Multiple users accessing Marimo

[jacksondelametter]

I am working on a project where we want to deploy Marimo and allow multiple users be able to access it from a cloud environment. If I create a new Python environment for each user as well as start a Marimo instances using marimo edit --port [port-number] for each user to individually use, then each user can still see notebooks that other users are working on even though the Marimo instances are running in a separate Python environment and bound to a different port number. Is there a way to have multiple Marimo instances completely isolated from one another?

[akshayka]

@jacksondelametter , thanks for the question.

Are they all running on the same server/same container, or in independent containers? I think you could isolate the marimo edit servers entirely by running in different Docker containers. Would that be possible?

each user can still see notebooks that other users are working on even though the Marimo instances are running in a separate Python environment and bound to a different port number

If they are running in the same container, I would guess it's it the recent files of other users that each of your users can see? Or what are they seeing?

[mscolnick]

@jacksondelametter you might find that you can achieve this with deploying marimo server's with JupyterHub

checking out this thread (near the bottom): #905
and this repo: https://github.com/jyio/jupyter-marimo-proxy

[dridk]

What about molab ?

[mikehogarth]

I would second the high value of molab as part of the Marimo offerings. I would go so far as to say I would pay to license molab even if Marimo itself is FOSS.