Docker : Using Create command, Starting an already created container, Attaching Shell and Docker Clustering with Swarm

Creating container, start & attach:

The docker create command creates a writeable container layer over the specified image and prepares it for running the specified command. The container ID is then printed to STDOUT. This is similar to docker run -d except the container is never started. You can then use the docker start <container_id> command to start the container at any point.

This is useful when you want to set up a container configuration ahead of time so that it is ready to start when you need it. The initial status of the new container is created.

-bash-4.2$ sudo docker create -it --name shan_container centos /bin/bash

0113ebdc09228e632be3799576e3a18ea1b5dec8c5bcefa480baf27c82c969c2

-bash-4.2$ sudo docker start shan_container

shan_container

-bash-4.2$ sudo docker attach shan_container

[root@0113ebdc0922 /]# ls

bin  dev  etc  home  lib  lib64  lost+found  media  mnt  opt  proc  root  run  sbin  srv  sys  tmp  usr  var

[root@0113ebdc0922 /]# exit

exit

-bash-4.2$ sudo docker ps -a | more

CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS                      PORTS

           NAMES

0113ebdc0922        centos              "/bin/bash"              3 minutes ago       Exited (0) 5 seconds ago

           shan_container

Build Custom Image and Run it as container:

The docker build command builds Docker images from a Dockerfile and a “context”. A build’s context is the set of files located in the specified PATH or URL. The build process can refer to any of the files in the context. Here, the docker image will be created following the steps mentioned in dockerfile.

-bash-4.2$ pwd

/home/dockmgr/customimage

-bash-4.2$ ls -la

total 8

drwxr-xr-x. 2 root root  40 Nov 30 08:21 .

drwxr-xr-x. 3 root root  25 Nov 30 07:54 ..

-rw-r--r--. 1 root root 230 Nov 30 08:21 dockerfile

-rwxr-xr-x. 1 root root  22 Nov 30 07:56 hello.py

-bash-4.2$ cat hello.py

print ('hello world')

-bash-4.2$ cat dockerfile

FROM centos:latest

MAINTAINER Shantanu Mukherjee (The guy from hell)

RUN yum update -y

RUN yum install -y python3 python3-pip

ADD hello.py /home/dockmgr/customimage/hello.py

CMD /usr/bin/python3 /home/dockmgr/customimage/hello.py

-bash-4.2$

-bash-4.2$ sudo docker build -t newcustomimage .

Sending build context to Docker daemon  3.072kB

Step 1/6 : FROM centos:latest

 ---> 0f3e07c0138f

Step 2/6 : MAINTAINER Shantanu Mukherjee (The guy from hell)

 ---> Using cache

 ---> 94df67869637

Step 3/6 : RUN yum update -y

 ---> Using cache

 ---> 4a9978cbf15b

Step 4/6 : RUN yum install -y python3 python3-pip

 ---> Using cache

 ---> 40d870e95e3b

Step 5/6 : ADD hello.py /home/dockmgr/customimage/hello.py

 ---> Using cache

 ---> de97be695628

Step 6/6 : CMD /usr/bin/python3 /home/dockmgr/customimage/hello.py

 ---> Running in e4ae408496e7

Removing intermediate container e4ae408496e7

 ---> 81c09795365b

Successfully built 81c09795365b

Successfully tagged newcustomimage:latest

-bash-4.2$ sudo docker run -it newcustomimage

hello world

-bash-4.2$ sudo docker logs thirsty_taussig

hello world

--------------------------------------------------------------------

Docker swarm:

The first thing to do is initialize the Swarm. We will SSH into the node1 machine and initialize the swarm in there.

[node1] (local) root@192.168.0.48 ~

$ ip r

default via 172.18.0.1 dev eth1

172.17.0.0/16 dev docker0 scope link  src 172.17.0.1

172.18.0.0/16 dev eth1 scope link  src 172.18.0.37

192.168.0.0/23 dev eth0 scope link  src 192.168.0.48

[node1] (local) root@192.168.0.48 ~

$ docker swarm init --advertise-addr 192.168.0.48

Swarm initialized: current node (w1ti8n29zbu4okyg4lnhbi6qb) is now a manager.

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

    docker swarm join --token SWMTKN-1-2mu7z4mcqwte82h3i01lawa8rw42mtou3yph4k5xnvghte9mu1-49b2h87lttz0vi9clew9gygqk 192.168.0.48:2377

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

-------------------------------------------------------------------------------------------------------------

Joining as Worker Node:

To find out what docker swarm command to use to join as a node, We will need to use the join-token <role> command.

