Sharing docker containers with other team members

Probably isn’t 100% related to cookiecutter-django specifically but i just followed: https://realpython.com/blog/python/development-and-deployment-of-cookiecutter-django-via-docker/

Which went real well and am thinking of using this as my base for future projects, just 1 question though. Say if i’m working on a development team, what is the ideal workflow for sharing dev & production containers with other members of the team?

[burhan]

A docker container is a running instance that is based from a docker image. You can think of a docker image as a “read only” version of a docker instance.

A dockerfile builds a docker image that is pre-configured, then you run it (and it becomes an image), Since docker images are read-only, generally they are packed with only the core files that are required for that docker to run - they don’t contain any runtime files (like say logs or temporary data), and in the case of application servers - like say Apache or nginx, it doesn’t contain the HTML documents or content.

For us, we are providing docker containers that are pre-configured to run django (along with all the dependencies). The dockers are linked with your workstation so that when the image is running, it has access to your django project.

You would normally share just the dockerfile, which contains information on how to create the server, and how to link it with the source.

The dockerfile is configured so that when the image runs, it looks for content at a specific path, which is configured in the dockerfile to be mapped to the source code. This source code is often subject to version control.

Anyone with the dockerfile can run the exact same application on their laptops; if multiple people need to share access to one instance - then you can run the docker image on a third party server (like a droplet), then to update the code - just do git pull on the droplet.

You can also - and many people do this - build an image that includes all your data and source code; this way all the other person has to do is download the image and run it. This makes changes difficult as after each change, you have to rebuild the docker image to “save” them.

Hope this helps.

[jayfk]

I think this can be closed now, excellent explanation @burhan.