Using MinIO to upload to a local S3 bucket in Django

Posted on Jan 10, 2021

Hi everyone!

Some weeks ago I was doing a demo to my teammates, and one of the things that was more suprising for them was that I was able to do S3 uploads locally using “MinIO”.

Let me set the stage:

Imagine you have a Django ImageField which uploads a picture to a AWS S3 bucket.

How do you setup your local development environment without using a “development” AWS S3 Bucket? For that, we use MinIO.

What is MinIO?

According to their GitHub README:

MinIO is a High Performance Object Storage released under Apache License v2.0. It is API compatible with Amazon S3 cloud storage service.

So MinIO its an object storage that uses the same API as S3, which means that we can use the same S3 compatible libraries in Python, like Boto3 and django-storages.

The setup

Here’s the docker-compose configuration for my django app:

version: "3"

      context: .
      - ./app:/app
      - 8000:8000
      - minio
    command: >
      sh -c "python migrate &&
             python runserver"      

    image: minio/minio
      - 9000:9000
      - MINIO_ACCESS_KEY=access-key
      - MINIO_SECRET_KEY=secret-key
    command: server /export

    image: minio/mc
      - minio
    entrypoint: >
      /bin/sh -c "
      apk add nc &&
      while ! nc -z minio 9000; do echo 'Wait minio to startup...' && sleep 0.1; done; sleep 5 &&
      /usr/bin/mc config host add myminio http://minio:9000 access-key secret-key;
      /usr/bin/mc mb myminio/my-local-bucket;
      /usr/bin/mc policy download myminio/my-local-bucket;
      exit 0;
  • app is my Django app. Nothing new here.
  • minio is the MinIO instance.
  • createbuckets is a quick instance that creates a new bucket on startup, that way we don’t need to create the bucket manually.

On my app, in

# S3 configuration

DEFAULT_FILE_STORAGE = "storages.backends.s3boto3.S3Boto3Storage"

AWS_ACCESS_KEY_ID = os.environ.get("AWS_ACCESS_KEY_ID", "access-key")
AWS_SECRET_ACCESS_KEY = os.environ.get("AWS_SECRET_ACCESS_KEY", "secret-key")
AWS_STORAGE_BUCKET_NAME = os.environ.get("AWS_STORAGE_BUCKET_NAME", "my-local-bucket")

    AWS_S3_ENDPOINT_URL = "http://minio:9000"

If we were in a production environment, the AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY and AWS_STORAGE_BUCKET_NAME would be read from the environmental variables, but since we haven’t set those up and we have DEBUG=True, we are going to use the default ones, which point directly to MinIO.

And that’s it! That’s everything you need to have your local S3 development environment.


First, let’s create our model. This is a simple mock model for testing purposes:

from django.db import models

class Person(models.Model):
    """This is a demo person model"""

    first_name = models.CharField(max_length=50)
    last_name = models.CharField(max_length=50)
    date_of_birth = models.DateField()
    picture = models.ImageField()

    def __str__(self):
        return f"{self.first_name} {self.last_name} {str(self.date_of_birth)}"

Then, in the Django admin we can interact with our new model:

If we go to the URL and change the domain to localhost, we should be able to see the picture we uploaded.

Bonus: The MinIO browser

MinIO has a local objects browser. If you want to check it out you just need to go to http://localhost:9000. With my docker-compose configuration, the credentials are:

username: access-key
password: secret-key

On the browser, you can see your uploads, delete them, add new ones, etc.


Now you can have a simple configuration for your local and production environments to work seamlessly, using local resources instead of remote resources that might generate costs for the development.

If you want to check out the project code, you can go to my git server here: or in Gitlab here:

See you in the next one!