How do i install new python packages in the hosted image

[AryanMithbawkar]

i want to install new packages in python environment of the n8n image how do i do so

[aldifayol]

Step-by-step guide:

1. Create a Requirements File
   First, create a file named requirements.txt in the same directory as your new Dockerfile. This file should list all the Python packages you need.

requirements.txt Example:

requests
pandas
numpy==1.26.4

Note: Specify package versions for better reproducibility.

2. Create the Custom Dockerfile
   Create a file named Dockerfile (with a capital 'D'). This file defines the steps to build your custom image. You'll switch to the root user to install system packages and Python, then switch back to the non-root node user for security.

Dockerfile Example: Dockerfile

# Start from the official n8n image
FROM n8nio/n8n:latest

# Switch to root to install system packages
USER root

# Install Python 3, pip, and required build tools (for packages that need compilation, e.g., numpy)
# n8n's base image uses Alpine Linux, so we use 'apk'
RUN apk add --update --no-cache \
    python3 \
    py3-pip \
    build-base \
    python3-dev \
    libffi-dev \
    openssl-dev

# Install python packages from requirements.txt
# Copy the requirements file into the container
COPY requirements.txt /tmp/requirements.txt

# Install packages using pip3. Use the --break-system-packages flag to bypass system warnings/restrictions.
RUN python3 -m pip install --no-cache-dir --break-system-packages -r /tmp/requirements.txt \
    && rm /tmp/requirements.txt

# Switch back to the non-root 'node' user (n8n's default user)
USER node

3. Build the Custom Image
   In your terminal, navigate to the directory where you saved the Dockerfile and requirements.txt, and run the following command to build your new image. Replace my-n8n-python with a name you prefer.

docker build -t my-n8n-python .

4. Update Your Docker Compose File (or Run Command)
   If you're using docker-compose.yml to run n8n, you need to update it to use your newly created image.

Before (Example): YAML

# ...
services:
  n8n:
    image: n8nio/n8n:latest # <--- CHANGE THIS LINE
# ...
After (Example):

YAML

# ...
services:
  n8n:
    image: my-n8n-python # <--- USE YOUR CUSTOM IMAGE NAME
    # ... rest of your config

If you were running n8n with a simple docker run command, replace n8nio/n8n:latest with my-n8n-python.

5. Restart n8n
   After updating your configuration, stop your current n8n container and start the new one.

If using Docker Compose: Bash

docker-compose down
docker-compose up -d

If using Docker Run (stop and start):

Stop the old container:

docker stop <n8n-container-name>

Run the new container (using your new image):

docker run -d --name n8n ... my-n8n-python

Your n8n container is now running with all the Python packages from your requirements.txt installed and available for use in nodes like the Execute Command node (using python3) or the Code node (if you've configured a custom Python task runner).

[AryanMithbawkar]

followed steps but it still shows ModuleNotFoundError: No module named 'openpyxl'
which i added in requirements file

[devangtomar]

@AryanMithbawkar Sharing this here as a potential solution since it’s not mentioned in the thread:

You can let n8n run Docker commands by mounting the host Docker socket + making the Docker CLI available inside the n8n container. Example:

n8n:
  ...
  volumes:
    - /var/run/docker.sock:/var/run/docker.sock
    - /usr/local/bin/docker:/usr/local/bin/docker
  user: root   # or match the docker group GID

Then n8n’s Execute Command node can run docker ps, docker run, etc. See screenshot.

This works well for local/self-hosted setups, but obviously comes with the usual Docker-socket security caveats. Still, it’s a simple and effective way to let workflows manage containers directly.

Curious what others (especially @aldifayol) think about including this as an optional/advanced example.