To find out the join command for a worker, fire the following command:

[node1] (local) root@192.168.0.48 ~

docker swarm join-token worker

docker swarm join --token SWMTKN-1-2mu7z4mcqwte82h3i01lawa8rw42mtou3yph4k5xnvghte9mu1-49b2h87lttz0vi9clew9gygqk 192.168.0.48:2377

On Workers:

[node2] (local) root@192.168.0.47 ~

$ docker swarm join --token SWMTKN-1-2mu7z4mcqwte82h3i01lawa8rw42mtou3yph4k5xnvghte9mu1-49b2h87lttz0vi9clew9gygqk 192.168.0.48:2377

This node joined a swarm as a worker.

[node3] (local) root@192.168.0.46 ~

$ docker swarm join --token SWMTKN-1-2mu7z4mcqwte82h3i01lawa8rw42mtou3yph4k5xnvghte9mu1-49b2h87lttz0vi9clew9gygqk 192.168.0.48:2377

This node joined a swarm as a worker.

[node4] (local) root@192.168.0.45 ~

$ docker swarm join --token SWMTKN-1-2mu7z4mcqwte82h3i01lawa8rw42mtou3yph4k5xnvghte9mu1-49b2h87lttz0vi9clew9gygqk 192.168.0.48:2377

This node joined a swarm as a worker.

------------------------------------------------------------------------------------------------------------

Keep in mind that We can have a node join as a worker or as a manager. At any point in time, there is only one LEADER and the other manager nodes will be as backup in case the current LEADER opts out.

$ [node1] (local) root@192.168.0.48 ~

$ docker node ls

ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION

w1ti8n29zbu4okyg4lnhbi6qb *   node1               Ready               Active              Leader              19.03.4

hl4e8ypdj30ipn9dgokhtfq51     node2               Ready               Active                                   19.03.4

n6gg9dcytsf6p7o3zdyajz9ei     node3               Ready               Active                                  19.03.4

k1urn3eqav6274lvg514qq0nf     node4               Ready               Active                                  19.03.4

Create a Service:

Now that we have our swarm up and running, it is time to schedule our containers on it. This is the whole beauty of the orchestration layer. We are going to focus on the app and not worry about where the application is going to run.

All we are going to do is tell the manager to run the containers for us and it will take care of scheduling out the containers, sending the commands to the nodes and distributing it.

To start a service, We would need to have the following:

What is the Docker image that We want to run. In our case, we will run the standard nginx image that is officially available from the Docker hub.

We will expose our service on port 80.

We can specify the number of containers (or instances) to launch. This is specified via the replicas parameter.

We will decide on the name for our service. And keep that handy.

What I am going to do then is to launch 2 replicas of the nginx container. To do that, I am again in the SSH session for my node1. And I give the following docker service create command:

[node1] (local) root@192.168.0.48 ~

$ docker service create --replicas 2 -p 80:80 --name web nginx

wxy2lg28lgia889uhqlv8dml8

overall progress: 2 out of 2 tasks

1/2: running   [==================================================>]

2/2: running   [==================================================>]

verify: Service converged

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS

rj6z01nxhfxy        web.1               nginx:latest        node2               Running             Running 11 seconds ago                      

ltla907oyhyd        web.2               nginx:latest        node1               Running             Running 11 seconds ago                      

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          2/2                 nginx:latest        *:80->80/tcp

Accessing the Service

We can access the service by hitting any of the manager or worker nodes. It does not matter if the particular node does not have a container scheduled on it. That is the whole idea of the swarm.

Try out a curl to any of the Docker Machine IPs (manager1 or worker1/2/3/4/5) or hit the URL (http://<machine-ip>) in the browser. We should be able to get the standard NGINX Home page.

Scaling up and Scaling down

Here I increased the number of replica to 3. So, it will check if there is any free worker available reporting to leader. If yes, then it will host the service on the new worker. If no, then it will host another instance of the service on the existing nodes where the service is already up.

