ZooKeeper集群部署(容器)

一、ZooKeeper基本概念

ZooKeeper是一个分布式且开源的分布式应用程序的协调服务(管理分布式服务)

ZooKeeper提供的主要功能包括：

• 配置管理
• 分布式锁
• 集群管理

二、ZooKeeper集群部署

1、前置环境准备

1.1 关闭防火墙等限制

systemctl disable firewalld --now
setenforce 0
sed  -i -r 's/SELINUX=[ep].*/SELINUX=disabled/g' /etc/selinux/config

1.2 安装docker环境

（1）安装docker

curl -o /etc/yum.repos.d/CentOS-Base.repo https://mirrors.aliyun.com/repo/Centos-7.repo
wget -O /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-7.repo
yum makecache# yum-utils软件用于提供yum-config-manager程序
yum install -y yum-utils# 使用yum-config-manager创建docker阿里存储库
yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repoyum install docker-ce-20.10.6 docker-ce-cli-20.10.6 -y

（2）配置docker国内加速器

mkdir /etc/docker
cat <<EOF > /etc/docker/daemon.json
{"registry-mirrors": [
"https://vm1wbfhf.mirror.aliyuncs.com",
"http://f1361db2.m.daocloud.io",
"https://hub-mirror.c.163.com",
"https://docker.mirrors.ustc.edu.cn",
"https://mirror.baidubce.com",
"https://ustc-edu-cn.mirror.aliyuncs.com",
"https://registry.cn-hangzhou.aliyuncs.com",
"https://ccr.ccs.tencentyun.com",
"https://hub.daocloud.io",
"https://docker.shootchat.top",
"https://do.nark.eu.org",
"https://dockerproxy.com",
"https://docker.m.daocloud.io",
"https://dockerhub.timeweb.cloud",
"https://docker.shootchat.top",
"https://do.nark.eu.org"],
"exec-opts": ["native.cgroupdriver=systemd"]
}
EOF

（3）启动docker并设置开机自启

systemctl enable docker --now
systemctl status docker

1.3 安装docker-compose环境

DOCKER_COMPOSE_VERSION="v2.27.0"
sudo curl -L "https://github.com/docker/compose/releases/download/${DOCKER_COMPOSE_VERSION}/docker-compose-$(uname -s)-$(uname -m)" \-o /usr/local/bin/docker-composechmod +x /usr/local/bin/docker-compose

2、ZooKeeper伪集群部署(可选)

ZooKeeper伪集群指的是，将集群部署到同一台服务器中

2.1 创建目录，添加docker-compose文件

mkdir /data/software/zookeeper-cluster -p
cd /data/software/zookeeper-cluster
vim docker-compose.ymlversion: '3.4'services:zk1:image: zookeeper:3.4.14restart: alwayshostname: zk1container_name: zk1ports:- 2181:2181- 2888:2888- 3888:3888volumes:- "./data/zk1-data:/data"- "./datalog/zk1-datalog:/datalog"- "./logs/zk1-logs:/logs"environment:ZOO_MY_ID: 1ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/Shanghaihealthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3networks:- zookeeper-netzk2:image: zookeeper:3.4.14restart: alwayshostname: zk2container_name: zk2ports:- 22181:2181- 22888:2888- 23888:3888volumes:- "./data/zk2-data:/data"- "./datalog/zk2-datalog:/datalog"- "./logs/zk2-logs:/logs"environment:ZOO_MY_ID: 2ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/Shanghaihealthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3networks:- zookeeper-netzk3:image: zookeeper:3.4.14restart: alwayshostname: zk3container_name: zk3ports:- 32181:2181- 32888:2888- 33888:3888volumes:- "./data/zk3-data:/data"- "./datalog/zk3-datalog:/datalog"- "./logs/zk3-logs:/logs"environment:ZOO_MY_ID: 3ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/Shanghaihealthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3networks:- zookeeper-net
networks:zookeeper-net:driver: bridge

2.2 启动zk集群：

cd /data/software/zookeeper-cluster
docker-compose up -d
docker-compose logs -f 

3、ZooKeeper集群部署(可选)

