Docker 搭建 ZooKeeper 集群

发表于 更新于 阅读次数： 评论数： 本文字数： 1.3k 阅读时长 ≈ 1 分钟

前言

本文将使用 Docker 与 Docker-Compose 在单机上快速安装 ZooKeeper 集群（伪集群）。

镜像地址

集群规划

节点名称节点 ID 监听端口映射端口版本说明
zookeeper011218121813.8.4
zookeeper022218121823.8.4
zookeeper033218121833.8.4

准备工作

  • 在宿主机中创建 ZooKeeper 节点一用于持久化数据的目录
1
2
sudo mkdir -p /usr/local/zookeeper/zookeeper01/data
sudo mkdir -p /usr/local/zookeeper/zookeeper01/datalog
  • 在宿主机中创建 ZooKeeper 节点二用于持久化数据的目录
1
2
sudo mkdir -p /usr/local/zookeeper/zookeeper02/data
sudo mkdir -p /usr/local/zookeeper/zookeeper02/datalog
  • 在宿主机中创建 ZooKeeper 节点三用于持久化数据的目录
1
2
sudo mkdir -p /usr/local/zookeeper/zookeeper03/data
sudo mkdir -p /usr/local/zookeeper/zookeeper03/datalog

数据卷使用说明

  • ZooKeeper 官方镜像在 /data 和 /datalog 处配置了数据卷，分别用于保存 Zookeeper 内存中的数据库快照和数据库更新的事务日志。
  • 请注意存放事务日志数据的磁盘位置，因为专用的事务日志磁盘是保持良好性能的关键，将日志存放在繁忙的磁盘上会对性能产生不利影响。
  • 若希望指定 ZooKeeper 的配置文件，可以将宿主机中的 ZooKeeper 配置文件 zoo.cfg 挂载到 ZooKeeper 容器内的 /conf/zoo.cfg 位置。

创建容器

  • 创建 docker-compose.yml 配置文件，并写入以下内容：
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
version: '3.5'

services:
  zookeeper01:
    image: zookeeper:3.8.4
    container_name: zookeeper01
    restart: always
    hostname: zookeeper01
    ports:
      - 2181:2181
    environment:
      TZ: Asia/Shanghai
      ZOO_MY_ID: 1
      ZOO_PORT: 2181
      ZOO_4LW_COMMANDS_WHITELIST: "ruok"
      ZOO_SERVERS: server.1=zookeeper01:2888:3888;2181 server.2=zookeeper02:2888:3888;2181 server.3=zookeeper03:2888:3888;2181
    healthcheck:
      test: ["CMD", "sh", "-c", "echo ruok | nc localhost 2181 | grep imok"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 20s
    volumes:
      - /usr/local/zookeeper/zookeeper01/data:/data
      - /usr/local/zookeeper/zookeeper01/datalog:/datalog
    networks:
      - distributed-network

  zookeeper02:
    image: zookeeper:3.8.4
    container_name: zookeeper02
    restart: always
    hostname: zookeeper02
    ports:
      - 2182:2181
    environment:
      TZ: Asia/Shanghai
      ZOO_MY_ID: 2
      ZOO_PORT: 2181
      ZOO_4LW_COMMANDS_WHITELIST: "ruok"
      ZOO_SERVERS: server.1=zookeeper01:2888:3888;2181 server.2=zookeeper02:2888:3888;2181 server.3=zookeeper03:2888:3888;2181
    healthcheck:
      test: ["CMD", "sh", "-c", "echo ruok | nc localhost 2181 | grep imok"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 20s
    volumes:
      - /usr/local/zookeeper/zookeeper02/data:/data
      - /usr/local/zookeeper/zookeeper02/datalog:/datalog
    networks:
      - distributed-network

  zookeeper03:
    image: zookeeper:3.8.4
    container_name: zookeeper03
    restart: always
    hostname: zookeeper03
    ports:
      - 2183:2181
    environment:
      TZ: Asia/Shanghai
      ZOO_MY_ID: 3
      ZOO_PORT: 2181
      ZOO_4LW_COMMANDS_WHITELIST: "ruok"
      ZOO_SERVERS: server.1=zookeeper01:2888:3888;2181 server.2=zookeeper02:2888:3888;2181 server.3=zookeeper03:2888:3888;2181
    healthcheck:
      test: ["CMD", "sh", "-c", "echo ruok | nc localhost 2181 | grep imok"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 20s
    volumes:
      - /usr/local/zookeeper/zookeeper03/data:/data
      - /usr/local/zookeeper/zookeeper03/datalog:/datalog
    networks:
      - distributed-network