[node1] (local) root@192.168.0.48 ~

$ docker service update web --replicas 3

web

overall progress: 3 out of 3 tasks

1/3: running   [==================================================>]

2/3: running   [==================================================>]

3/3: running   [==================================================>]

verify: Service converged

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          3/3                 nginx:latest        *:80->80/tcp

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS

rj6z01nxhfxy        web.1               nginx:latest        node2               Running             Running 2 minutes ago                       

ltla907oyhyd        web.2               nginx:latest        node1               Running             Running 2 minutes ago                       

vql6z4y6ryr5        web.3               nginx:latest        node3               Running             Running 19 seconds ago                      

[node1] (local) root@192.168.0.48 ~

$ docker service scale web=4 -- This command will also scale up the service.

web scaled to 4

overall progress: 4 out of 4 tasks

1/4: running   [==================================================>]

2/4: running   [==================================================>]

3/4: running   [==================================================>]

4/4: running   [==================================================>]

verify: Service converged

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          4/4                 nginx:latest        *:80->80/tcp

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE                ERROR               PORTS

rj6z01nxhfxy        web.1               nginx:latest        node2               Running             Running 3 minutes ago                           

ltla907oyhyd        web.2               nginx:latest        node1               Running             Running 3 minutes ago                           

vql6z4y6ryr5        web.3               nginx:latest        node3               Running             Running about a minute ago                      

m9mhoz07vcts        web.4               nginx:latest        node4               Running             Running 13 seconds ago                          

[node1] (local) root@192.168.0.48 ~

[node1] (local) root@192.168.0.48 ~

$ docker service scale web=8   -- Further scaling up to 8 instances of the service.

web scaled to 8

overall progress: 8 out of 8 tasks

1/8: running   [==================================================>]

2/8: running   [==================================================>]

3/8: running   [==================================================>]

4/8: running   [==================================================>]

5/8: running   [==================================================>]

6/8: running   [==================================================>]

7/8: running   [==================================================>]

8/8: running   [==================================================>]

verify: Service converged

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS

rj6z01nxhfxy        web.1               nginx:latest        node2               Running             Running 7 minutes ago                        

ltla907oyhyd        web.2               nginx:latest        node1               Running             Running 7 minutes ago                       

vql6z4y6ryr5        web.3               nginx:latest        node3               Running             Running 5 minutes ago                       

m9mhoz07vcts        web.4               nginx:latest        node4               Running             Running 4 minutes ago                       

fj48c9k8ylxp        web.5               nginx:latest        node2               Running             Running 55 seconds ago                      

s6yjswf31ab7        web.6               nginx:latest        node3               Running             Running 55 seconds ago                      

8nkb2f5ao6cd        web.7               nginx:latest        node4               Running             Running 54 seconds ago                      

qy65i1jgndqh        web.8               nginx:latest        node1               Running             Running 54 seconds ago                      

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          8/8                 nginx:latest        *:80->80/tcp

[node1] (local) root@192.168.0.48 ~

$ docker service scale web=5    --- Scaling down the service from 8 to 5 instances.

web 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

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          5/5                 nginx:latest        *:80->80/tcp

[node1] (local) root@192.168.0.48 ~

If the node is ACTIVE, it is ready to accept tasks from the Master i.e. Manager. That is why if we need to do some maintenance on a node or if we want to restrict a node to get instructions from master, we drain the particular node. When we drain the instance of the service hosted gets migrated to some other node.

$ docker node update --availability drain node2  

node2

[node1] (local) root@192.168.0.48 ~

$ docker service ls

ID                  NAME                MODE                REPLICAS            IMAGE               PORTS

wxy2lg28lgia        web                 replicated          5/5                 nginx:latest        *:80->80/tcp

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS

xv3g1t50tnqz        web.1               nginx:latest        node1               Running             Running 6 seconds ago                        

rj6z01nxhfxy         \_ web.1           nginx:latest        node2               Shutdown            Shutdown 8 seconds ago                      

ltla907oyhyd        web.2               nginx:latest        node1               Running             Running 10 minutes ago                      

vql6z4y6ryr5        web.3               nginx:latest        node3               Running             Running 8 minutes ago                       

m9mhoz07vcts        web.4               nginx:latest        node4               Running             Running 7 minutes ago                       

klcjxyq1c6om        web.5               nginx:latest        node3               Running             Running 6 seconds ago                       

fj48c9k8ylxp         \_ web.5           nginx:latest        node2               Shutdown            Shutdown 7 seconds ago                      

[node1] (local) root@192.168.0.48 ~

This is straight forward. In case We have an updated Docker image to roll out to the nodes, all We need to do is fire an service update command.

$ docker service update --image nginx:latest web

web

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

[node1] (local) root@192.168.0.48 ~

Remove the Service

We can simply use the service rm command as shown below:

$ docker service rm web

web

[node1] (local) root@192.168.0.48 ~

$ docker service ps web

no such service: web

[node1] (local) root@192.168.0.48 ~

$