[RUBY] Puma --Nignx is an escape route when rails s -e production -d doesn't work in the environment

Premise

Deployed to EC2 with Rails6 + Puma + Nginx And at the terminal SSH connected to EC2

rails s -e production 

Then the service is displayed without any problem

rails s -e production -d

Then the error screen is displayed?

Yeah ~ I'm still connected to SSH to keep the service running

It's for you who have troubles!

Solution

In config / puma.rb in the Rails root folder

config/puma.rb

daemonize

I will add it! with this

rails s -e production -d

I can't start it

rails s -e production 

This will make it a daemon, so it's a solution!

People who think that this is inconvenient even locally as a daemon

config/puma.rb

if Rails.env == 'production'
  daemonize
end

Let's write! It's a workaround that anyone can think of! I'm sorry to write it as I knew it!

Cause

I haven't completely investigated it, but as a result of becoming a net surfer and diving in the sea of google, it seems that puma.rb is not loaded when it is daemonized with "rails s -d" in puma. When I started with rails s -e production -d, I could not connect to the Nginx error log with the socket set in puma.rb! Because it is described, I think that it is probably correct.

This guy is bad! Please point out any points!