Based on feedback from this meta post I’ve created a basic Profile that should work for most any environment. This also pairs well with a poorly written upstart script that I hope to refine and eventually add to the repo.
I actually think it should probably be a Procfile.sample, cause ideally people want to be running multiple thins, any way you can comment it to explain how they would and add comment links to any relevant discussion.
@SamSaffron Unfortunately, there aren’t any ways to add comments ddollar/foreman#12 Furthermore, you control how many workers for each category with “a .procfile” file (or via the command line). So doing so you can adjust concurrency, etc. Would it be better to just have a Procfile.sample or a .procfile.sample (with comments since it’s yaml) or both?
I’ve left the
Procfile as is, upon further reading the idea is to extract away what commands are responsible for what services. In this respect if the project moved away from thin and used gunicorn or another server then changing the
web key in the Procfile will allow everyone to continue running without a problem and everything should just magic. As for concurrency for of tasks I’ve added a
.procfile.sample with glorious comments about how it works.
Looks good to me! Thanks!
A commit that appears in this pull request is being discussed here.