三节点hadoop集群搭建教程

一、安装VMware虚拟机

二、创建第一个Linux虚拟机节点，本机使用的centOS7.6 64位版本

三、创建2个克隆节点

四、 3节点都关闭防火墙

命令:systemctl stop firewalld

关闭后查看防火墙状态确认是否关闭成功：systemctl status firewalld

五、关闭 selinux

vi etc/selinux/config

SELINUX=disabled

六、配置网络设置

vi /etc/sysconfig/network-scripts/ifcfg-ens33

扫描二维码关注公众号，回复： 7424372 查看本文章

BOOTPROTO=static

ONBOOT=yes

IPADDR=192.168.XX.XX

NETMASK=255.255.255.0

GATEWAY=192.168.XX.2

DNS1=8.8.8.8

配置完毕，需重启网络服务

service network restart

七、三台虚拟机更改主机名

hostnamectl set-hostname node01

hostnamectl set-hostname node02

hostnamectl set-hostname node03

并更改hosts文件配置主机名与ip地址的映射关系

vi etc/hosts

重启虚拟机后生效

八、三台机器配置时间同步，这里选择aliyun的ntp服务器

yum -y install ntpdate

crontab -e

*/1 * * * * /usr/sbin/ntpdate time1.aliyun.com

九、添加hadoop专用用户，并赋予sudo权限

useradd hadoop

passwd XXXXXX

visudo

配置文件中添加

Hadoop ALL=(ALL) ALL

九、为hadoop应用创建专用目录

mkdir -p /hadoop/soft

mkdir -p /hadoop/install

将文件夹所有者更改为hadoop用户

chown -R hadoop:hadoop /hadoop

十、三台机器安装jdk

切换到hadoop用户，解压并安装jdk,此处使用1.8版本

cd /hadoop/soft/

配置hadoop用户的环境变量

cd /home/hadoop

vi .bash_profile

配置完成后,重新加载配置文件:source .bash_profile

验证配置：java -version

十一、配置hadoop用户免密登录

三台机器执行命令

ssh-keygen -t rsa

生成公钥与私钥

将公钥拷贝到节点1

ssh-copy-id node01

再从node01将公钥文件复制到其他两个节点

cd /home/hadoop/.ssh/

scp authorized_keys node02:$PWD

scp authorized_keys node03:$PWD

验证配置

三台虚拟机分别执行ssh命令连接到其他到其他机器，如连接失败，可删除hadoop用户目录下的.ssh文件夹重复该步骤。

十二、

解压安装hadoop，本机使用的是CDH发行版的hadoop

配置环境变量，此处仍然配置在hadoop用户下

配置完成后，重新加载环境变量：source .bash_profile

验证配置：java -version

hadoop version

如能正常输出版本信息，即验证配置成功

十三、

以下为hadoop的配置文件配置，建议使用远程连接工具登录虚拟机进行编辑

1、配置hadoop-env.sh文件

cd /hadoop/install/hadoop-2.6.0-cdh5.14.2/etc/hadoop

vi hadoop-env.sh

此文件内只需导入jdk的安装目录

export JAVA_HOME=/hadoop/install/jdk1.8.0_141

2、配置core-site.xml

cd /hadoop/install/hadoop-2.6.0-cdh5.14.2/etc/hadoop

vi core-site.xml

<name>fs.defaultFS</name>

</property>

<name>hadoop.tmp.dir</name>

<value>/hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/tempDatas</value>

</property>

<name>io.file.buffer.size</name>

</property>

<name>fs.trash.interval</name>

<description>检查点被删除后的分钟数。如果为零，垃圾桶功能将被禁用。

该选项可以在服务器和客户端上配置。如果垃圾箱被禁用服务器端，则检查客户端配置。

如果在服务器端启用垃圾箱，则会使用服务器上配置的值，并忽略客户端配置值。</description>

</property>

<name>fs.trash.checkpoint.interval</name>

<description>垃圾检查点之间的分钟数。应该小于或等于fs.trash.interval。

如果为零，则将该值设置为fs.trash.interval的值。每次检查指针运行时，

它都会从当前创建一个新的检查点，并删除比fs.trash.interval更早创建的检查点。</description>

</property>

</configuration>

3、配置hdfs-site.xml

<!-- 集群动态上下线

<name>dfs.hosts</name>

<value>/hadoop/install/hadoop-2.6.0-cdh5.14.2/etc/hadoop/accept_host</value>

</property>

<name>dfs.hosts.exclude</name>

<value>/hadoop/install/hadoop-2.6.0-cdh5.14.2/etc/hadoop/deny_host</value>

</property>

-->

<name>dfs.namenode.secondary.http-address</name>

</property>

<name>dfs.namenode.http-address</name>

</property>

<name>dfs.namenode.name.dir</name>

<value>file:///hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/namenodeDatas</value>

</property>

<name>dfs.datanode.data.dir</name>

<value>file:///hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/datanodeDatas</value>

</property>

<name>dfs.namenode.edits.dir</name>

<value>file:///hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/nn/edits</value>

</property>

<name>dfs.namenode.checkpoint.dir</name>

<value>file:///hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/snn/name</value>

</property>

<name>dfs.namenode.checkpoint.edits.dir</name>

<value>file:///hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/nn/snn/edits</value>