3.1 集群环境说明

序号	IP地址	主机名称
1	16.32.15.116	zk1
2	16.32.15.200	zk2
3	16.32.15.201	zk3

3.2 zk1主机相关操作

（1）创建目录，添加docker-compose文件

mkdir /data/software/zookeeper -p
cd /data/software/zookeeper
vim docker-compose.ymlversion: '3.4'services:zk1:                                  # 三个节点对应不同名称 [ zk2 | zk3 ]image: zookeeper:3.4.14restart: alwayshostname: zk1                       # 三个节点对应不同名称 [ zk2 | zk3 ]container_name: zk1                 # 三个节点对应不同名称 [ zk2 | zk3 ]network_mode: "host"volumes:- "./data:/data"- "./datalog:/datalog"- "./logs:/logs"environment:ZOO_MY_ID: 1                      # 三个节点对应不同ID [ 2 | 3 ]ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/ShanghaiJVMFLAGS: "-Xmx1024m -Xms521m"extra_hosts:                        # 设置容器hosts为宿主机IP地址- "zk1:16.32.15.116"- "zk2:16.32.15.200"- "zk3:16.32.15.201"healthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3mem_limit: 2g                         # 内存硬限制mem_reservation: 1500m                # 内存软限制logging:                              # 日志大小限制driver: json-fileoptions:max-size: "50m"max-file: "10"

（2）启动zk1容器：

cd /data/software/zookeeper
docker-compose up -d
docker-compose logs -f 

3.3 zk2主机相关操作

（1）创建目录，添加docker-compose文件

mkdir /data/software/zookeeper -p
cd /data/software/zookeeper
vim docker-compose.ymlversion: '3.4'services:zk2:image: zookeeper:3.4.14restart: alwayshostname: zk2container_name: zk2network_mode: "host"volumes:- "./data:/data"- "./datalog:/datalog"- "./logs:/logs"environment:ZOO_MY_ID: 2ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/ShanghaiJVMFLAGS: "-Xmx1024m -Xms521m"extra_hosts:                        # 设置容器hosts为宿主机IP地址- "zk1:16.32.15.116"- "zk2:16.32.15.200"- "zk3:16.32.15.201"healthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3mem_limit: 2g                         # 内存硬限制mem_reservation: 1500m                # 内存软限制logging:                              # 日志大小限制driver: json-fileoptions:max-size: "50m"max-file: "10"

（2）启动zk2容器：

cd /data/software/zookeeper
docker-compose up -d
docker-compose logs -f 

3.4 zk3主机相关操作

（1）创建目录，添加docker-compose文件

mkdir /data/software/zookeeper -p
cd /data/software/zookeeper
vim docker-compose.ymlversion: '3.4'services:zk3:image: zookeeper:3.4.14restart: alwayshostname: zk3container_name: zk3network_mode: "host"volumes:- "./data:/data"- "./datalog:/datalog"- "./logs:/logs"environment:ZOO_MY_ID: 3ZOO_SERVERS: server.1=zk1:2888:3888 server.2=zk2:2888:3888 server.3=zk3:2888:3888TZ: Asia/ShanghaiJVMFLAGS: "-Xmx1024m -Xms521m"extra_hosts:                        # 设置容器hosts为宿主机IP地址- "zk1:16.32.15.116"- "zk2:16.32.15.200"- "zk3:16.32.15.201"healthcheck:test: ["CMD", "sh", "-c", "nc -z 127.0.0.1 2181"]interval: 10stimeout: 5sretries: 3mem_limit: 2g                         # 内存硬限制mem_reservation: 1500m                # 内存软限制logging:                              # 日志大小限制driver: json-fileoptions:max-size: "50m"max-file: "10"       

（2）启动zk3容器：

cd /data/software/zookeeper
docker-compose up -d
docker-compose logs -f 

三、ZooKeeper集群验证

1、查看集群角色

yum -y install nc
zkList=(16.32.15.116 16.32.15.200 16.32.15.201)
for zkhost in ${zkList[@]};do zkMode=$(echo stat | nc ${zkhost} 2181 | grep Mode);echo [${zkhost}] ${zkMode};done

2、数据同步测试

# zk1主机创建数据
docker exec -it zk1 bin/zkCli.sh
create /test "QIN TEST 666...."# zk2主机查看数据
docker exec -it zk2 bin/zkCli.sh
get /test

3、选举leader测试

1. 查看leader主机IP地址

zkList=(16.32.15.116 16.32.15.200 16.32.15.201)
for zkhost in ${zkList[@]};do zkMode=$(echo stat | nc ${zkhost} 2181 | grep Mode);echo [${zkhost}] ${zkMode};done

2. 将leader主机停止掉模仿服务器宕机

# leader主机操作(zk2)
cd /data/software/zookeeper
docker-compose down

3. 查看选举出新的leader主机

zkList=(16.32.15.116 16.32.15.200 16.32.15.201)
for zkhost in ${zkList[@]};do zkMode=$(echo stat | nc ${zkhost} 2181 | grep Mode);echo [${zkhost}] ${zkMode};done