搭建hadoop分布式集群以及大数据开发环境(配置hdfs,yarn,mapreduce等)

网友投稿 280 2022-11-24

搭建hadoop分布式集群以及大数据开发环境(配置hdfs,yarn,mapreduce等)

一、hadoop集群

1.节点

master:

master1:  ip:192.168.75.137

master2:  ip:192.168.75.138

slave:

slave1:  ip:192.168.75.139

slave2:  ip:192.168.75.140

操作:

(1)查看ip

ifconfig

(2)更改hostname主机名

hostnamectl set-hostname 主机名

(3)添加域名映射

vim /etc/hosts

(4)查看是否存在.ssh

[root@master1 ~]# ls -a

如果有则输入:rm -rf /root/.ssh卸载 (5)生成ssh

ssh-keygen -t rsa

(6)给钥匙

master上执行

scp id_rsa.pub root@master1:/root/

scp id_rsa.pub root@slave1:/root/

等等等

(7)加保险

master和slave上都需要

cat id_rsa.pub>>.ssh/authorized_keys (8)测试

[root@master1 ~]# ssh slave1 Last login: Tue Jul 17 09:52:38 2018 from 192.168.75.1 [root@slave1 ~]#

2.配置java环境变量

(1)vim /etc/profile

末尾添加:

export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.151-1.b12.el7_4.x86_64/jre export CLASSPATH=.:$CLASSPATH:$JAVA_HOME/lib export PATH=$PATH:$JAVA_HOME/bin (2)执行source /etc/profile生效

(3)查看

$PATH

3.集群搭建

(1)配置文件路径:

配置集群的配置文件目录:cd $HADOOP_HOME/etc/hadoop/conf (2)增加slave节点

[root@master1 conf]# vim /etc/hadoop/conf/slaves

添加

master1 master2 slave1 slave2 (3)配置集群core-site.xml

[root@master2 ~]# vim /etc/hadoop/conf/core-site.xml 添加:

(4)配置集群hdfs-site.xml

[root@master1 conf]# vim hdfs-site.xml

添加:

(5)创建hdfs需要用的文件目录

[root@master1 ~]# mkdir /usr/hdp/tmp -p [root@master1 ~]# mkdir /hadoop/hdfs/{data,name} -p [root@master1 ~]# chown -R hdfs:hadoop /hadoop [root@master1 ~]# chown -R hdfs:hadoop /usr/hdp/tmp

(6)初始化hdfs文件系统

在master1上操作:

[root@master1 ~]# sudo -E -u hdfs hdfs namenode -format

(7)启动hdfs文件系统

启动master1节点上的服务:

[root@master1 ~]# systemctl start hadoop-hdfs-namenode [root@master1 ~]# systemctl start hadoop-hdfs-datanode

启动master2节点上的服务:

[root@master2 ~]# systemctl start hadoop-hdfs-datanode [root@master2 ~]# systemctl start hadoop-hdfs-secondarynamenode

启动slave1、slave2节点上的服务:

[root@slave1 ~]# systemctl start hadoop-hdfs-datanode [root@slave2 ~]# systemctl start hadoop-hdfs-datanode

(8)使用jps命令查看

4.网址查看

192.168.75.137:50070

二、大数据开发环境

1.准备程序运行目录

[root@master1 ~]# su - hdfs

-bash-4.2$ hadoop fs -mkdir /tmp

-bash-4.2$ hadoop fs -chmod -R 1777 /tmp

-bash-4.2$ hadoop fs -mkdir -p /var/log/hadoop-yarn

-bash-4.2$ hadoop fs -chown yarn:mapred /var/log/hadoop-yarn

-bash-4.2$ hadoop fs -mkdir /user

-bash-4.2$ hadoop fs -mkdir /user/hadoop

-bash-4.2$ hadoop fs -mkdir /user/history

-bash-4.2$ hadoop fs -chmod 1777 /user/history

-bash-4.2$ hadoop fs -chown mapred:hadoop /user/history

2.配置yarn-site.xml

[root@master1 conf]# vim yarn-site.xml 添加:

3.配置mapred-site.xml

[root@master1 conf]# vim mapred-site.xml

添加:

4.配置yarn的本地目录

[root@master1 ~]# touch /etc/hadoop/conf/yarn-env.sh [root@master1 ~]# mkdir -p /hadoop/yarn/local [root@master1 ~]# chown yarn:yarn -R /hadoop/yarn/local

5.启动服务

在master2上开启resourcemanager:

[root@master2 ~]# systemctl start hadoop-yarn-resourcemanager

访问web后台master2:8088

在slave1、slave2上开启historyserver

[root@slave1 ~]# systemctl start hadoop-mapreduce-historyserver

[root@slave2 ~]# systemctl start hadoop-mapreduce-historyserver

在所有启动datanode的节点上开nodemanager

[root@slave2 ~]# systemctl start hadoop-yarn-nodemanager

6.验证

master2:8088

slave1:19888

版权声明:本文内容由网络用户投稿,版权归原作者所有,本站不拥有其著作权,亦不承担相应法律责任。如果您发现本站中有涉嫌抄袭或描述失实的内容,请联系我们jiasou666@gmail.com 处理,核实后本网站将在24小时内删除侵权内容。

上一篇:Hadoop Word Count程序
下一篇:「易路」完成近10亿元融资,HR赛道独角兽即将诞生
相关文章

 发表评论

暂时没有评论,来抢沙发吧~