Related

Tool NGINX Config Generator Tool
Unable to connect on deploying a rails app with Nginx, Capistrano and Puma on Ubuntu 16.04 Question Question
Subdomain not being served by Nginx Question Question

Question

Rails one-click (16.04) problem with unicorn.sock not found

Posted June 15, 2017 3.7k views
Ruby on RailsNginxUbuntu 16.04

I’m trying to setup my Rails app onto the new one-click install using Ubuntu 16.04. I was trying to set Let’s encrypt when I realize the server did not work and the nginx log show me this:

2017/06/15 10:49:17 [crit] 3304#3304: *1 connect() to unix:/run/DigitalOceanOneClick/unicorn.sock failed (2: No such file or directory) while connecting to upstream, client: x.x.x.x, server: staging-api.batiment.nc, request: "GET /favicon.ico HTTP/1.1", upstream: "http://unix:/run/DigitalOceanOneClick/unicorn.sock:/favicon.ico", host: "staging-api.batiment.nc", referrer: "http://staging-api.batiment.nc/"

Checking at the /run folder, it seems that DigitalOceanOneClick directory is erased somehow. Whenever I create it, something erase it.

Here are the thing I did on the machine (the rails_project demo page was working fine before):

  • Use rvm to change Ruby version (was 2.4.0 but needed to change to 2.2.5)
  • Install my app and its dependencies (every worked out)
  • Point unicorn to the right directory (the new app)
  • Configure nginx to point to the right directory and start let’s encrypt but then it failed to find well-known... public URL

Anyone of you experiencing that?

Add a comment
lkartono
By:
lkartono
Add a comment

These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.

×
7 answers
didierbanlock June 17, 2017

Hello,
i have the same issue. it’s very strange

Reply Report
lkartono June 19, 2017

With the previous one-click setup for Rails (on Ubuntu 14.04), it was much easier to manage Ruby and its Gems. Right now, as soon as I’m trying to change Ruby version and restart unicorn and nginx, the same error appear, the server crash and I cannot seem to be able to make it work again.
Any member of the team will be available to give some insight or help? or should I open a priority ticket with the support team?

Reply Report
mtaqweemsultan August 4, 2017

I got same issue. Have you able to resolve this issue ?

Reply Report
iumersheikh September 9, 2017

same issue with me.

Reply Report
espensommer September 28, 2017

Same issue.

Reply Report
espensommer September 30, 2017

For me the solution was to bundle install –deployment so the unicorn server is installed in the vendor folder of the new rails project.
And then you need to point the paths in the .unicorn.sh file to the correct place. In my case there where to extra folders added to the path. /ruby/2.4.0 was created inside vendor/bundle so in .unicorn.sh change to:
export GEM_HOME=“/home/rails/project/vendor/bundle/ruby/2.4.0” and so on for the other paths too.

Hope it helps!

Reply Report
pekkle April 5, 2018

I’m running into the same issue too.

2018/04/05 23:29:19 [crit] 1501#1501: *15 connect() to unix:/run/DigitalOceanOneClick/unicorn.sock failed (2: No such file or directory) while connecting to upstream, client: X.X.X.X, server: _, request: “GET / HTTP/1.1”, upstream: “http://unix:/run/DigitalOceanOneClick/unicorn.sock:/”, host: “192.241.169.48”

Reply Report
Submit an Answer

Related

Looking for something else?

Ask a new question Search for more help