docker-swarm

简介

Swarm的特性：
1.工作节点的注册和发现
2.管理节点收集集群中所有的信息
3.管理节点支持HA
4.管理节点能感知集群的变化，在节点挂掉后重新调度上面的container
5.提供filter和scheduler的调度策略调度集群里的容器

Swarm概念
swarm：是一组docker引擎的集群
node：是单个docker引擎的实例，可以在一个物理机上也可以在多个
application：是应用
manager node：部署应用的时候会有一个manager node节点
Worker nodes：对应的就是Worker nodes
service：然后service是一堆被workder执行的任务
replicated services：是负载均衡节点
global services：则是全局的，在所有节点上都会执行的一个服务
task：一个task就是一个docker的容器，是Swarm的工作单元

部署swarm

环境：
swarm 192.168.1.10
node1 192.168.1.20
node2 192.168.1.30

搭建私有仓库

docker pull registry:2
mkdir -p /opt/data/registry
docker run -dit -p 5000:5000 --restart always -v /opt/data/registry/:/var/lib/registry --name registry registry:2
vim /usr/lib/systemd/system/docker.service 		##添加--insecure-registry 192.168.1.10:5000

ExecStart=/usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock --insecure-registry 192.168.1.10:5000
##其余两台worker同样添加

##下载httpd打包
docker pull httpd
docker tag httpd:latest  192.168.1.10:5000/httpd
docker push 192.168.1.10:5000/httpd
curl 192.168.1.10:5000/v2/_catalog
{"repositories":["httpd","visualizer"]}

部署集群

vim /etc/hosts
192.168.1.20 node1
192.168.1.30 node2

ssh-keygen ##一直回车
ssh-copy-id node1
ssh-copy-id node2

##开启防火墙相应端口
firewall-cmd --add-port=2377/tcp --permanent 
firewall-cmd --add-port=7946/tcp --permanent 
firewall-cmd --add-port=7946/udp --permanent 
firewall-cmd --add-port=4789/udp --permanent 
firewall-cmd --add-port=4789/tcp --permanent 
firewall-cmd --reload

初始化

docker swarm init --advertise-addr 192.168.1.10	##会有提示如何加入集群
docker swarm join-token manager ##查看加入集群命令

docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION
mil1gpiau7ujhfjgnvp5hdzc7 *   swarm               Ready               Active              Leader              19.03.12

##node1与node2   ##加入集群
docker swarm join --token SWMTKN-1-15skd1ufc98589qgpxwzksft7e4ml4bxrkjj2j0lkqooa4pe1p-88tzyad1zi5i77lveqbxmqstn 192.168.1.10:2377


##swarm
docker node ls
docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION
n9okoh4p9gihvr80u9hiwoetu     node1               Ready               Active                                  19.03.12
nvnjwoixsql85aveph6247lwe     node2               Ready               Active                                  19.03.12
mil1gpiau7ujhfjgnvp5hdzc7 *   swarm               Ready               Active              Leader              19.03.12

下载可视化工具

docker pull dockersamples/visualizer
docker run -dit -p 8888:8080 -e HOST=192.368.1.10 -v /var/run/docker.sock:/var/run/docker.sock 192.168.1.10:5000/visualizer
##访问  ip：8888

部署service

##swarm
docker service create --name web_server 192.168.1.10:5000/httpd

docker service ls
ID                  NAME                MODE                REPLICAS            IMAGE                            PORTS
n164bmq4u7uc        web_server          replicated          3/3                 192.168.1.10:5000/httpd:latest   

REPLICAS 显示当前副本信息，0/1 的意思是 web_server 这个 service 期望的容器副本数量为 1，目前已经启动的副本数量为 0。也就是当前 service 还没有部署完成。命令 docker service ps 可以查看 service 每个副本的状态。

docker service ps web_server 
ID                  NAME                IMAGE                            NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS
8amf6th0zfh9        web_server.1        192.168.1.10:5000/httpd:latest   node2               Running             Running 15 seconds ago                       

service伸缩

docker service scale web_server=5
web_server scaled to 5
overall progress: 5 out of 5 tasks 
1/5: running   [==================================================>] 
2/5: running   [==================================================>] 
3/5: running   [==================================================>] 
4/5: running   [==================================================>] 
5/5: running   [==================================================>] 
verify: Service converged 


docker service ps web_server 
ID                  NAME                IMAGE                            NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS
8amf6th0zfh9        web_server.1        192.168.1.10:5000/httpd:latest   node2               Running             Running 3 minutes ago                        
rog6rlrpke84        web_server.2        192.168.1.10:5000/httpd:latest   swarm               Running             Running 18 seconds ago                       
vt9goq7efi0v        web_server.3        192.168.1.10:5000/httpd:latest   node2               Running             Running 18 seconds ago                       
nlc7x2st1ay8        web_server.4        192.168.1.10:5000/httpd:latest   node1               Running             Running 18 seconds ago                       
g44z1ksntkk8        web_server.5        192.168.1.10:5000/httpd:latest   swarm               Running             Running 18 seconds ago                       

实现 Failover

##关闭node1
poweroff
##还是5个服务

访问 service

在创建service时添加 --publish-add 暴露端口如：

docker service create  --publish 8080:80 192.168.1.10:5000/httpd

对于已经创建的service使用更新的命令

docker service update --publish-add 8080:80 web_server

验证

curl 192.168.1.10:8080

It works!

curl 192.168.1.20:8080

It works!

curl 192.168.1.30:8080

It works!