Kubernetes project

This Kubernetes project was edited by Mr. Nicolas Muller in parallel with the Treeptik learning platform.

Student informations

Firstname : Charley Lastname : GEOFFROY Filiere : DO3 Year : 2022-2023

Table of contents

How it works

This project is a Kubernetes project that will be used to deploy a frontend, a backend and a database. The frontend will be a SSR application using Next.js, the backend will be a REST API using Express and the database will be a Postgres database.

A CI github actions is configured to build the docker images and push them to the github registry. The images are then pulled from the registry and deployed on the Kubernetes cluster.

Images build by the CI

• charley-kube-frontend
• charley-kube-backend

Install kind

[ $(uname -m) = x86_64 ] && curl -Lo ./kind https://kind.sigs.k8s.io/dl/v0.20.0/kind-$(uname)-amd64
chmod +x ./kind
sudo mv ./kind /usr/local/bin/kind

Add autocomplete to kind by editing your ~/.bashrc file and adding the following line:

source <(kind completion bash)

Create cluster

kind create cluster tp-kube

Create all namespaces needed

For each namespace you want to create, you can use the following command (frontend, backend, db):

kubectl create namespace <namespace>

Create dry-run deployments

For each deployment you want to create, you can adapt the following command in the appropriate working directory, this will create a deployment.yaml file that you can adapt and apply:

kubectl create deploy <deployment-name> --image=<image> -n <namespace-required> --dry-run=client -o yaml > deployment.yaml

Create dry-run services

For each service you want to create, you can adapt the following command in the appropriate working directory, this will create a service.yaml file that you can adapt and apply:

kubectl create service <service-type> <service-name> --tcp=<port>:<port> -n <namespace-required> --dry-run=client -o yaml > service.yaml

Create dry-run secret

For each secret you want to create, you can adapt the following command in the appropriate working directory, this will create a secret.yaml file that you can adapt and apply:

kubectl create secret generic <secret-name> --from-literal=<key>=<value> -n <namespace-required> --dry-run=client -o yaml > secret.yaml

Run the project

Create Namespace

kubectl create -R -f infra/ns

Apply and delete all kubernetes objects

From the working directory, you can apply this command to apply all kubernetes objects:

kubectl apply -R -f infra/

You can also delete all kubernetes objects:

kubectl delete -R -f infra/

To apply specific objects, you can use the following command:

kubectl apply -f <object-name>.yaml

Apply port-forwarding to access the webapp

kubectl port-forward services/webapp-service 5000:80 -n frontend

Then you can access on heathcheck page on ==> localhost:5000/healthz

In order to access the backend API, the frontend sends a request to backend-service.backend.svc.cluster.local:5000 you can find the serverRuntimeConfig in the frontend/next.config.js file.

TP - Monitoring and Dashboarding