</property>

<name>dfs.replication</name>

</property>

<name>dfs.permissions</name>

<value>false</value>

</property>

<name>dfs.blocksize</name>

</property>

</configuration>

4、配置mapred-site.xml

vi mapred-site.xml

<name>mapreduce.framework.name</name>

</property>

<name>mapreduce.job.ubertask.enable</name>

</property>

<name>mapreduce.jobhistory.address</name>

</property>

<name>mapreduce.jobhistory.webapp.address</name>

</property>

</configuration>

5、配置yarn-site.xml

<name>yarn.resourcemanager.hostname</name>

</property>

<name>yarn.nodemanager.aux-services</name>

<value>mapreduce_shuffle</value>

</property>

<name>yarn.log-aggregation-enable</name>

</property>

<name>yarn.log.server.url</name>

<value>http://node01:19888/jobhistory/logs</value>

</property>

<name>yarn.log-aggregation.retain-seconds</name>

</property>

<name>yarn.nodemanager.log.retain-seconds</name>

</property>

<name>yarn.nodemanager.log-aggregation.compression-type</name>

</property>

<name>yarn.nodemanager.local-dirs</name>

<value>/hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/yarn/local</value>

</property>

<name>yarn.resourcemanager.max-completed-applications</name>

</property>

</configuration>

6、创建文件存放目录

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/tempDatas

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/namenodeDatas

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/datanodeDatas

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/nn/edits

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/snn/name

[root@node01 ~]# mkdir -p /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/nn/snn/edits

十四、格式化hadoop（此步骤是在hadoop集群启动前，在namenode(主节点)执行）

hdfs namenode -format

以下为部分日志，供参考

19/08/23 04:32:34 INFO namenode.NameNode: STARTUP_MSG:

/************************************************************

STARTUP_MSG: Starting NameNode

STARTUP_MSG: user = hadoop

STARTUP_MSG: host = node01.kaikeba.com/192.168.52.100

STARTUP_MSG: args = [-format]

STARTUP_MSG: version = 2.6.0-cdh5.14.2

STARTUP_MSG: classpath = /hadoop/install/hadoop-2.6.0-19/08/23 04:32:35 INFO common.Storage: Storage directory /hadoop/install/hadoop-2.6.0-

#显示格式化成功。。。

cdh5.14.2/hadoopDatas/namenodeDatas has been successfully formatted.

19/08/23 04:32:35 INFO common.Storage: Storage directory /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/dfs/nn/edits has been successfully formatted.

19/08/23 04:32:35 INFO namenode.FSImageFormatProtobuf: Saving image file /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/namenodeDatas/current/fsimage.ckpt_0000000000000000000 using no compression

19/08/23 04:32:35 INFO namenode.FSImageFormatProtobuf: Image file /hadoop/install/hadoop-2.6.0-cdh5.14.2/hadoopDatas/namenodeDatas/current/fsimage.ckpt_0000000000000000000 of size 323 bytes saved in 0 seconds.

19/08/23 04:32:35 INFO namenode.NNStorageRetentionManager: Going to retain 1 images with txid >= 0

19/08/23 04:32:35 INFO util.ExitUtil: Exiting with status 0

19/08/23 04:32:35 INFO namenode.NameNode: SHUTDOWN_MSG:

#此处省略部分日志

/************************************************************

SHUTDOWN_MSG: Shutting down NameNode at node01.kaikeba.com/192.168.52.100

************************************************************/

十五、启动集群

start-all.sh

This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh

19/08/23 05:18:09 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

Starting namenodes on [node01]

node01: starting namenode, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/hadoop-hadoop-namenode-node01.kaikeba.com.out

node01: starting datanode, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/hadoop-hadoop-datanode-node01.kaikeba.com.out

node03: starting datanode, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/hadoop-hadoop-datanode-node03.kaikeba.com.out

node02: starting datanode, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/hadoop-hadoop-datanode-node02.kaikeba.com.out

Starting secondary namenodes [node01]

node01: starting secondarynamenode, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/hadoop-hadoop-secondarynamenode-node01.kaikeba.com.out

19/08/23 05:18:24 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

starting yarn daemons

starting resourcemanager, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/yarn-hadoop-resourcemanager-node01.kaikeba.com.out

node03: starting nodemanager, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/yarn-hadoop-nodemanager-node03.kaikeba.com.out

node02: starting nodemanager, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/yarn-hadoop-nodemanager-node02.kaikeba.com.out

node01: starting nodemanager, logging to /hadoop/install/hadoop-2.6.0-cdh5.14.2/logs/yarn-hadoop-nodemanager-node01.kaikeba.com.out

[hadoop@node01 ~]$

在浏览器地址栏输入http://192.168.52.100:50070/dfshealth.html#tab-overview查看namenode的web界面

十六、运行mapreduce程序

1、使用 hdfs dfs -ls / 命令浏览hdfs文件系统

hdfs dfs -ls

由于集群刚搭建，此时没有目录显示

2、创建测试目录

hdfs dfs -mkdir /test

hdfs dfs -ls /test

19/08/23 05:22:25 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

Found 2 items

drwxr-xr-x - hadoop supergroup 0 2019-08-23 05:21 /test/

3、使用touch命令在linux本地创建一个words文件

touch words

vi words

sadfasdfasdfas2rzxcvzr3r23

sadfasdfhszcxvhh8

4、将创建的本地words文件上传到hdfs的test目录上

hdfs dfs -put words /test

查看文件是否上传成功

hdfs dfs -ls -r /test

执行命令，统计/test/words文件中的单词个数，并输出到test/output文件中，output文件不能是已经存在的，否则会报错

hadoop jar /hadoop/install/hadoop-2.6.0-cdh5.14.2/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.6.0-cdh5.14.2.jar wordcount /test/words /test/output

十七、关闭集群

stop-all.sh

简易hadoop三节点集群搭建总结

三节点hadoop集群搭建教程

猜你喜欢