Containerizing Python Apps

I have an extensive Python script that I currently run in a dev environment. Normally, I’d use n8n to automate workflows like this, but n8n’s Python node runs Pyodide, which is a WebAssembly-based Python interpreter. That’s fine for light tasks but it doesn’t support many critical features, including:

• File system access
• Custom dependencies (e.g. psycopg2, pandas, etc.)
• Making raw requests with custom headers
• Multithreading or subprocess usage

All of which my script leverages. Because of these limitations, I decided to containerize the script, deploy it through Portainer, and monitor it with Uptime Kuma. It’s a very clean and repeatable way to deploy Python apps on any Linux box or VM.

1. Project Structure

Create a project folder on the server, in this case we’ll call it python-script:

mkdir python-script && cd python-script

Inside this folder, create:

python-script/
├── Dockerfile
├── requirements.txt
└── app.py

---

2. Dockerfile

FROM python:3.11-slim

WORKDIR /app

# Install dependencies
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Copy the script
COPY app.py .

# Run it
CMD ["python", "app.py"]

Change the Python version tag if needed (e.g., 3.10-slim or 3.12-slim).

---

3. Build the Docker image

From inside the project folder:

docker build -t python-script:latest .

---

4. Deploy it with Portainer

You can absolutely run this with Docker Compose locally or as a standalone container. But I used Portainer so I could monitor logs more easily from the web UI.

In Portainer:

1. Go to Stacks > Add Stack
2. Give it a name like python-script
3. Paste in the following YAML:

version: '3.8'

services:
  python_script:
    image: python-script:latest
    container_name: python-script
    restart: always
    command: python -u app.py  # -u forces unbuffered output so logs show up live in Portainer

Without -u, Docker will buffer stdout and your print() logs won’t show up in Portainer until the buffer flushes (which might never happen if your script runs indefinitely).

---

5. Monitor with Uptime Kuma

If your script runs continuously or at regular intervals, you can:

• Add a heartbeat call to Healthchecks.io or a similar service
• Or expose a simple /health endpoint via Flask or FastAPI inside the container and point Uptime Kuma at it
• Monitor the container itself with Uptime Kuma's Docker Container monitor.

For example, to use a basic heartbeat with Healthchecks.io:

import requests

requests.get("https://hc-ping.com/YOUR-UUID/start")

# your logic...

requests.get("https://hc-ping.com/YOUR-UUID")

---

Wrap-up

This method is lightweight and works great for:

• Data ETL jobs
• API polling scripts
• Scheduled sync tasks
• Anything you’d otherwise try to cram into cron or n8n