Dock Swarm 配置以及应用

最新推荐文章于 2024-09-09 08:56:13 发布

原创最新推荐文章于 2024-09-09 08:56:13 发布 · 499 阅读

1 ·

CC 4.0 BY-SA版权

Docker 专栏收录该内容

3 篇文章

订阅专栏

本文档详细介绍了如何配置Docker Swarm环境，包括准备三台虚拟机节点，设置master节点，加入worker节点，以及在Swarm模式下创建service。通过`docker swarm init`、`docker swarm join`命令进行集群初始化和节点加入，并展示了`docker node ls`和`docker network ls`命令用于检查节点和网络状态。最后，创建了名为test1的alpine服务以及一个运行Nginx的service，以实现服务部署。

Docker swarm 基于docker api设计出来，不需要额外的安装

环境准备

三台虚拟机

192.168.1.19 swarm1

192.168.1.6 swarm2

192.168.1.7 swarm3

选择master 节点

docker swarm init --advertise-addr 192.168.1.19

加入swarm cluster

docker swarm join --token SWMTKN-1-46uwk99r3z3toqznatpu1kphrl4d0ut6t5cpv5ipts7bcjfopi-bw0roqkyq5pir4dr1l465l469 192.168.1.19:2377

[root@slave2 ~]# docker swarm join --token SWMTKN-1-46uwk99r3z3toqznatpu1kphrl4d0ut6t5cpv5ipts7bcjfopi-bw0roqkyq5pir4dr1l465l469 192.168.1.19:2377

This node joined a swarm as a worker.

master上查看

[root@slave1 ~]# docker network ls

NETWORK ID NAME DRIVER SCOPE

48abc9b488ff bridge bridge local

3677ccc8add2 docker_gwbridge bridge local

f0cc6cb394a7 host host local

tc5zi3s5sy8f ingress overlay swarm

b11d743337ed none null local

查看角色

[root@slave1 ~]# docker node ls

ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION

w5c5xhybt202zvrqtzts6mrfb * slave1 Ready Active Leader 18.09.4

vfwl8ny249uf5ox2dxzc9zlhx slave2 Ready Active 18.09.4

g5yagqmhha0xki9qwk67ruwy5 slave3 Ready Active 18.09.4

提升slave2 manager

[root@slave1 ~]# docker node promote slave2

Node slave2 promoted to a manager in the swarm.

尝试创建一个service

docker service create --name test1 alpine ping www.baidu.com

docker service create --name nginx --detach=false nginx

暴露映射Nginx的默认端口到宿主机的8080端口


	docker service update --publish-add 8080:80 --detach=false nginx 访问： http://192.168.1.19:8080/ http://192.168.1.7:8080/ http://192.168.1.6:8080/ 三台节点都可以扩容 nginx, swarm将会部署成负载均衡 docker service scale nginx=3 --detatch=false [root@slave1 ~]# docker service ls ID NAME MODE REPLICAS IMAGE PORTS 9iv6vilrha3p nginx replicated 3/3 nginx:latest :8080->80/tcp, :8087->80/tcp lcuojb81tc0c nginx1 replicated 1/1 nginx:latest yb6jivpmol4z test1 replicated 1/1 alpine:latest 使用自定义网络部署应用首先删除 services [root@slave1 ~]# docker service rm nginx nginx1 test1 nginx nginx1 test1 docker network create -d overlay imooc-overlay --detach指定后台运行 docker service create --network imooc-overlay -p 8080:80 --detach=false nginx docker service create --network imooc-overlay --name alpine --detach=false alpine ping www.baidu.com 上alpine所在的机器进入容器内部ping 刚启动的nginx服务（刚才没有给nginx service name, 所以分配了一个随机名称 infallible_brattain） root@slave1 ~]# docker service ps alpine ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS 36g8ok640n6t alpine.1 alpine:latest slave2 Running Running 8 minutes ago docker exec -it d04 sh / # ping infallible_brattain PING infallible_brattain (10.0.0.2): 56 data bytes 64 bytes from 10.0.0.2: seq=0 ttl=64 time=0.064 ms 64 bytes from 10.0.0.2: seq=1 ttl=64 time=0.065 ms 64 bytes from 10.0.0.2: seq=2 ttl=64 time=0.066 ms ^C --- infallible_brattain ping statistics --- 3 packets transmitted, 3 packets received, 0% packet loss round-trip min/avg/max = 0.064/0.065/0.066 ms Docker stack 管理一组服务 Service.yml version: "3.4" services: alpine: image: alpine command: - "ping" - "www.baidu.com" networks: - "imooc-overlay" deploy: replicas: 2 restart_policy: condition: on-failure resources: limits: cpus: "0.1" memory: 50M depends_on: - nginx nginx: image: nginx networks: - "imooc-overlay" ports: - "8080:80" networks: imooc-overlay: external: true docker stack deploy -c service.yml test [root@slave1 ~]# docker stack ls NAME SERVICES ORCHESTRATOR test 2 Swarm [root@slave1 ~]# docker stack services test ID NAME MODE REPLICAS IMAGE PORTS 1n137ttzosyg test_nginx replicated 1/1 nginx:latest *:8080->80/tcp yggli8y0e3tg test_alpine replicated 2/2 alpine:latest Docker swarm 负载均衡两种模式，负载均衡有两种模式：VIP、DNSRR VIP：分配独立的虚拟IP，DNS记录解析到服务名中作为代理IP。（如果需要外部直接访问可以采取这种） dnsrr：DNS记录不解析VIP，而去解析每个容器内的IP。dnsrr模式不支持端口对外暴露。（如果不需要外部访问，只是微服务组件之间的通讯可以考虑这种）其他资料参考 https://www.cnblogs.com/xiangsikai/p/9938862.html