docker如何部署etcd集群

更新时间：2023年03月16日 15:28:32 作者：AlexClownfish

这篇文章主要介绍了docker如何部署etcd集群问题，具有很好的参考价值，希望对大家有所帮助。如有错误或未考虑完全的地方，望不吝赐教

–name：设置成员节点的别名，建议为每个成员节点配置可识别的命名
–advertise-client-urls：广播到集群中本成员的监听客户端请求的地址
–initial-advertise-peer-urls：广播到集群中本成员的Peer监听通信地址
–listen-client-urls：客户端请求的监听地址列表
–listen-peer-urls：Peer消息的监听服务地址列表
–initial-cluster-token：启动集群的时候指定集群口令，只有相同token的几点才能加入到同一集群
–initial-cluster：所有集群节点的地址列表
–initial-cluster-state：初始化集群状态，默认为new，也可以指定为exi-string表示要加入到一个已有集群

创建etcd数据目录

mkdir -p ./etcd-node{1,2,3}

创建docker网络

docker network create --driver bridge --subnet 172.62.0.0/16 --gateway 172.62.0.1 etcd-cluster

etcd-cluster-compose.yml

version: '3'

networks:
  etcd-cluster:
    external: true

services:
  etcd-node1:
    image: quay.io/coreos/etcd:v3.3.1
    container_name: etcd-node1
    ports:
      - "12379:2379"
      - "12380:2380"
    restart: always
    volumes:
      - ./etcd-node1:/data/app/etcd
    command: etcd --name etcd-node1 --data-dir /data/app/etcd/ --advertise-client-urls http://172.62.0.10:2379 --initial-advertise-peer-urls http://172.62.0.10:2380 --listen-client-urls http://0.0.0.0:2379 --listen-peer-urls http://0.0.0.0:2380 --initial-cluster-token etcd-cluster --initial-cluster "etcd-node1=http://172.62.0.10:2380,etcd-node2=http://172.62.0.11:2380,etcd-node3=http://172.62.0.12:2380" --initial-cluster-state new
    networks:
      etcd-cluster:
        ipv4_address: 172.62.0.10

  etcd-node2:
    image: quay.io/coreos/etcd:v3.3.1
    container_name: etcd-node2
    ports:
      - "22379:2379"
      - "22380:2380"
    restart: always
    volumes:
      - ./etcd-node2:/data/app/etcd
    command: etcd --name etcd-node2 --data-dir /data/app/etcd/ --advertise-client-urls http://172.62.0.11:2379 --initial-advertise-peer-urls http://172.62.0.11:2380 --listen-client-urls http://0.0.0.0:2379 --listen-peer-urls http://0.0.0.0:2380 --initial-cluster-token etcd-cluster --initial-cluster "etcd-node1=http://172.62.0.10:2380,etcd-node2=http://172.62.0.11:2380,etcd-node3=http://172.62.0.12:2380" --initial-cluster-state new
    networks:
      etcd-cluster:
        ipv4_address: 172.62.0.11

  etcd-node3:
    image: quay.io/coreos/etcd:v3.3.1
    container_name: etcd-node3
    ports:
      - "32379:2379"
      - "32380:2380"
    restart: always
    volumes:
      - ./etcd-node3:/data/app/etcd
    command: etcd --name etcd-node3 --data-dir /data/app/etcd/ --advertise-client-urls http://172.62.0.12:2379 --initial-advertise-peer-urls http://172.62.0.12:2380 --listen-client-urls http://0.0.0.0:2379 --listen-peer-urls http://0.0.0.0:2380 --initial-cluster-token etcd-cluster --initial-cluster "etcd-node1=http://172.62.0.10:2380,etcd-node2=http://172.62.0.11:2380,etcd-node3=http://172.62.0.12:2380" --initial-cluster-state new
    networks:
      etcd-cluster:
        ipv4_address: 172.62.0.12

启动并验证集群

启动

[root@k8s-node1 etcd-cluster]# docker-compose -f etcd-cluster-compose.yml up -d
Pulling etcd-node1 (quay.io/coreos/etcd:v3.3.1)...
v3.3.1: Pulling from coreos/etcd
ff3a5c916c92: Pull complete
dec5fcc85a18: Pull complete
3944f16f0112: Pull complete
0b6d29b049fe: Pull complete
d8c39ae91d38: Pull complete
42fcea4864ba: Pull complete
Digest: sha256:454e69370d87554dcb4272833b8f07ce1b5d457caa153bda4070b76d89a1cc97
Status: Downloaded newer image for quay.io/coreos/etcd:v3.3.1
Creating etcd-node1 ... done
Creating etcd-node2 ... done
Creating etcd-node3 ... done

