Running Python applications on Nodion
Learn more about how to run Python applications like Flask or Django on Nodion
Create a new application
To get started you can simply create a new application after signing in. If you simply want to try out our platform you can use one of our starter templates to get up and running quickly.
Deploying with Buildpacks
Buildpacks will automatically detect that you are running a Python application if you have a
Pipfile file in the root directory of your application. Please provide the desired Python version in a
runtime.txt file, the Python Buildpack will install all dependencies as well as the specified Python version.
It's necessary to set a
PORT Environment Variable within your application settings. If this is not set, our load balancing system is not able to redirect to your application and an error will occur.
If you need to run certain tasks when deploying you can use the release stage of the Procfile. The command provided there will be run right before spawning the new containers of that deployment.
web: gunicorn app.wsgi release: python manage.py migrateProcfile
Celery is well known and a popular library to handle background processing tasks. With Nodion you can simply provide a worker start command with the
worker type to run Celery. For example:
web: gunicorn app.wsgi worker: python -m celery -A django_celery worker release: bundle exec rails db:migrateProcfile
Celery requires Redis to work. You can use our Managed Redis service to launch Redis instances.
We support multiple Buildpack Builders, depending on which Builder you chose there are different packages pre-installed. You can use additional Buildpacks, like the FFmpeg Buildpack to install a specific package. In general, all Builders that are supported by Nodion are based on Ubuntu. That means you can also install any package by adding a Buildpack called apt and creating an
Aptfile to provide a list of packages.