Docker Three Musketeers: Docker swarm

Read the catalog:

Docker Machine creates a Docker host
Docker Swarm configuration cluster node
Docker Service deploys a single cluster service
Docker Stack deploys multiple cluster services and GUI management pages
Common commands of docker-machine, docker swarm, docker node, docker service and docker stack

Docker SwarmLike Docker Compose, they are all official Docker container orchestration projects, but the difference is that Docker Compose is a tool for creating multiple containers on a single server or host, while Docker Swarm can create container clusters on multiple servers or hosts Services, for the deployment of microservices, obviously Docker Swarm will be more suitable.

Starting from Docker 1.12.0 version, Docker Swarm has been included in the Docker engine (docker swarm), and the service discovery tool has been built in, we do not need to configure Etcd or as beforeConsulCome to configure service discovery.

1. Docker Machine creates a Docker host

Before configuring Docker Swarm, we also need to talk about another official Docker tool, Docker Machine (also one of the three Docker Musketeers), whose role is to quickly help us build a Docker host environment. For example, if we want to use Docker Swarm, we must have a lot Docker host to operate, Docker Machine is the most ideal tool.

Because I am operating on Mac OS, and Docker for Mac already includes Docker Machine (docker machine), so I don't need to install it anymore. If you use Linux system, the installation is very simple, the command:

$ sudo curl -L https://github.com/docker/machine/releases/download/v0.13.0/docker-machine-`uname -s`-`uname -m` > /usr/local/bin/docker-machine
$ sudo chmod +x /usr/local/bin/docker-machine

Well, we first use Docker Machine to create four Docker hosts, the command:

$ docker-machine create -d virtualbox manager1 && 
docker-machine create -d virtualbox manager2 && 
docker-machine create -d virtualbox worker1 && 
docker-machine create -d virtualbox worker2

Running pre-create checks...
(worker1) No default Boot2Docker ISO found locally, downloading the latest release...
(worker1) Latest release for github.com/boot2docker/boot2docker is v17.11.0-ce
(worker1) Downloading /Users/xishuai/.docker/machine/cache/boot2docker.iso from https://github.com/boot2docker/boot2docker/releases/download/v17.11.0-ce/boot2docker.iso...

Execute the above command, you will find the speed is very slow (as above), the reason is to download one from GitHubboot2docker.isoFile (no way for domestic network), how to solve it? Very simple, we use X browser to download manuallyboot2docker.isoFile, and then copy to the corresponding directory (my computer's directory/Users/xishuai/.docker/machine/cache/), and then execute the above command to find a batch of fast.

We can view the Docker host information created below, the command:

$ docker-machine ls
NAME       ACTIVE   DRIVER       STATE     URL                         SWARM   DOCKER        ERRORS
manager1   -        virtualbox   Running   tcp://192.168.99.100:2376           v17.11.0-ce   
manager2   -        virtualbox   Running   tcp://192.168.99.101:2376           v17.11.0-ce   
worker1    -        virtualbox   Running   tcp://192.168.99.102:2376           v17.11.0-ce   
worker2    -        virtualbox   Running   tcp://192.168.99.103:2376           v17.11.0-ce

As you can see, we have created four Docker hosts (two Managers and two Workers), and we can also connect to any server to operate, the command:

$ docker-machine ssh manager1
                        ##         .
                  ## ## ##        ==
               ## ## ## ## ##    ===
           /"""""""""""""""""\___/ ===
      ~~~ {~~ ~~~~ ~~~ ~~~~ ~~~ ~ /  ===- ~~~
           \______ o           __/
             \    \         __/
              \____\_______/
 _                 _   ____     _            _
| |__   ___   ___ | |_|___ \ __| | ___   ___| | _____ _ __
| '_ \ / _ \ / _ \| __| __) / _` |/ _ \ / __| |/ / _ \ '__|
| |_) | (_) | (_) | |_ / __/ (_| | (_) | (__|   <  __/ |
|_.__/ \___/ \___/ \__|_____\__,_|\___/ \___|_|\_\___|_|
Boot2Docker version 17.11.0-ce, build HEAD : e620608 - Tue Nov 21 18:11:40 UTC 2017
Docker version 17.11.0-ce, build 1caf76c

2. Docker Swarm configuration cluster node

We execute the following command:

$ docker-machine ssh manager1 "docker swarm init --advertise-addr 192.168.99.100"
Swarm initialized: current node (n0ub7dpn90rxjq97dr0g8we0w) is now a manager.

