Demo DevOps Python Application

This repository contains a simple Django application that demonstrates various DevOps practices. The application is deployed on a Kubernetes cluster using Helm and GitHub Actions for CI/CD pipelines.

Table of Contents

• Architecture
• Tools
• Setup Instructions
• CI/CD Pipelines
• Kubernetes Deployment
• Contributing
• License

Architecture

The Django API is deployed on a DigitalOcean Kubernetes cluster with the following components:

• Deployment: The API is deployed with 2 replicas, managed by a Horizontal Pod Autoscaler (HPA) that can scale up to 3 replicas.
• Service Account: A basic service account is used to follow security standards.
• LoadBalancer: Exposes the API to the world, accessible via the domain api.algova.dev. It may take about 1 minute.
• DNS and SSL: Managed using Route53 and Let's Encrypt. The domain api.algova.dev points to the LoadBalancer with automatic SSL certificate renewal.
• Static Files: Served from an S3 bucket with proper ACL configurations.

Tools

The following tools were used to develop this project:

• Django: Python library used to power the backend.
• Helm: Used to manage Kubernetes deployments in an organized and clean manner.
• Kubernetes: Where our API is deployed, providing scalability and security.
• Docker: Our API is containerized for easy deployment in any environment.
• Github Actions: Used to deploy our API and to test our PRs before merging.
• Pylint: Keeps our code following high standards.
• Pytest: Ensures proper testing of our API to avoid downtimes.
• Trivy: Keeps us alert to vulnerabilities in our Docker images.

Setup Instructions

The project was divided according to the task to be performed: In all cases, we must clone the repo bash git clone https://github.com/Adriana618/devops-deemo.git cd devops-deemo

Develop:

You must have Minikube, Docker, and Helm installed. bash sudo apt-get update sudo apt-get install -y python3 python3-pip docker.io kubectl minikube pip3 install -r requirements/dev.txt

Test:

You must have Minikube, Docker, and Helm installed. bash sudo apt-get update sudo apt-get install -y python3 python3-pip docker.io kubectl minikube pip3 install -r requirements/test.txt

Deploy:

You must have a Kubernetes cluster with 2 nodes with 1CPU and 2GB of memory at a minimum.

After having the necessary setup for each stage, proceed to fill in the Helm values.yaml file or overwrite the variables at the time of installing the app.

• DJANGO_SECRET_KEY
• DATABASE_NAME='db.sqlite3'
• DJANGO_ALLOWED_HOSTS
• AWS_STORAGE_BUCKET_NAME
• AWS_ACCESS_KEY_ID
• AWS_SECRET_ACCESS_KEY
• AWS_DEFAULT_REGION

2. Dockerize the Application:
   • Dockerfile located at docker-conf/Dockerfile-django-api

CI/CD Pipelines

Deploy API

This GitHub Action deploys commits merged into the master branch to a Kubernetes cluster on DigitalOcean.

See the file .github/workflows/deploy-api.yml

Test PR

This GitHub Action tests pull requests by running static code analysis, code coverage, and vulnerability scans.

See the file .github/workflows/deploy-pr.yml

Kubernetes Deployment

Deployment Configuration

See the file devops-deemo/templates/api/api-deployment.yaml

Service Configuration

See the file devops-deemo/templates/api/api-services.yaml

Secrets

See the files devops-deemo/templates/api/api-secrets.yaml

Contributing

Please follow the contribution guidelines to propose improvements or report issues.

License

This project is licensed under the MIT License - see the LICENSE file for details.