docker swarm 集群的简单实现

1、系统环境：三台CENTOS7的机器

192.168.156.75 管理节点

192.168.156.72 工作节点

192.168.156.77 工作节点

2、分别在三台机器安装docker

yum install docker-io -y

docker pull swarm

修改/lib/systemd/system/docker.service

在中间修改一行

[root@v72 tmp]# cat /lib/systemd/system/docker.service | grep ExecStart
ExecStart=/usr/bin/dockerd-current -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock \

修改/etc/sysconfig/docker-storage

[root@v72 tmp]# cat /etc/sysconfig/docker-storage | grep -v ^#

DOCKER_STORAGE_OPTIONS="--storage-driver overlay"

3、在管理节点上初始化集群

Status: Downloaded newer image for docker.io/swarm:latest
[root@v75 ~]# docker swarm init --advertise-addr 192.168.156.75
Swarm initialized: current node (842qfz9f65cue2cxavyj57p65) is now a manager.

To add a worker to this swarm, run the following command:

docker swarm join \
--token SWMTKN-1-212t8ykaqawrwjb6v4yk7xwas0tyuvuuitxza0r2z9klzcwx9z-7e36rbznzdraxo3n4vtz0hc88 \
192.168.156.75:2377

#上面这一行用于节点加入群集，token是全球唯一的

To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.

此时群集已经激活，并且只有自身一个节点

[root@v75 ~]# docker info|grep -i swarm
WARNING: You're not using the default seccomp profile
Swarm: active
[root@v75 ~]# netstat -tnlp|grep docker
tcp6 0 0 :::2377 :::* LISTEN 8329/dockerd-curren
tcp6 0 0 :::7946 :::* LISTEN 8329/dockerd-curren
[root@v75 ~]# docker network ls
NETWORK ID NAME DRIVER SCOPE
69f2aa51355f bridge bridge local
96f7efaa2b9b docker_gwbridge bridge local
1ec550fc11c2 host host local
p4wufy9vhwsm ingress overlay swarm
212782e1b92b none null local
[root@v75 ~]# docker node ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS
842qfz9f65cue2cxavyj57p65 * v75 Ready Active Leader
[root@v75 ~]# ls /var/lib/docker/swarm
certificates docker-state.json raft state.json worker

4、找到刚才群集初始化的输出，在两台节点机器上执行，加入群集

[root@v77 shell]# docker swarm join --token SWMTKN-1-212t8ykaqawrwjb6v4yk7xwas0tyuvuuitxza0r2z9klzcwx9z-7e36rbznzdraxo3n4vtz0hc88 192.168.156.75:2377
This node joined a swarm as a worker.

此时在群集上面能看到三个节点了

[root@v75 ~]# docker node ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS
842qfz9f65cue2cxavyj57p65 * v75 Ready Active Leader
ojdjeddpgcgfwvndwfwxh4amw vm Ready Active
vje0xnfw59zxb0qyjl37s1qcm v72 Ready Active

5、创建服务

[root@v75 ~]# docker service create --name webs --mode global nginx
xupykl4zucj1018cyazqgq7e3

服务名：webs 模式：global在三个节点上面跑镜像：nginx

将会自动下载镜像启动容器

[root@v75 ~]# docker service ps webs
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS
sa3pt1g93y1x webs.vje0xnfw59zxb0qyjl37s1qcm nginx:latest v72 Running Running 25 minutes ago
rsn23lc0rwwr webs.842qfz9f65cue2cxavyj57p65 nginx:latest v75 Running Running 25 minutes ago
8soldedzagn2 webs.ojdjeddpgcgfwvndwfwxh4amw nginx:latest vm Running Running 25 minutes ago

6、添加到容器到主机的端口映射

[root@v75 ~]# docker service update --publish-add 8001:80 webs
webs

此时通过节点所有主机的IP地址加映射的8001端口，都能访问到容器的80端口 ,只要容器有实例在任意一台机器上面跑即可

docker swarm 集群的简单实现

猜你喜欢