python-fastapi

Python FastAPI

A FastAPI application running on Uvicorn, managed by Velos.

What this configures

• API: uvicorn main:app with 4 workers, 512 MB memory limit
• Auto-restart: restarts on crash, up to 10 attempts
• Log rotation: 20 MB per file, 14 rotated files retained
• Env profiles: production and development

Quick start

# Install dependencies
pip install -r requirements.txt

# Start with Velos
velos start --config velos.toml

# Start in development (with file watching)
# Set watch = true in velos.toml first
velos start --config velos.toml --env development

# Test the API
curl http://localhost:8000/
curl http://localhost:8000/health

# View logs
velos logs api

Adding Celery

The config includes a commented-out Celery worker section. To enable it:

1. Install Celery: pip install celery[redis]
2. Uncomment the [apps.celery] section in velos.toml
3. Start a Redis broker: redis-server
4. Run: velos start --config velos.toml

Notes

• PYTHONUNBUFFERED=1 ensures Python output appears in logs immediately
• Adjust --workers based on your CPU cores (rule of thumb: 2 * cores + 1)
• For development, set watch = true to auto-reload on code changes