networks:
  distributed-network:
    driver: bridge

配置说明

  • ZOO_MY_ID：用于指定 ZooKeeper 节点的 ID，它在集群中必须是唯一，并且其值应在 1 到 255 之间。另外，ZOO_MY_ID: 1 对应 server.1=zookeeper01:2888:3888;2181 中的 server.1。
  • ZOO_4LW_COMMANDS_WHITELIST:"ruok" 是让 ZooKeeper 启用 ruok"四字命令"（4LW，Four Letter Words），用于简单的健康检查。在默认情况下，部分 ZooKeeper 配置可能禁用了这些命令。
  • server.1=zookeeper01:2888:3888;2181 中的 zookeeper01 是在 Docker Compose 中定义的 ZooKeeper 服务的名称，用作其他服务（如 Kafka）访问 ZooKeeper 的主机名，Docker Compose 会自动将这个名称解析为 ZooKeeper 容器的 IP 地址。
  • 创建并启动 ZooKeeper 容器
1
sudo docker-compose up -d

测试容器

查看容器状态

  • 查看所有 ZooKeeper 容器的运行状态
1
sudo docker ps -a
1
2
3
82d40df530ec   zookeeper:3.8.4   "/docker-entrypoint.…"   About a minute ago   Up About a minute (healthy)   2888/tcp, 3888/tcp, 0.0.0.0:2181->2181/tcp, :::2181->2181/tcp, 8080/tcp   zookeeper01
4ae126d4ea36   zookeeper:3.8.4   "/docker-entrypoint.…"   About a minute ago   Up About a minute (healthy)   2888/tcp, 3888/tcp, 8080/tcp, 0.0.0.0:2182->2181/tcp, :::2182->2181/tcp   zookeeper02
c70d199dd8ce   zookeeper:3.8.4   "/docker-entrypoint.…"   About a minute ago   Up About a minute (healthy)   2888/tcp, 3888/tcp, 8080/tcp, 0.0.0.0:2183->2181/tcp, :::2183->2181/tcp   zookeeper03
  • 若 ZooKeeper 容器启动失败，可以通过以下命令查看容器的启动日志来排查问题
1
sudo docker logs -f --tail 100 zookeeper01

查看集群状态

  • 查看 ZooKeeper 节点一的集群状态
1
sudo docker exec -it zookeeper01 zkServer.sh status
1
2
3
4
ZooKeeper JMX enabled by default
Using config: /conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: follower
  • 查看 ZooKeeper 节点二的集群状态
1
sudo docker exec -it zookeeper02 zkServer.sh status
1
2
3
4
ZooKeeper JMX enabled by default
Using config: /conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: follower
  • 查看 ZooKeeper 节点三的集群状态
1
sudo docker exec -it zookeeper03 zkServer.sh status
1
2
3
4
ZooKeeper JMX enabled by default
Using config: /conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: leader

数据同步测试

  • 使用客户端连接 ZooKeeper 节点一，并创建目录
1
2
# 客户端连接节点一
sudo docker exec -it zookeeper01 zkCli.sh -server localhost:2181
1
2
3
4
5
6
7
# 创建目录
[zk: localhost:2181 (CONNECTED) 0] create /test ""
Created /test

# 查看目录信息
[zk: localhost:2181 (CONNECTED) 1] ls /
[test, zookeeper]
  • 使用客户端连接其他任意节点（比如节点二），可以发现 test 目录会同步创建，即在任何一个集群节点进行操作，其他集群节点也会同步更新
1
2
# 客户端连接节点二
sudo docker exec -it zookeeper02 zkCli.sh -server localhost:2181
1
2
3
# 查看目录信息
[zk: localhost:2181 (CONNECTED) 0] ls /
[test, zookeeper]

集群重新选举测试

  • 关闭 Leader 节点（比如节点三）
1
sudo docker stop zookeeper03
  • 等待超时时间到了之后，重新查看其他节点的状态（比如节点一、节点二），会发现节点二被选举为新的 Leader。
1
sudo docker exec -it zookeeper01 zkServer.sh status
1
2
3
4
ZooKeeper JMX enabled by default
Using config: /conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: follower
1
sudo docker exec -it zookeeper02 zkServer.sh status
1
2
3
4
ZooKeeper JMX enabled by default
Using config: /conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: leader

参考资料