Time: Approximately 10 minutes
Difficulty: Easy
- Create a Docker Hub account
- Set Up 'Play With Docker'
- Introduction to Docker Contexts
- How to save your work
By the end of this exercise you will:
- Have created a Docker Hub account (unless you already have one)
- Learn how to use Play With Docker
- Learn how to install new Docker CLI Plugins
- Learn how to use Docker Contexts
- Learn how to save your workshop files
-
If you already have a Docker Hub account, go to step 5
-
Go to https://hub.docker.com and click on the "Sign up for Docker Hub" button.
Details
-
Complete the registration form by providing your Docker ID, password, and an email address.
Details
-
Check your mailbox and confirm your email address.
Details
-
Sign in to Docker Hub to validate your credentials work.
This workshop will be using Play With Docker to help everyone have the same experience and reduce our dependency on the conference WiFi to push/pull GBs of images. You can still follow the instructions locally with your own engine.
-
Go to https://labs.play-with-docker.com
Details
-
Click on the
Logindropdown, chooseDockerand sign in using yourDocker HubcredentialsDetails
-
The
Loginbutton should turn into aStartbutton. Click theStartbutton to get started!Details
-
A new session is opened, which will be closed in 4 hours. You can add multiple new instances and delete them. Each one has it's own engine and own prompt. You can also ssh directly to it if you feel more comfortable with your own terminal.
For this workshop, we will simulate a basic workflow with two environments:
- A local development environment, represented as a single non-Swarm node
- A remote Swarm cluster
-
Create a swarm cluster by clicking on the wrench (:wrench:) and selecting the
1 Manager and 1 Workertemplate.Details
It just added 2 nodes, one manager1 and one worker1.
-
On the manager1 node, run
docker node lsand validate you have two nodes. If you don't, rundocker swarm join-token worker, copy the token, and run the copieddocker swarm joincommand on the worker1 node.Details
On manager1 node:
$ docker node ls ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION du64yyv64u3gpypkqwb8cli39 * manager1 Ready Active Leader 18.09.4 [manager1] (local) root@192.168.0.43 ~ $ docker swarm join-token worker To add a worker to this swarm, run the following command: docker swarm join --token SWMTKN-1-4v4o696r1t6zwcimz8pazuxnwwuh2vjol7zejt33d2erp5wp4g-bbp8kcw4uk7mt9snusp6xh2io 192.168.0.43:2377
Copy the
docker swarm joincommand and run it on the worker1 node.docker swarm join --token SWMTKN-1-4v4o696r1t6zwcimz8pazuxnwwuh2vjol7zejt33d2erp5wp4g-bbp8kcw4uk7mt9snusp6xh2io 192.168.0.43:237 This node joined a swarm as a worker.
Then, on the manager1 node,
docker node lsshould show two nodes.$ docker node ls ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION du64yyv64u3gpypkqwb8cli39 * manager1 Ready Active Leader 18.09.4 z6l2z7terttlxp33kym0rg4tn worker1 Ready Active 18.09.4
-
Now let's add a local development node: click on + ADD NEW INSTANCE
Details
-
We will be doing most of our work from this new instance (node1). Select it.
-
Login to Docker Hub using the
docker logincommand and your credentials.$ docker login Login with your Docker ID to push and pull images from Docker Hub. If you don't have a Docker ID, head over to https://hub.docker.com to create one. Username: dapworkshop Password: WARNING! Your password will be stored unencrypted in /root/.docker/config.json. Configure a credential helper to remove this warning. See https://docs.docker.com/engine/reference/commandline/login/#credentials-store Login Succeeded
To install Docker App, you can download the release directly from the Releases page on GitHub and install it. But, for simplicity, we have packaged it up into a container image to make it easier.
Docker CLI plugins are installed in the ~/.docker/cli-plugins directory.
-
Run the
dapworkshop/workshop:0.9image, mounting the home directory into the container.$ docker run -v ~/:/workshop dapworkshop/workshop:0.9 -
Check
docker appis installed properly$ docker app --version Version: v0.9.0-zeta1 Git commit: 64e685eb Built: Tue Oct 15 15:49:04 2019 OS/Arch: linux/amd64
-
Run
tree .dockerand you'll see the plugin installed!$ tree ~/.docker/ /root/.docker/ ├── cli-plugins │ └── docker-app └── config.json 1 directory, 2 files
Congratulations 👏 You just installed your first Docker CLI plugin on your own!
Now the workshop can really start!
Docker Contexts is a new Docker CLI 19.03 feature. For those who already know about Kubernetes contexts, you will notice that Docker Contexts look familiar.
With contexts, you can quickly switch between your different environments, such as:
- Local daemon
- Remote cluster
- Even locally switch between orchestrators Swarm/Kubernetes
The context commands let you manage all your contexts:
$ docker context
Usage: docker context COMMAND
Manage contexts
Commands:
create Create a context
export Export a context to a tar or kubeconfig file
import Import a context from a tar file
inspect Display detailed information on one or more contexts
ls List contexts
rm Remove one or more contexts
update Update a context
use Set the current docker contextDocker will always come with a default context:
$ docker context ls
NAME DESCRIPTION DOCKER ENDPOINT KUBERNETES ENDPOINT OHESTRATOR
default * Current DOCKER_HOST based configuration unix:///var/run/docker.sock swarmNote: The little star (*) indicates which context is currently in use.
Let's create a new context to let us deploy to our swarm cluster from our development instance. To do so, we will use the docker context create command.
