I’m working with a legacy Rails application that requires the puma-dev
tool in order to develop locally. The app needs to be able to resolve domains like client-a.myapp.test
to localhost:3000.
The problem is this app is so old that me and my team need to docker-ize it in order to have any hope of standing it up locally, due to various outdated components.
The problem, continued, is that I can’t just plop puma-dev into my docker container because Linux support for the tool is extremely lacking and highly unclear how to translate the instructions to a docker environment. After struggling with that route for quite a while I decided to take another approach.
Right now I’ve created a separate docker image just for puma-dev from a golang image on dockerhub. I decided to try and run it in parallel with my other docker container, containing the ruby app, in hopes that they can work together (the ruby app serving on port 3000, with puma-dev listening), or with me ultimately listing it as a dependency in docker-compose for the ruby app and bundling them together. However, I’m not sure if that makes any sense at all to do, or if it’s at all a logically sound approach.
I guess my question is, is it possible for me to do it this way? Or, better yet, how can I get puma-dev working with Docker?