[root@k8s-node1 etcd-cluster]# docker-compose -f etcd-cluster-compose.yml ps -a
   Name                 Command               State                        Ports
------------------------------------------------------------------------------------------------------
etcd-node1   etcd --name etcd-node1 --d ...   Up      0.0.0.0:12379->2379/tcp, 0.0.0.0:12380->2380/tcp
etcd-node2   etcd --name etcd-node2 --d ...   Up      0.0.0.0:22379->2379/tcp, 0.0.0.0:22380->2380/tcp
etcd-node3   etcd --name etcd-node3 --d ...   Up      0.0.0.0:32379->2379/tcp, 0.0.0.0:32380->2380/tcp

验证集群

通过etcdctl member list命令可以查询出所有集群节点的列表并且结果一致即为成功

[root@k8s-node1 etcd-cluster]# docker exec -it  etcd-node1 /bin/sh
/ # etcdctl member list
8cef47d732d4acff: name=etcd-node1 peerURLs=http://172.62.0.10:2380 clientURLs=http://172.62.0.10:2379 isLeader=false
c93af917b643516f: name=etcd-node3 peerURLs=http://172.62.0.12:2380 clientURLs=http://172.62.0.12:2379 isLeader=true
cdee7114ad135065: name=etcd-node2 peerURLs=http://172.62.0.11:2380 clientURLs=http://172.62.0.11:2379 isLeader=false
/ # exit
[root@k8s-node1 etcd-cluster]# docker exec -it  etcd-node2 /bin/sh
/ # etcdctl member list
8cef47d732d4acff: name=etcd-node1 peerURLs=http://172.62.0.10:2380 clientURLs=http://172.62.0.10:2379 isLeader=false
c93af917b643516f: name=etcd-node3 peerURLs=http://172.62.0.12:2380 clientURLs=http://172.62.0.12:2379 isLeader=true
cdee7114ad135065: name=etcd-node2 peerURLs=http://172.62.0.11:2380 clientURLs=http://172.62.0.11:2379 isLeader=false
[root@k8s-node1 etcd-cluster]# docker exec -it  etcd-node3 /bin/sh
/ # etcdctl member list
8cef47d732d4acff: name=etcd-node1 peerURLs=http://172.62.0.10:2380 clientURLs=http://172.62.0.10:2379 isLeader=false
c93af917b643516f: name=etcd-node3 peerURLs=http://172.62.0.12:2380 clientURLs=http://172.62.0.12:2379 isLeader=true
cdee7114ad135065: name=etcd-node2 peerURLs=http://172.62.0.11:2380 clientURLs=http://172.62.0.11:2379 isLeader=false

k/v操作

CURL

新增

[root@k8s-node1 etcd-cluster]# docker-compose -f etcd-cluster-compose.yml ps -a
   Name                 Command               State                        Ports
------------------------------------------------------------------------------------------------------
etcd-node1   etcd --name etcd-node1 --d ...   Up      0.0.0.0:12379->2379/tcp, 0.0.0.0:12380->2380/tcp
etcd-node2   etcd --name etcd-node2 --d ...   Up      0.0.0.0:22379->2379/tcp, 0.0.0.0:22380->2380/tcp
etcd-node3   etcd --name etcd-node3 --d ...   Up      0.0.0.0:32379->2379/tcp, 0.0.0.0:32380->2380/tcp
[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:12379/version
{"etcdserver":"3.3.1","etcdcluster":"3.3.0"}[root@k8s-node1 etcd-cluster]#
[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:12379/v2/keys/Alexclownfish -X PUT -d value=https://blog.alexcld.com
{"action":"set","node":{"key":"/Alexclownfish","value":"https://blog.alexcld.com","modifiedIndex":19,"createdIndex":19}}