When creating a new context, you are able to define some information for the new context. Some of this information might include:
- A name
- A description
- A default orchestrator on this context
- Some Docker endpoint options, in a comma separated
key=valueformat - Some Kubernetes endpoint options, in a comma separated
key=valueformat
Full details of flags and options available for docker context create
$ docker context create --help
Usage: docker context create [OPTIONS] CONTEXT
Create a context
Docker endpoint config:
NAME DESCRIPTION
from Copy named context's Docker endpoint configuration
host Docker endpoint on which to connect
ca Trust certs signed only by this CA
cert Path to TLS certificate file
key Path to TLS key file
skip-tls-verify Skip TLS certificate validation
Kubernetes endpoint config:
NAME DESCRIPTION
from Copy named context's Kubernetes endpoint configuration
config-file Path to a Kubernetes config file
context-override Overrides the context set in the kubernetes config file
namespace-override Overrides the namespace set in the kubernetes config file
Example:
$ docker context create my-context --description "some description" --docker "host=tcp://myserver:2376,ca=~/ca-file,cert=~/cert-file,key=~/key-file"
Options:
--default-stack-orchestrator string Default orchestrator for stack operations to use with this context (swarm|kubernetes|all)
--description string Description of the context
--docker stringToString set the docker endpoint (default [])
--from string create context from a named context
--kubernetes stringToString set the kubernetes endpoint (default [])-
First we need to get the Swarm manager IP address. Look in the left sidebar for the IP address for the manager1 node.
# Swap in your real IP address MANAGER_IP=192.168.0.7 -
Let's create our context to target the swarm cluster we created. We will simply name it
swarm.$ docker context create swarm --description "remote cluster" --default-stack-orchestrator=swarm --docker "host=tcp://${MANAGER_IP}:2375" swarm Successfully created context "swarm"
-
Run
docker context lsto see the new context.$ docker context ls NAME DESCRIPTION DOCKER ENDPOINT KUBERNETES ENDPOINT ORCHESTRATOR default * Current DOCKER_HOST based configuration unix:///var/run/docker.sock swarm swarm remote cluster tcp://192.168.0.7:2375 swarm
-
To validate the context is setup correctly, let's start a container on the manager node.
- Select the manager1 instance and run
docker run hello-world. - Back on the node instance, run
docker ps -a. You should only see the workshop container we ran earlier.$ docker ps -a CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES dc823bc61e60 dapworkshop/dc2019-workshop "/run.sh" 2 minutes ago Exited (0) 2 minutes ago cocky_minsky
- Now run
docker --context=swarm ps -a. You should now see thehello-worldcontainer.$ docker --context=swarm ps -a CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 1541322b8f2b hello-world "/hello" 1 minute ago Exited (0) 1 minute ago reverent_ptolemy
- Select the manager1 instance and run
-
While the
--contextflag is great, it would be nice to switch our entire CLI to use the swarm cluster. To do so, we use thedocker context usecommand and give it the name of the context to use. Rundocker context use swarmand then see that the various commands are running against that engine.$ docker context use swarm swarm Current context is now "swarm" $ docker context ls NAME DESCRIPTION DOCKER ENDPOINT KUBERNETES ENDPOINT ORCHESTRATOR default Current DOCKER_HOST based configuration unix:///var/run/docker.sock swarm swarm * remote cluster tcp://192.168.0.7:2375 swarm $ docker ps -a CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 1541322b8f2b hello-world "/hello" 2 minutes ago Exited (0) 2 minutes ago reverent_ptolemy
-
We can also create a context from another one. Let's create a development context, targeting our local daemon:
$ docker context create dev --from default --description "development node" $ docker context ls NAME DESCRIPTION DOCKER ENDPOINT KUBERNETES ENDPOINT ORCHESTRATOR default Current DOCKER_HOST based configuration unix:///var/run/docker.sock swarm dev development node unix:///var/run/docker.sock swarm swarm * remote cluster tcp://192.168.0.7:2375 swarm
-
Let's switch back to our development context:
$ docker context use dev dev Current context is now "dev"
A very handy pair of commands is docker context import/export. With these commands, you can quickly configure a new Docker CLI with all the contexts you are used to work with. The export command creates a tarball file with:
- All the context information (host, name, description...)
- All the TLS/Certificates files associated with the context
NOTE: It can also export to a kubeconfig file all your Kubernetes cluster configuration.
-
Use docker context export to export your
swarmcontext.$ docker context export swarm Written file "swarm.dockercontext"
-
Import the context as a new context using
docker context import.$ docker context import imported-swarm swarm.dockercontext imported-swarm Successfully imported context "imported-swarm" $ docker context ls NAME DESCRIPTION DOCKER ENDPOINT KUBERNETES ENDPOINT ORCHESTRATOR default Current DOCKER_HOST based configuration unix:///var/run/docker.sock swarm dev * development node unix:///var/run/docker.sock swarm imported-swarm remote cluster tcp://192.168.0.7:2375 swarm swarm remote cluster tcp://192.168.0.7:2375 swarm
NOTE: import/export support the standard input/output using
-, so you can import/export in one line$ docker context export swarm - | docker context import imported-swarm2 - imported-swarm2 Successfully imported context "imported-swarm2"
-
Go ahead and remove the new context to prevent confusion going forward. Use the
docker context rmcommand to do so.$ docker context rm imported-swarm imported-swarm2 imported-swarm imported-swarm2
You may want to set a reminder to save your exercises locally before leaving the workshop. We will remind you at the end as well..
For those using PWD, you can copy/paste the SSH address and use it to SCP (Secure File Copy through SSH) all your files:
# run this locally
$ scp -rp ip172-18-0-23-bg01fj0v0j3g00d0atfg@direct.labs.play-with-docker.com:/root .