Getting Started

🎯 Goal

This section explains how to run the service locally and execute CommandRequest end-to-end.

🧰 Prerequisites

• Python (Poetry-managed dependencies)
• PostgreSQL and Redis
• Playwright Chromium

If you use Docker, Chromium and system dependencies are already included in the Docker image.

🐳 Local run with Docker Compose

The gsc-gui-tools repository contains docker-compose.yml which starts:

• web (Django + Gunicorn)
• celery_worker
• celery_scheduler (Celery Beat)
• command_request_starter (management command that dispatches NEW requests)
• db (Postgres)
• redis

🏗️ 1) Build the image

docker-compose.yml references image gsc-gui-tools-v1, so build it first:

docker build -t gsc-gui-tools-v1 .

▶️ 2) Start the stack

docker compose up

After startup:

• Web: http://localhost:8000/
• Admin: http://localhost:8000/admin/
• API docs: http://localhost:8000/api/docs/

🖥️ 3) VNC (watch the browser UI)

The celery_worker container runs a VNC server to help debug Playwright automation.

• In Docker Compose, VNC is exposed on host port 5901 (mapped to container port 5900).
• The VNC password is currently hardcoded in scripts/Dockerfile — change it before using this setup outside local development.

🐍 Run without Docker (Poetry)

⚙️ Settings module

For local development (DB/Redis on localhost) use config.local_settings:

export DJANGO_SETTINGS_MODULE=config.local_settings

🌐 Web (Django)

poetry install
poetry run python manage.py migrate
poetry run python manage.py createsuperuser
poetry run python manage.py runserver

🧵 Workers / scheduler

Run these in separate terminals:

poetry run celery -A config worker --loglevel=info
poetry run celery -A config beat --loglevel=info
poetry run python manage.py command_request_starter

✅ How to verify

1. Create an Account in Admin
2. Create a CommandRequest (via API or Admin)
3. Ensure command_request_starter dispatches the task
4. Ensure celery_worker executes it and status becomes COMPLETED or FAILED
5. If callback_url was set, ensure a callback is sent