Get help from the marimo community

Updated 7 months ago

Best practices for deployment?

At a glance

The community member is looking for guidance on deploying their marimo dashboards, which they have created for their team but want to keep private. They are considering hosting the dashboards on a single server, perhaps using Docker and nginx. The community members discuss that while marimo is not special, it would be helpful to have more comprehensive deployment documentation. Some suggestions include using Docker, deploying on fly.io, and exploring tools like ShinyProxy that can help with deploying dockerized apps. The community members express interest in contributing to the marimo deployment documentation and are open to learning from each other's experiences.

Useful resources
I have seen a few threads discussing deployment of marimo apps, e.g. on AWS. Are there complete examples that I could learn from? Any pointers would be appreciated!

For context: I have created multiple marimo dashboards, and would like to share them with the rest of my team. (They are currently private, e.g. I don't want to share them with the rest of the world.) They are not resource intensive, and I could easily host them on a single server (perhaps dockerized, and routed via nginx?). A starting page that lists the available apps would be useful, too.

Maybe others have solved this problem already? (Or perhaps something akin to Shiny Server in the R world is already in the works somewhere?)
A
M
s
7 comments
We don’t have a comprehensive doc on deployment yet, but as people find what works for them we’d happily accept contributions to our own docs: https://docs.marimo.io/guides/deploying/index.html
marimo is not really special, so whatever people do for other apps that use a mix of http/websockets should work for marimo. Still I understand the value in having instructions
Dockerized is the best and recommended route
I prefer fly.io for deploying docker containers
Thanks a lot for inviting contributions! I understand that marimo can be deployed like other apps. But I also suspect that marimo is / will be highly valuable to users without a strong software engineering background (myself included). So a little help from other marimo enthusiasts will be much appreciated from us data science types!
One option that might offer a low barrier is https://www.shinyproxy.io/ Fully open source, and with lots of examples. (It enabled me to deploy shiny apps successfully in the past - but now supports virtually any app that has been dockerized.) I will work out an example with marimo and report back!
Awesome, thanks! Curious to hear how it goes
Add a reply
Sign up and join the conversation on Discord