Configuring gunicorn workers on App Platform with Django

Hi there,

I believe that when using the DigitalOcean’s App Platform with Django, it automatically uses gunicorn as the application server if you haven’t specified a custom start command. However, you have full control over the configuration by using the “Run Command” setting in the App Platform.

To set the number of workers you can update the “Run Command” setting. Here, you can specify your custom gunicorn command:

gunicorn --worker-tmp-dir /dev/shm --workers 4 django_app.wsgi

You can adjust the --workers number as required.

If you are still getting the error that gunicorn is not found, could you share your requirements.txt file here?

Regarding your database connections issue, in your Django settings, you can use persistent database connections. This can help reduce the overhead of establishing a new connection every time one is requested:

DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'NAME': 'mydatabase',
        ...
        'CONN_MAX_AGE': 300,
    }
}

The CONN_MAX_AGE setting specifies the lifetime of a database connection in seconds. Setting it to a positive value like 300 (5 minutes) allows Django to use a persistent connection for that duration.

Also, if you’re experiencing high connection loads, consider using a connection pooler like pgbouncer:

https://docs.digitalocean.com/products/databases/postgresql/how-to/manage-connection-pools/

Also keep in mind that the number of gunicorn workers is not directly related to the number of database connections. However, each worker can handle a request, and if each request results in a new database connection, they can quickly add up, especially if you’re not using persistent connections or a connection pooler.

Let me know how it goes!

Best,

Bobby