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

    docker swarm join --token SWMTKN-1-5uwpqibnvmho1png8zmhcw8274yanohee32jyrcjlait9djhsk-envtxo4dl6df2ar3qldcccfdg 192.168.99.100:2377

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

The above ismanager1On the host, create a Docker Swarm management node (when initializing the cluster, the current node will be automatically set as the management node).

Then we areworker1withworker2On the host, create two working nodes and join them to the cluster. Command:

$ docker-machine ssh worker1 "docker swarm join --token SWMTKN-1-5uwpqibnvmho1png8zmhcw8274yanohee32jyrcjlait9djhsk-envtxo4dl6df2ar3qldcccfdg 192.168.99.100:2377"
This node joined a swarm as a worker.

$ docker-machine ssh worker2 "docker swarm join --token SWMTKN-1-5uwpqibnvmho1png8zmhcw8274yanohee32jyrcjlait9djhsk-envtxo4dl6df2ar3qldcccfdg 192.168.99.100:2377"
This node joined a swarm as a worker.

There is another onemanager2The host needs to be configured as a management node, we need to firstmanager1On the host, get the correspondingtoken, And then configured as a management node, the command:

$ docker-machine ssh manager1 "docker swarm join-token manager"
To add a manager to this swarm, run the following command:

    docker swarm join --token SWMTKN-1-5uwpqibnvmho1png8zmhcw8274yanohee32jyrcjlait9djhsk-0koz1b98sco8r5cn3g61eahnu 192.168.99.100:2377

$ docker-machine ssh manager2 "docker swarm join --token SWMTKN-1-5uwpqibnvmho1png8zmhcw8274yanohee32jyrcjlait9djhsk-0koz1b98sco8r5cn3g61eahnu 192.168.99.100:2377"
This node joined a swarm as a manager.

After configuration, we entermanager1In the host (the above command can also be executed in the host), and then view the information of the cluster nodes, the command:

$ docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS
n0ub7dpn90rxjq97dr0g8we0w *   manager1            Ready               Active              Leader
t4cy67qp0bf2spgabsutwxnzt     manager2            Ready               Active              Reachable
if0kmzp4ww3oy57y7cha7v36t     worker1             Ready               Active              
jgg61cujzaeb3du5796fm0x2g     worker2             Ready               Active

LeaderOf course the head of the cluster,ReachableCan be understood as the candidate of the head, it will go up as soon as the head hangs.

It should be noted that after I configured it that day, all Docker hosts werestopAgain, and then the next daystartAfter that, the following problems appeared:

docker node ls
Error response from daemon: rpc error: code = Unknown desc = The swarm does not have a leader. It's possible that too few managers are online. Make sure more than half of the managers are online.

It seems that the cluster node has lost its head. Related questions:How to deal with "The swarm does not have a leader" problem of docker swarm cluster, According to the article to solve:

$ docker swarm init --force-new-cluster
Error response from daemon: could not choose an IP address to advertise since this system has multiple addresses on different interfaces (10.0.2.15 on eth0 and 192.168.99.102 on eth1) - specify one with --advertise-addr
$ docker swarm init --force-new-cluster --advertise-addr 192.168.99.102
Error response from daemon: This node is not a swarm manager. Worker nodes can't be used to view or modify cluster state. Please run this command on a manager node or promote the current node to a manager.
$ docker node ls
Stuck
$ docker-machine restart manager1 
Can't restart, keep turning

No way, I can only delete four Docker hosts and recreate it.

3. Docker Service deploys a single cluster service

Before deploying the cluster service, we need to do some preparations, because the Docker image acceleration address is not configured in the Docker host, so when pulling the official image, it will definitely be very slow. In addition to configuring the Docker image acceleration address, we can also Use Docker private image warehouse to solve this problem.

