CentOS 7 hadoop 3.13 高可用集群安装部署

发表评论
531 views

CentOS 7 hadoop 3.13 高可用集群安装部署

一、服务器规划

本次部署有5台服务器，namenode主备两台，datanode三台，具体规划如下表：

服务器IP 域名 角色 备注
192.168.31.218 HD01.MYHDFS.COM namenode,resourcemanager
192.168.31.153 HD02.MYHDFS.COM datanoade,nodemanager,zoomkeeper,journal
192.168.31.236 HD03.MYHDFS.COM datanode,nodemanager,zoomkeeper,journal
192.168.31.61 HD04.MYHDFS.COM datanode,nodemanager,zoomkeeper,journal
192.168.31.11 HD05.MYHDFS.COM namenode,resourcemanager

二、前置配置

1、主机host配置

在每台主机/etc/hosts 添加一下hosts 绑定，方便后面用主机访问

192.168.31.218 HD01 HD01.MYHDFS.COM
192.168.31.153 HD02 HD02.MYHDFS.COM
192.168.31.236 HD03 HD03.MYHDFS.COM
192.168.31.61 HD04 HD04.MYHDFS.COM
192.168.31.11 HD05 HD05.MYHDFS.COM

2、创建hadoop用户

useradd hadoop
echo "hadoop" | passwd --stdin hadoop
# 将hadoop添加到root组
usermode -g root hadoop

3、配置ssh免密码登陆

# 在每台主机都进行如下操作，以HD01为例
su - hadoop
ssh-keygen
ssh-copy-id -i /home/hadoop/.ssh/id_rsa.pub  hadoop@HD01
ssh-copy-id -i /home/hadoop/.ssh/id_rsa.pub  hadoop@HD02
ssh-copy-id -i /home/hadoop/.ssh/id_rsa.pub  hadoop@HD03
ssh-copy-id -i /home/hadoop/.ssh/id_rsa.pub  hadoop@HD04
ssh-copy-id -i /home/hadoop/.ssh/id_rsa.pub  hadoop@HD05

4、安装并配置java环境

wget https://tool-box.oss-cn-hangzhou.aliyuncs.com/jdk-8u111-linux-x64.tar.gz
tar zxvf jdk-8u111-linux-x64.tar.gz -C  /usr/local
cat  >> /etc/profile << EOF
export JAVA_HOME=/usr/local/jdk1.8.0_111
export JRE_HOME=${JAVA_HOME}/jre
export CLASSPATH=.:${JAVA_HOME}/lib:${JRE_HOME}/lib
export PATH={JAVA_HOME}/bin:$PATH
EOF
source /etc/profile

三、ZOOKEEPER安装

1、下载安装

su - hadoop
wget https://tool-box.oss-cn-hangzhou.aliyuncs.com/apache-zookeeper-3.5.6-bin.tar.gz
tar zxvf apache-zookeeper-3.5.6-bin.tar.gz -C /home/hadoop/
ln -s apache-zookeeper-3.5.6-bin zookeeper
mkdir -p /home/hadoop/zookeeper/data
mkdir -p /home/hadoop/zookeeper/log

修改zoo.cfg  如下：
echo >> /home/hadoop/zookeeper/conf/zoo.cfg << EOF
tickTime=2000
initLimit=10
syncLimit=5
dataDir=/home/hadoop/zookeeper/data
dataLogDir=/home/hadoop/zookeeper/log
clientPort=2181
server.1=HD02:2888:3888
server.2=HD03:2888:3888
server.3=HD04:2888:3888
EOF

2、配置ID

# HD02上
echo 1 > /home/hadoop/zookeeper/data/myid
# HD03上
echo 2 > /home/hadoop/zookeeper/data/myid
# HD04上
echo 3 > /home/hadoop/zookeeper/data/myid

3、启动zookeeper集群

# HD02上
su - hadoop
/home/hadoop/zookeeper/bin/zkServer.sh start
# 查看状态
/home/hadoop/zookeeper/bin/zkServer.sh status
# HD03上
su - hadoop
/home/hadoop/zookeeper/bin/zkServer.sh start
# 查看状态
/home/hadoop/zookeeper/bin/zkServer.sh status
# HD04上
su - hadoop
/home/hadoop/zookeeper/bin/zkServer.sh start
# 查看状态
/home/hadoop/zookeeper/bin/zkServer.sh status

四、安装部署HADOOP

1、下载安装（只在HD01）

su - haddop
wget https://tool-box.oss-cn-hangzhou.aliyuncs.com/hadoop-3.1.3.tar.gz
tar zxvf hadoop-3.1.3.tar.gz -C /home/hadoop/
ln -s hadoop-3.1.3 hadoop

2、在namenode（HD01）主节点配置

2.1 配置core-site.yml

cat > /home/hadoop/hadoop/etc/hadoop/core-site.xml < EOF
<configuration>
    <property>
        <name>fs.defaultFS</name>
        <value>hdfs://mycluster</value>
    </property>
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/home/hadoop/hadoop/data</value>
    </property>
    <property>
        <name>ha.zookeeper.quorum</name>
        <value>HD02:2181,HD03:2181,HD04:2181</value>
    </property>
</configuration>
EOF

