Skip to content

README.md

Latest commit

 

History

History
117 lines (60 loc) · 2.52 KB

README.md

File metadata and controls

117 lines (60 loc) · 2.52 KB

Maths Exam Builder

A lightweight exam-building tool designed for tutors and teachers to generate GCSE-style maths exams. Select from a curated question bank and generate clean, printable PDFs using LaTeX formatting.

This project was born out of a real need I experienced as a maths tutor — I often wanted to create structured, exam-style papers for my students but couldn’t find a tool that was simple, fast, and free-to-use. So I built one. Now, I can generate tailored practice exams for my students in just a few clicks and I hope to scale it so teachers and educators across the UK can use it too.

Exam Builder – Observability + Deployment Case Study

🚀 Overview

I revived an older project of mine, a LaTeX-based exam builder, and used it as a playground for learning real DevOps fundamentals: observability, Docker-based architecture, CI/CD, Linux troubleshooting, and production deployment.

This project now includes:

  • Node backend with /health endpoint

  • Docker Compose architecture

  • CI/CD for SSH deployment with DockerHub

  • Observability stack: Node Exporter, Prometheus, Grafana

  • UptimeRobot external monitoring

  • Nginx reverse proxy + SSL

  • Automated deployments with a custom deploy.sh

🔧 Architecture Diagram

Components:

  • User

  • Nginx reverse proxy

    • Serves backend API

    • Routes grafana.tacknowledge.co.uk to Grafana container

    • Handles SSL termination

  • Backend container (Node)

    • Exposes /api/health

    • Prometheus-ready

  • Node Exporter container

    • System metrics

  • Prometheus container

    • Scrapes Node Exporter

    • Provides metrics to Grafana

  • Grafana container

    • Dashboards + alerts

  • DigitalOcean Droplet

    • All containers running via Docker Compose

    • Volumes persist Grafana + Prometheus data

Exam Builder Backend Architecture

Grafana Node Export Panel

⚙️ Deployment Flow

  1. Push to GitHub

  2. GitHub Actions SSH into droplet with main.yml

  3. Run deploy.sh

  4. Script pulls new backend image from Docker Hub, restarts entire stack using:

docker-compose up -d
  1. Observability stack stays running, with dashboards kept inside Docker volumes.

📊 Observability Features

  • Node Exporter

    • CPU, RAM, Disk, Network, Load

  • Prometheus

    • Scrapes metrics every 15s

  • Grafana

    • Imported "Node Exporter Full (ID: 1860)" dashboard

    • Custom CPU Usage panel

    • SSL-secured grafana.tacknowledge.co.uk

  • UptimeRobot

    • Monitors /api/health

    • External uptime verification