Hi team,

I’m deploying a Django app on Railway with Gunicorn/Uvicorn as the ASGI server and Django Q for background tasks/scheduled jobs (push notifications, etc.).

Right now, I’m starting qcluster inside my container entrypoint.sh alongside Gunicorn. It kind of works, but I’m noticing:

• Extremely slow response times on web requests (sometimes hanging for 1–2 minutes).

• Logs suggest qcluster it is competing with Gunicorn workers

I know the “ideal” solution is to run separate services (web and worker) like:

web: gunicorn fitease_backend.asgi:application -w 2 -k uvicorn.workers.UvicornWorker -b 0.0.0.0:$PORT
worker: python manage.py qcluster

But that would count as two Railway services, which I’m trying to avoid for cost reasons.

This is my entrypoint script and below is my dockerfile.

# Stage 1: Builder FROM python:3.13 AS builder RUN pip install pipenv WORKDIR /app COPY Pipfile Pipfile.lock ./ RUN pipenv install --deploy --system # Stage 2: Final image FROM python:3.13 WORKDIR /app # Copy installed deps COPY --from=builder /usr/local/lib/python3.13 /usr/local/lib/python3.13 COPY --from=builder /usr/local/bin /usr/local/bin # Copy project files COPY . . RUN apt-get update && \ apt-get install -y --no-install-recommends ffmpeg libjemalloc2 && \ rm -rf /var/lib/apt/lists/* ENV LD_PRELOAD=libjemalloc.so.2 # optional tuning ENV MALLOC_CONF=background_thread:true,dirty_decay_ms:500,muzzy_decay_ms:500 # Copy entrypoint script COPY entrypoint.sh /app/entrypoint.sh RUN chmod +x /app/entrypoint.sh # Create non-root user RUN useradd -m appuser RUN chown -R appuser:appuser /app USER appuser ENV PYTHONUNBUFFERED=1 \ PYTHONDONTWRITEBYTECODE=1 \ PYTHONPATH=/app # Run your script first, then launch the app ENTRYPOINT ["/app/entrypoint.sh"] CMD ["gunicorn", "fitease_backend.asgi:application", "--bind", "0.0.0.0:8000", "-w", "1", "-k", "uvicorn_worker.UvicornWorker", "--max-requests", "2000", "--max-requests-jitter", "20"]