Skip to content

Latest commit

 

History

History
201 lines (145 loc) · 7.51 KB

README.MD

File metadata and controls

201 lines (145 loc) · 7.51 KB
Raw

LAB-K8S-06: ConfigMaps

Description: In this lab, participants will learn how to use ConfigMap, as a first step, a configMap (css file) that will overwrite the existing files.

Duration: ±40m

Goals

At the end of this laboratory, each participant will be able to set up a configMap and its volume to simply modify configurations such as for example a virtual host, an application configuration file etc...

Prerequisites

ConfigMap

When you start developping/integrating an application, what a pleasure to be able to modify parameters in one or more configuration files without having to completely rebuild an image.

A traditional approach is to decorrelate application code from its configuration. (key/value)

And that's where configMaps come in.

WARN : A configMap must not contain sensitive data.

It's time to create our first configMap :

configmap-example

Creating a ConfigMap :

The goal here will be to set up a configmap that will feed a css stylesheet that will override the style of your future application.

content-configmap

  • Deploy this newly created file :
    $ kubectl create -f configmap-css-additional.yml
  • Let's make sure it's been created :
    $ kubectl get configmaps

  • Congratulations, now we'll have to make sure to declare our configmap volume in our deployment :

  • Retrieve the deployment you have used in the past :

kubectl get deployment my-todo-deployment -o yaml > my-todo-deployment.yml

mountPath must match : /usr/src/app/static/css/css-additional (solution here) :

  • Once finished, apply the new configuration of your deployment :
    $ kubectl apply -f my-todo-deployment.yml
  • A little routine check :
    $ kubectl get deployments -o wide -w
    $ kubectl get pods -o wide -w

Using ConfigMaps as an Environment variables

We can create also ConfigMaps with the imperative way:

kubectl create configmap configmap-single-env --from-literal=name=HanSolo

And use the value of the config map inside an Environment variable.

  • ✅ Check documentation
  • ✅ Rework the deployment file to use the value of the key: name from the configmap configmap-single-env in our application container ORG environment variable (solution here)

Using ConfigMaps for multiple variables

We've seen in the first section how to use ConfigMaps directly mounted as volumes which result in a structure of files named by config map keys and containing the corresponding values. We've also seen how to pick an individual environment variable values from a specific configmap key's value.

As part of this new application release, the application team has included a dedicated page for the landing zone specific planet.

  • ✅ Check it out (navigation menu)

By default, it shows the Tatooine information, but our application team has provided a way to customize with the following environment variables:

  • SW_PLANET
  • SW_CLIMATE
  • SW_TERRAIN
  • SW_POPULATION
  • SW_RESIDENT

We don't want to declare and map these enviroment variables one by one, as it would be really long and would result in a complex declaration file. Fortunately, Kubernetes provides a way to declare a set of environment variables with their names and values coming from a ConfigMap.

  • ✅ Read the documentation on how to use this particular notation

  • ✅ If you can't provide all StarWards planets information, you use this link and look for the following information:

    • Planet name
    • Climate
    • Terrain
    • Population
    • Name of a resident of this planet

  • ✅ Create a configmap with the following specifications (solution here) :

  • name is configmap-multiple-env
  • data with correct content: SW_PLANET: "Tatooine" SW_CLIMATE: "Arid" SW_TERRAIN: "Desert" SW_INFO_POPULATION: "200000" SW_INFO_RESIDENT: "Beru Whitesun lars"

  • ✅ Deploy this newly created file :

    kubectl create -f configmap-env-multiple.yml

  • ✅ Verify it's been created :

    kubectl get configmaps

Now we need to modify our deployment to use our configmap as environment variables for our container:

  • ✅ Retrieve the deployment you have used in the past :

    kubectl get deployment my-todo-deployment -o yaml > my-todo-deployment.yml

  • ✅ Rewrite the environment section to use the configmap as a reference. (solution here)

  • ✅ Apply the changes:

    kubectl apply -f my-todo-deployment.yml

  • ✅ Verify on your landing page's dedicated planet section

🍸 Congratulations, that's it for the configMaps, let's move on to Secrets.🍸

Tips

USAGES EXAMPLES
--from-literal sw.parameters.hero: 'Dark Vador!'
--from-file sw.parameters.file: |-
sw.parameters.hero.1 = 'Dark Vador!'
sw.parameters.hero.2 = 'Luke'
CHARACTER SHORTCUT USAGE
> MAJ + < Single-line format
| ALT-G + 6 Multi-line format

Examples :

apiVersion: v1
kind: ConfigMap
metadata:
  name: example
  labels:
    app: example
data:
  example-single-line.yml: > Single-line example

---

apiVersion: v1
kind: ConfigMap
metadata:
  name: example-multi-line
  labels:
    app: example-multi-line
data:
  example-multi-line.yml: | 
  key1:
    value1: 'example1'
  key2:
    value2: 'example1'

Next lab