查询

[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:12379/v2/keys/Alexclownfish -X GET
{"action":"get","node":{"key":"/Alexclownfish","value":"https://blog.alexcld.com","modifiedIndex":19,"createdIndex":19}}
[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:22379/v2/keys/Alexclownfish -X GET
{"action":"get","node":{"key":"/Alexclownfish","value":"https://blog.alexcld.com","modifiedIndex":19,"createdIndex":19}}
[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:32379/v2/keys/Alexclownfish -X GET
{"action":"get","node":{"key":"/Alexclownfish","value":"https://blog.alexcld.com","modifiedIndex":19,"createdIndex":19}}

修改

同新建一样

删除

[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:12379/v2/keys/Alexclownfish -X DELETE
{"action":"delete","node":{"key":"/Alexclownfish","modifiedIndex":20,"createdIndex":19},"prevNode":{"key":"/Alexclownfish","value":"https://blog.alexcld.com","modifiedIndex":19,"createdIndex":19}}
[root@k8s-node1 etcd-cluster]# curl -L http://127.0.0.1:12379/v2/keys/Alexclownfish -X GET
{"errorCode":100,"message":"Key not found","cause":"/Alexclownfish","index":20}

etcdctl

新增

/ # etcdctl set clownfish 1234567
1234567
/ # etcdctl get clownfish
1234567

查询

/ # etcdctl get clownfish
1234567

修改

/ # etcdctl get clownfish
1234567
/ # etcdctl set clownfish 987654321ddd
987654321ddd
/ # etcdctl get clownfish
987654321ddd

删除

/ # etcdctl rm clownfish
PrevNode.Value: 987654321ddd
/ # etcdctl get clownfish
Error:  100: Key not found (/clownfish) [23]

总结

以上为个人经验，希望能给大家一个参考，也希望大家多多支持脚本之家。

您可能感兴趣的文章:

如何给Docker配置官方国内加速镜像
在国内访问 Docker 官方的镜像，一直以来速度都慢如蜗牛。为了快速访问 Docker 官方镜像都会配置三方加速器
2017-06-06
使用Docker快速安装grafana的基本步骤
Grafana 是一个强大的开源监控和数据分析平台,它能够与各种数据源集成,并提供灵活的可视化和仪表盘功能,下面给大家分享Docker 中安装 Grafana 的基本步骤,感兴趣的朋友一起看看吧
2024-01-01
docker覆盖镜像默认命令之docker entrypoint详解
entrypoint命令就是覆盖ENTRYPOINT命令的,本文给大家介绍了docker覆盖镜像默认命令之docker entrypoint的相关知识,需要的朋友可以参考下
2023-10-10
迁移Docker中MySQL容器的全过程
在这篇文章中,我们将介绍如何将在 Docker 中运行的 MySQL 容器从一台机器（我们称之为A机器）迁移到另一台机器（我们称之为B机器）,这个过程主要包括五个步骤：提交容器为镜像,保存镜像为 tar 包等,文中有详细的代码示例供大家参考,需要的朋友可以参考下
2023-12-12
docker空间爆满导致的进入容器失败的解决方案
这篇文章主要介绍了docker空间爆满导致的进入容器失败的解决方案，具有很好的参考价值，希望对大家有所帮助。一起跟随小编过来看看吧
2021-03-03
详解Docker Compose 中可用的环境变量问题
Compose 的多个部分在某种情况下处理环境变量。这篇文章主要介绍了Docker Compose 中可用的环境变量问题，对大家的学习或者工作具有一定的参考学习价值，需要的朋友们下面随着小编来一起学习学习吧
2018-12-12
docker 动态映射运行的container端口实例详解
这篇文章主要介绍了 docker 动态映射运行的container端口实例详解的相关资料,需要的朋友可以参考下
2016-10-10
基于docker 搭建Prometheus+Grafana的过程详解
这篇文章主要介绍了基于docker 搭建Prometheus+Grafana的过程详解,本文通过图文并茂的形式给大家介绍的非常详细，对大家的学习或工作具有一定的参考借鉴价值，需要的朋友可以参考下
2021-03-03
docker安装RabbitMq添加用户实现方式
这篇文章主要介绍了docker安装RabbitMq添加用户实现方式,具有很好的参考价值,希望对大家有所帮助,如有错误或未考虑完全的地方,望不吝赐教
2024-07-07
从进程中去理解 Docker隔离技术
这篇文章主要介绍了从进程中去理解 Docker隔离技术，Docker 公司发布的 Docker 项目具有里程碑式的意义；Docker 项目通过“容器镜像”，解决了应用打包这个根本性难题，下文更多相关资料,需要的朋友可以参考一下
2022-04-04