2.2 配置 hdfs-site.yml

cat > /home/hadoop/hadoop/etc/hadoop/hdfs-site.xml < EOF
<configuration>
         <!--指定hdfs的nameservice为mycluster，需要和core-site.xml中的保持一致 -->
         <property>
                <name>dfs.nameservices</name>
                <value>mycluster</value>
         </property>
         <!-- mycluster下面有两个NameNode，分别是hdfs01，hdfs05 -->
         <property>
                <name>dfs.ha.namenodes.mycluster</name>
                <value>HD01,HD05</value>
        </property>
        <!-- hdfs01的RPC通信地址 -->
        <property>
                <name>dfs.namenode.rpc-address.mycluster.HD01</name>
                <value>HD01:9000</value>
        </property>
        <!-- hdfs05的RPC通信地址 -->
        <property>
                <name>dfs.namenode.rpc-address.mycluster.HD05</name>
                <value>HD05:9000</value>
        </property>
        <!-- hdfs01的http通信地址 -->
        <property>
                <name>dfs.namenode.http-address.mycluster.HD01</name>
                <value>HD01:50070</value>
        </property>
        <!-- hdfs05的http通信地址 -->
        <property>
                <name>dfs.namenode.http-address.mycluster.HD05</name>
                <value>HD05:50070</value>
        </property>
        <!-- 指定NameNode的元数据在JournalNode上的存放位置 -->
        <property>
                <name>dfs.namenode.shared.edits.dir</name>
                <value>qjournal://HD02:8485;HD03:8485;HD04:8485/mycluster</value>
        </property>
        <property>
                <name>dfs.datanode.data.dir</name>
                <value>/home/hadoop/mnt/disk01,/home/hadoop/mnt/disk02,/home/hadoop/mnt/disk03,/home/hadoop/mnt/disk04,/home/hadoop/mnt/disk05</value>
                </property>
        <!-- 指定JournalNode在本地磁盘存放数据的位置 -->
        <property>
                <name>dfs.journalnode.edits.dir</name>
                <value>/home/hadoop/hadoop/data/journal</value>
        </property>
        <!-- 开启NameNode失败自动切换 -->
        <property>
                <name>dfs.ha.automatic-failover.enabled</name>
                <value>true</value>
        </property>
        <!-- 配置失败自动切换实现方式 -->
        <property>
                <name>dfs.client.failover.proxy.provider.mycluster</name>
                <value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
        </property>
        <!-- 配置隔离机制方法，多个机制用换行分割，即每个机制暂用一行-->
        <property>
                <name>dfs.ha.fencing.methods</name>
                <value>
                        shell(/bin/true)
                </value>
        </property>

        <!-- 设置副本数-->
         <property>
                <name>dfs.replication</name>
                <value>2</value>
         </property>
        <!-- 设置块大小-->
         <property>
                 <name>dfs.block.size</name>
                 <value>64M</value>
         </property>
</configuration>
EOF

2.3 配置 yarn-site.yml

cat > /home/hadoop/hadoop/etc/hadoop/yarn-site.xml < EOF
<configuration>
        <!-- 开启RM高可靠 -->
        <property>
                <name>yarn.resourcemanager.ha.enabled</name>
                <value>true</value>
        </property>
        <!-- 指定RM的cluster id -->
        <property>
                <name>yarn.resourcemanager.cluster-id</name>
                <value>cluster1</value>
        </property>
        <!-- 指定RM的名字 -->
        <property>
                <name>yarn.resourcemanager.ha.rm-ids</name>
                <value>rm1,rm2</value>
        </property>
        <!-- 分别指定RM的地址 -->
        <property>
                <name>yarn.resourcemanager.hostname.rm1</name>
                <value>HD01</value>
        </property>
        <property>
                <name>yarn.resourcemanager.hostname.rm2</name>
                <value>HD05</value>
        </property>
        <property>
                <name>yarn.resourcemanager.recovery.enabled</name>
                <value>true</value>
        </property>

        <property>
                <name>yarn.resourcemanager.store.class</name>
                <value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value>
        </property>
        <!-- 指定zk集群地址 -->
        <property>
                <name>yarn.resourcemanager.zk-address</name>
                <value>HD02:2181,HD03:2181,HD04:2181</value>
        </property>

        <property>
                <name>yarn.nodemanager.aux-services</name>
                <value>mapreduce_shuffle</value>
        </property>
</configuration>
EOF

2.4 配置worker

cat >> /home/hadoop/hadoop/etc/hadoop/worker << EOF
HD02
HD03
HD04
EOF

2.5 初始化HA在zookeeper上的状态(HD01上执行)

su - hadoop
/home/hadoop/hadoop/bin/hdfs zkfc -formatZK

3、将hadoop 同步到其他4台主机

scp -r /home/hadoop/hadoop-3.13 hadoop@HD02:~/
scp -r /home/hadoop/hadoop-3.13 hadoop@HD03:~/
scp -r /home/hadoop/hadoop-3.13 hadoop@HD04:~/
scp -r /home/hadoop/hadoop-3.13 hadoop@HD05:~/

4、在HD02,HD03,HD04上分别启动journal

/home/hadoop/hadoop-3.1.3/sbin/hadoop-daemon.sh start journalnode
# 用jps命令查看状态
[hadoop@HD02 ~]$ jps
14374 Jps
9801 JournalNode
8410 QuorumPeerMain

5、初始化namenode

# 在HD01（主namenode）
su - hadoop
/home/hadoop/hadoop/bin/hdfs namenode -format mycluster

# 在HD05（备namenode）
su - hadoop
/home/hadoop/hadoop/bin/hdfs namenode -bootstrapStandby

6、启动namenode、datanode

# 在HD01上
/home/hadoop/hadoop/sbin/start-dfs.sh

7、启动reourcemanager、nodemanager

# 在HD01上
/home/hadoop/hadoop/sbin/start-yarn.sh

8、验证namenode、resourcemanager

在浏览器输入http://HD01:50070 ,http://HD05:50070, http://HD01:8088

五、各服务总结