Kubernetes and Helm training

The problem statement

You work in an assigment that is creating the next revolutiary TODO application.

You are the engineer maintaining these two applications: One frontend end user web application and a back end application used by the front end to process incoming requests. These two applications can be built and deployed locally via Docker using the docker-composer.yml provided.

This works well for development, but to ensure production scale deployment, you need to make use of Azure Kubernetes Service.

Your tasks

Manually writing Kubernetes manifests can be tedious and error prone, so you will use Helm to better manage and package your applications.

Create the relevant Helm chart under the helm directory to produce a compatible Deployment that will achieve the following criteria:

• Two instances of the front end application are always running and one instance of the back end application;
• The front end application is public available (reachable from the internet) and the back end is not;
• A Load balancer will must evenly distribute the requests (round-robbin).

Expected deliverables

• A public link where the application can be accessed;
• A tour to the created k8s manifests and rationale of each of them;
• A way to reproduce the deployment multiple times;
• If you deploy multiple instances of the back end, you will notice that the application doesn't behave as intended, can understand the problem and propose a solution to fix that? BONUS: can you implement that solution?

Guidelines and resources

• Use your personal Azure subscription (not the corporate account);
• Use AKS Quickstart for sample instructions.

Technical requirements and development environment

• You need Docker;
• You need Python3.

In your WSL environment create a Python3 virtual environment:

$ sudo apt install python3.8-venv
$ python3 -m venv k8s-training
$ . k8s-training/bin/activate

Building the applications:

$ ./build.bash

Start the applications via:

$ docker-compose up

Launch a web-browser at http://localhost:5001,