Reference article:Ubuntu Docker Registry builds a private warehouse

Here, I will briefly explain the configuration steps. First, execute the following command on Mac OS:

$ docker run -d -v /Users/xishuai/Documents/Docker:/var/lib/registry -p 5000:5000 --restart=always --name registry registry

$ docker tag nginx 192.168.99.1:5000/nginx:latest && 
docker push 192.168.99.1:5000/nginx:latest && 
docker pull 192.168.99.1:5000/nginx:latest

$ curl http://192.168.99.1:5000/v2/_catalog
{"repositories":["nginx"]}

We created a private warehouse container on Mac OS and putnginxThe image is placed in a private warehouse, because Https is not used, so when pulling and pushing the image, the following errors will be reported (Mac OS and Docker hosts will report errors):

$ docker pull 192.168.99.1:5000/nginx:latest
The push refers to a repository [192.168.99.1:5000/nginx]
Get https://192.168.99.1:5000/v1/_ping: http: server gave HTTP response to HTTPS client

Solution, we need to add configuration to the four Docker hosts separately (Docker for Mac can be configured in the management interface), the command:

$ sudo touch /etc/docker/daemon.json && 
sudo chmod 777 /etc/docker/daemon.json && 
sudo echo '{ "insecure-registries":    ["192.168.99.1:5000"] }' > /etc/docker/daemon.json

Then restart the four Docker hosts (Docker for Mac also needs to be restarted), the command:

$ docker-machine restart manager1 && 
docker-machine restart manager2 && 
docker-machine restart worker1 && 
docker-machine restart worker2

The above is more verbose, we will officially deploy cluster services next, or takenginxMirror as an example, the command (docker service createcommandDetailed description）：

$ docker service create --replicas 4 -p 8088:80 --name nginx 192.168.99.1:5000/nginx:latest
ap8h8srb8yh3mni0h2nz61njz
overall progress: 4 out of 4 tasks 
1/4: running   [==================================================>] 
2/4: running   [==================================================>] 
3/4: running   [==================================================>] 
4/4: running   [==================================================>] 
verify: Service converged

have to be aware of is,--replicas 4Represents the number of service creation instances (the default is one), what does it mean? For example, 4 is to create one on each of the four Docker hostsnginxService, if it is 3, it is three Docker hosts, or you can understand the number of Docker hosts, in addition,REPLICASThere will be a progress display, and the execution is asynchronous.

We can also manually set the number of instances, the command:

$ docker service scale nginx=4

After the service is deployed, we can view it. Command:

$ docker service ls
ID                  NAME                MODE                REPLICAS            IMAGE                            PORTS
ap8h8srb8yh3        nginx               replicated          4/4                 192.168.99.1:5000/nginx:latest   *:8080->8080/tcp

$ docker service ps nginx
ID                  NAME                IMAGE                            NODE                DESIRED STATE       CURRENT STATE                ERROR               PORTS
l2rdrwzs5zog        nginx.1             192.168.99.1:5000/nginx:latest   manager1            Running             Running about a minute ago                       
vsfczzbwanx3        nginx.2             192.168.99.1:5000/nginx:latest   manager2            Running             Running about a minute ago                           
qtbgw5h6dsi9        nginx.3             192.168.99.1:5000/nginx:latest   worker              Running             Running about a minute ago                           
za2ejnvb3n6z        nginx.4             192.168.99.1:5000/nginx:latest   worker2             Running             Running about a minute ago

We arbitrarily use one of the four Docker host IP addresses, and the browser opens:http://192.168.99.100:8088/

4. Docker Stack deploys multiple cluster services and GUI management page

docker serviceDeploying a single service, we can usedocker stackFor multi-service orchestration deployment, the same is useddocker-compose.ymlConfiguration file, example:

version: "3"

services:
  nginx:
    image: 192.168.99.1:5000/nginx:latest
    ports:
      - 8088:80
    deploy:
      mode: replicated
      replicas: 4

  visualizer:
    image: 192.168.99.1:5000/dockersamples/visualizer:latest
    ports:
      - "8080:8080"
    volumes:
      - "/var/run/docker.sock:/var/run/docker.sock"
    deploy:
      replicas: 1
      placement:
        constraints: [node.role == manager]

  portainer:
    image: 192.168.99.1:5000/portainer/portainer:latest
    ports:
      - "9000:9000"
    volumes:
      - "/var/run/docker.sock:/var/run/docker.sock"
    deploy:
      replicas: 1
      placement:
        constraints: [node.role == manager]

As shown above, we need to deploy three services in total.nginxService as an example,visualizer（Official address)withportainer（Official address) Are cluster GUI management services.

