <value>*</value>
          <description>配置成*的意义，表示任意节点使用 hadoop 集群的代理用户hadoop 都能访问 hdfs 集群</description>
     </property>
     <property>
          <name>hadoop.proxyuser.hadoop.groups</name>
          <value>*</value>
          <description>代理用户所属的组</description>
     </property>
</configuration>

7-3-6) hadoop-env.sh

配置JAVA_HOME

   # Licensed to the Apache Software Foundation (ASF) under one
   # or more contributor license agreements.  See the NOTICE file
   # distributed with this work for additional information
   # regarding copyright ownership.  The ASF licenses this file
   # to you under the Apache License, Version 2.0 (the
   # "License"); you may not use this file except in compliance
   # with the License.  You may obtain a copy of the License at
   #
   #     http://www.apache.org/licenses/LICENSE-2.0
  #
  # Unless required by applicable law or agreed to in writing, software
  # distributed under the License is distributed on an "AS IS" BASIS,
  # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  # See the License for the specific language governing permissions and
  # limitations under the License.
 
  # Set Hadoop-specific environment variables here.
 
  # The only required environment variable is JAVA_HOME.  All others are
  # optional.  When running a distributed configuration it is best to
  # set JAVA_HOME in this file, so that it is correctly defined on
  # remote nodes.
 
  # The java implementation to use.
  #export JAVA_HOME=${JAVA_HOME}
  export JAVA_HOME=/usr/lib/jvm/java-1.8.-openjdk-1.8.0.181-.b13.el7_5.x86_64

8. 配置环境变量(每台机器都必须做)

vi /etc/profile

在文件尾部添加：

#####set jdk enviroment
export JAVA_HOME=/usr/java/jdk1..0_172-amd64
export JRE_HOME=$JAVA_HOME/jre
export CLASSPATH=$JAVA_HOME/lib:$JRE_HOME/lib:$CLASSPATH
export PATH=$JAVA_HOME/bin:$JRE_HOME/bin:$PATH
 
##### set hadoop_home enviroment
export HADOOP_HOME=/opt/hadoop-2.7.
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib"
export HADOOP_CONF_DIR=${HADOOP_HOME}/etc/hadoop
export YARN_HOME=/home/lenmom/workspace/software/hadoop-2.7.
export YARN_CONF_DIR=${YARN_HOME}/etc/hadoop
export PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin
###enable hadoop native library
export LD_LIBRARY_PATH=$HADOOP_HOME/lib/native 

命令行终端执行 source  /etc/profile，让配置的环境变量生效

source /etc/profile ####make the env variable to take effect right now.

9. 启动：

NameNode:(master 252)
#格式化namenode

hdfs namenode -format

#启动dfs　

start-dfs.sh # (master )

#启动Yarn：　 yarn节点(253)
#注意：Namenode和ResourceManger如果不是同一台机器，
#不能在NameNode上启动 yarn，
#应该在ResouceManager所在的机器上启动yarn。

start-yarn.sh

#验证启动情况：
jps #查看java进程
http://namenode:50070/

10 Hadoop启动停止方式

）各个服务组件逐一启动
分别启动hdfs组件： hadoop-daemon.sh  start|stop  namenode|datanode|secondarynamenode
   启动yarn：     yarn-daemon.sh    start|stop  resourcemanager|nodemanager
 
）各个模块分开启动（配置ssh是前提）常用
start|stop-dfs.sh     start|stop-yarn.sh
 
）全部启动（不建议使用）
start|stop-all.sh
 
4) 开启historyserver(任意节点启动即可)
mr-jobhistory-daemon.sh start|stop  historyserver

reference:

1. https://www.cnblogs.com/baierfa/p/6689022.html