mlinda/devtrack-backend repository overview

DevTrack backend — a realtime, board-based task tracker API.

A Django 5 + Django REST Framework service-layer API for DevTrack, a full-stack task tracker. Boards and tasks are the two core resources; a generic data-driven workflow engine replaces hard-coded statuses with states and transitions stored as data, and Django Channels + Redis push task changes and viewer presence to every client over WebSockets.

The container runs database migrations automatically on startup, then serves the API (ASGI via Daphne) over HTTP and WebSockets.

⁠What's inside

• Django 5 + Django REST Framework service-layer API
• Django Channels (ASGI) + Redis channel layer for realtime
• JWT auth via django-rest-framework-simplejwt
• OpenAPI 3 schema + Swagger UI + ReDoc (drf-spectacular)
• PostgreSQL 16 database, Redis 7 cache / channel layer
• Role-based access, user invites, admin workflow configuration
• Append-only TransitionLog audit trail

⁠Environment

The backend reads configuration from the environment. Set at minimum:

• SECRET_KEY — Django secret key
• DEBUG — True/False
• DJANGO_ALLOWED_HOSTS — comma-separated allowed hosts
• DATABASE_* — PostgreSQL connection settings
• REDIS_* — Redis connection settings

See the repo docs/deployment.md for the full variable list.

⁠Ports

• 8000 — HTTP + WebSocket (health check: GET /api/v1/health/)

⁠Tags

• latest — newest push to main
• main — branch tag
• sha- — immutable build tag for pinning rollouts

⁠Run

The images are designed to run together via docker-compose.prod.yml (with a PostgreSQL, Redis, and the frontend container):

docker compose -f docker-compose.prod.yml pull
docker compose -f docker-compose.prod.yml up -d

Built from backend/Dockerfile.prod (python:3.12-slim base).