Deployment command:

$ docker stack deploy -c docker-compose.yml deploy-demo
Creating service deploy-demo_nginx
Creating service deploy-demo_visualizer
Creating service deploy-demo_portainer

After the deployment is successful, we can view the specific details, the command:

$ docker stack ls
NAME                SERVICES
deploy-demo         3

ViewvisualizerGUI cluster management, browser opens:http://192.168.99.100:8080/

ViewportainerGUI cluster management, you need to configure account information first, the browser opens:http://192.168.99.100:9000/

can be seen,portainerratiovisualizerIt's so powerful, even all our operations can be doneportainerComplete.

5. Common commands of docker-machine, docker swarm, docker node, docker service and docker stack

docker-machine common commands

command	Explanation
docker-machine create	Create a Docker host (usually`-d virtualbox`）
docker-machine ls	View all Docker hosts
docker-machine ssh	SSH to the host to execute commands
docker-machine env	Show environment variables needed to connect to a host
docker-machine inspect	Output host more information
docker-machine kill	Stop a host
docker-machine restart	Restart a host
docker-machine rm	Delete a host
docker-machine scp	Copy files between hosts
docker-machine start	Start a host
docker-machine status	View host status
docker-machine stop	Stop a host

docker swarm common commands

command	Explanation
docker swarm init	Initialize the cluster
docker swarm join-token worker	View the token of the worker node
docker swarm join-token manager	View the token of the management node
docker swarm join	Join the cluster

docker node common commands

command	Explanation
docker node ls	View all cluster nodes
docker node rm	Delete a node (`-f`Forced deletion)
docker node inspect	View node details
docker node demote	Node downgrade from management node to working node
docker node promote	Node upgrade, upgrade from working node to management node
docker node update	Update node
docker node ps	View the tasks in the node

docker service common commands

command	Explanation
docker service create	Deployment Service
docker service inspect	View service details
docker service logs	See a service log
docker service ls	View all service details
docker service rm	Delete a service (`-f`Forced deletion)
docker service scale	Set a number of services
docker service update	Update a service

docker stack common commands

command	Explanation
docker stack deploy	Deploy new stack or update existing stack
docker stack ls	List existing stacks
docker stack ps	List tasks in the stack
docker stack rm	Delete stack
docker stack services	List services in the stack
docker stack down	Remove a stack (will not delete data)

Intelligent Recommendation

Docker Three Musketeers: Machine, Compose, Swarm

Docker Three Musketeers In order to maximize the advantages of containerization technology, docker company has launched three major technologies docker-machine docker-compose docker-swarm They can be ...

Docker Series 5: Docker Three Musketeers Docker Swarm

Title: Docker Series 5: Docker Three Musketeers Docker Swarm categories: Docker tags: - Docker timezone: Asia/Shanghai date: 2019-02-19 basic concept node A host running Docker can proactively initial...

Docker Three Musketeers docker-swarm: Know docker-swarm and single-room docker-swarm cluster

docker-swarm One, know docker-swarm 2. Build a docker-swarm cluster One, know docker-swarm 2. Build a docker-swarm cluster Upgrade docker-ce swarm initialization Add server2 and server3 to the managem...

Docker-learning summary (the Docker-swarm practice supplement of the Three Musketeers)

Article Directory Docker-swarm practice supplement Service Custom Network Service mount custom data volume Docker-swarm practice supplement Service Custom Network Overlay is a virtual network created ...

Docker(2)-Docker Swarm cluster of Docker Three Musketeers (manually build, stretch, shrink, cluster monitoring)

Overview of Docker Swarm Docker Swarm, like Docker Compose, is an official Docker container orchestration project But the difference is that Docker Compose is a tool for creating multiple containers o...