Notes Application - Docker & Kubernetes Deployment

Overview

This project is a Notes Application deployed using Docker and Kubernetes (MicroK8s). The application allows users to:

• Create notes
• Delete notes
• Upload files
• Download files

It follows a 3-tier architecture:

• Frontend (Nginx)
• Backend (Node.js)
• Database (MongoDB)

Architecture

User -> Frontend (Nginx) -> Backend (Node.js) -> MongoDB

Containers

Frontend

• Image: nginx:alpine
• Role: Serves static UI
• Port: 80

Backend

• Image: node:18
• Role: API server (handles notes & files)
• Port: 3000

Database

• Image: mongo:6.0
• Role: Stores notes and file data
• Port: 27017

Kubernetes Resources

• Namespace: z2
• Deployments:
  • frontend
  • backend
• StatefulSet:
  • mongo
• Services:
  • frontend (NodePort -> 30007)
  • backend (ClusterIP)
  • mongo (Headless Service)

Storage

• Persistent Volume Claim: mongo-storage
• Size: 1Gi
• Mount path: /data/db
• Purpose: Store MongoDB data persistently

Setup Instructions

1. Prepare Application

./prepare-app.sh

• Builds Docker images
• Loads them into MicroK8s

2. Start Application

./start-app.sh

• Deploys Kubernetes resources
• Starts all services

3. Stop Application

./stop-app.sh

• Stops running pods
• Keeps configuration

4. Delete Application

microk8s kubectl delete all --all -n z2
microk8s kubectl delete pvc --all -n z2

• Removes everything including database

Access Application

Open in browser:

http://<server-ip>:30007

Example:

http://147.232.204.210:30007

Useful Commands

Check pods:

microk8s kubectl get pods -n z2

Check services:

microk8s kubectl get svc -n z2

View logs:

microk8s kubectl logs <pod-name> -n z2

Common Issues

Mongo Crash (version error)

Fix:

microk8s kubectl delete pvc --all -n z2

Frontend shows "Welcome to nginx"

Fix:

• Ensure index.html is copied in Dockerfile
• Rebuild using:

./prepare-app.sh

Honesty Statement

This project was developed with the assistance of AI tools (such as ChatGPT) for:

• Debugging errors
• Understanding Kubernetes concepts
• Identifying fixes for deployment issues

All configurations, testing, and final implementation were completed independently, and the AI was used strictly as a learning aid.