一．创建三台虚拟机
- node1 10.24.227.111
- node2 10.24.227.112
- node3 10.24.227.113
二．安装jdk环境
方法一：rpm安装：预先下载好 jdk-21_linux-x64_bin.rpm
将包拉进根目录，执行：rpm -ivh jdk-21_linux-x64_bin.rpm
方法二：yum install -y java-1.8.0-openjdk*
三．关闭防火墙
systemctl stop firewalld && systemctl disable firewalld
### 关闭selinux
vi /etc/selinux/config
模式改成permissive
setenforce 0
四．安装zookeeper
tar -zxvf apache-zookeeper-3.9.2-bin.tar.gz -C /usr/local/
cd /usr/local/
mv apache-zookeeper-3.9.2-bin zookeeper-3.9.2
cd /usr/local/zookeeper-3.9.2/
mkdir data
cd conf/
mv zoo_sample.cfg zoo.cfg
vi zoo.cfg
`\# zookeeper 数据存储地址`
`dataDir=/usr/local/zookeeper-3.9.2/data`
`\# zookeeper 集群地址`
`server.1=10.24.227.111:2888:3888`
`server.2=10.24.227.112:2888:3888`
`server.3=10.24.227.113:2888:3888`
`\# admin.server端口`
`admin.serverPort=8888`
scp -r zookeeper-3.9.2 node2:/usr/local/
scp -r zookeeper-3.9.2 node3:/usr/local/
node1:
echo 1 > data/myid
node2:
echo 2 > data/myid
node3:
echo 3 > data/myid
启动zookeeper：
sh bin/zkServer.sh start
sh bin/zkServer.sh status //注意，要启动三台才能看见它的运行状态，我们可以通过bin目录下面的zookeeper.out来查看问题原因，这上面说的是myid文件缺失，可能涉及到Leader（Master）选举的问题，需要先启动master，然后再启动follower。
安装Kafka
tar -zxvf kafka_2.12-3.6.0.tgz -C /usr/local/
cd /usr/local/kafka_2.12-3.6.0/
vi config/server.properties
broker.id=1 //不同节点id不同
host.name=10.24.227.111
listeners=PLAINTEXT://10.24.227.111:9092
advertised.listeners=PLAINTEXT://10.24.227.111:9092
zookeeper.connect=10.24.227.111:2181,10.24.227.112:2181,10.24.227.113:2181/kafka
scp -r kafka_2.12-3.6.0 node2:/usr/local/
scp -r kafka_2.12-3.6.0 node3:/usr/local/
启动Kafka
sh bin/kafka-server-start.sh -daemon config/server.properties
一共三个node节点，1leader，2follower集群成功，三个卡夫卡成功运行
版权属于：JiuXia2025
本文链接：https://blog.inekoxia.com/index.php/archives/684.html
转载时须注明出处及本声明
