This is a basic hello world Django example for Google App Engine Managed VMs.
You can run locally using django's manage.py:
$ python manage.py runserver
Follow the standard deployment instructions in
the top-level README. Google App Engine runs the application
using gunicorn as defined by entrypoint in
app.yaml. You can use a different WSGI container if you want, as
long as it listens for web traffic on port $PORT and is declared in
requirements.txt.
This project was created using standard Django commands:
$ virtualenv env
$ source env/bin/activate
$ pip install django gunicorn
$ pip freeze > requirements.txt
$ django-admin startproject project_name
$ python manage.py startapp helloworld
Then, we added a simple view in hellworld.views, added the app to
project_name.settings.INSTALLED_APPS, and finally added a URL rule to
project_name.urls.
In order to deploy to Google App Engine, we created a simple
app.yaml.
This sample project uses Django's default sqlite database. This isn't suitable for production as your application can run multiple instances and each will have a different sqlite database. Additionally, instance disks are emphemmeral, so data will not survive restarts.
For production applications running on Google Cloud Platform, you have the following options:
- Use Cloud SQL, a fully-managed MySQL database. There is a Flask CloudSQL sample that should be straightfoward to adapt to Django.
- Use any database of your choice hosted on Google Compute Engine. The Cloud Launcher can be used to easily deploy common databases.
- Use third-party database services, or services hosted by other providers, provided you have configured access.