Setting up a rails app on puma, rails, nginx everything is running, but nginx sends error

问题

I am trying to run a rails app. I followed these instructions: http://luugiathuy.com/2014/11/setup-nginx-puma-on-ubuntu/ to set up a server.

But, when I got to the end, specifically: puma -e production -d -b unix:///tmp/app_name.sock --pidfile /tmp/puma.pid Upon running it I got the response:

Puma starting in single mode...
* Version 2.16.0 (ruby 2.1.7-p400), codename: Midwinter Nights Trance
* Min threads: 0, max threads: 16
* Environment: production
* Daemonizing...

The external IP just kept returning the nginx error:

We're sorry, but something went wrong.

If you are the application owner check the logs for more information.

So, looking elsewhere, I tried: RACK_ENV=production bundle exec puma -p 3000

And, it seemed that this did more, I got:

Puma starting in single mode...
* Version 2.15.3 (ruby 2.1.7-p400), codename: Autumn Arbor Airbrush
* Min threads: 0, max threads: 16
* Environment: production
Rails Error: Unable to access log file. Please ensure that /home/myusername/myappname/log/production.log exists and is writable (ie, make it writable for user and group: chmod 0664 /home/myusername/myappname/log/production.log). The log level has bee
n raised to WARN and the output directed to STDERR until the problem is fixed.
** [Bugsnag] Bugsnag exception handler 3.0.0 ready, api_key=#####################
* Listening on tcp://0.0.0.0:3000
Use Ctrl-C to stop

But the external web page returned the same nginx error.

How should I proceed? It seems that things are running but puma and nginx just are not talking?

EDIT 1

I did:  `sudo chmod -R 0777 /home/myunsername/appname/log/`

Then:

$RACK_ENV=production bundle exec puma -p 3000

And, now the output is below and the nginx response at the external IP remains the same.

Puma starting in single mode...
* Version 2.15.3 (ruby 2.1.7-p400), codename: Autumn Arbor Airbrush
* Min threads: 0, max threads: 16
* Environment: production
* Listening on tcp://0.0.0.0:3000
Use Ctrl-C to stop

Now, both methods: RACK_ENV=production bundle exec puma -p 3000 and puma -e production -d -b unix:///tmp/web-app.sock --pidfile /tmp/puma.pid create puma processes that don't seem to communicate with nginx but do not return any errors.

EDIT 2 Checking Logs

I ran:

cat log/production.log

and it returned:

I, [2016-02-02T##:##:##.###### #28802] INFO -- : ** [Bugsnag] Bugsnag exception handler 3.0.0 ready, api_key=####################

EDIT 2.5 Checking More Logs I ran:

tail -f /var/log/nginx/error.log

and it returned:

2016/02/02 11:09:52 [emerg] 28220#0: unknown directive "tream" in /etc/nginx/sites-enabled/web-appname.com:1 2016/02/02 11:10:26 [emerg] 28273#0: unknown directive "tream" in /etc/nginx/sites-enabled/web-appname.com:1

Those errors were caused a bit ago from my file missing the first 3 characters, that error has since been fixed. In conclusion: there seem to be no errors in the logs that I have listed above that give me any hints.

EDIT 3

My guess currently is that I set up my /etc/nginx/sites-available/myapp.com file incorrectly. Here is all the naming that I have used.

Currently it is named: web-app.com, the url is just the ip address, lets say https://104.199.155.166/. My application is in a folder called web-app

upstream web-app {
  server unix:///tmp/web-app.sock;
}

server {
  listen 80;
  server_name web-app; # change to match your URL
  root /home/myusername/web-app/public; # change to match your rails app public folder

  location / {
    proxy_pass http://web-app; # match the name of upstream directive which is defined in line 1
    proxy_set_header Host $host;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_set_header X-Real-IP $remote_addr;
  }

  location ~* ^/assets/ {
    # Per RFC2616 - 1 year maximum expiry
    expires 1y;
    add_header Cache-Control public;

    # Some browsers still send conditional-GET requests if there's a
    # Last-Modified header or an ETag header even if they haven't
    # reached the expiry date sent in the Expires header.
    add_header Last-Modified "";
    add_header ETag "";
    break;
  }
}

EDIT 4

I have made many changes to the above config file, and nothing that I do changes anything. I currently have it at what I believe to be the correct configuration. I am at a loss and would appreciate any suggestions.

Small question: if there are other users on the same VM and they may have nginx installed and configure differently, could that be causing the issues?

回答1:

StackOverflow was being too annoying about "code not properly idented", so I decided to put all the answer as code.

You have to create config/puma.rb manually. I recommend reading DigitalOcean's article: 
["How To Deploy a Rails App with Puma and Nginx on Ubuntu 14.04"][1]
I'll paste the puma.rb from the article:
    # config/puma.rb
    # Change to match your CPU core count
    workers 2

    # Min and Max threads per worker
    threads 1, 6

    app_dir = File.expand_path("../..", __FILE__)
    shared_dir = "#{app_dir}/shared"

    # Default to production
    rails_env = ENV['RAILS_ENV'] || "production"
    environment rails_env

    # Set up socket location
    # bind "unix://#{shared_dir}/sockets/puma.sock"
    # in your case
    bind "unix:///tmp/web-app.sock"

    # Logging
    stdout_redirect "#{shared_dir}/log/puma.stdout.log", "#shared_dir}/log/puma.stderr.log", true

    # Set master PID and state locations
    pidfile "#{shared_dir}/pids/puma.pid"
    state_path "#{shared_dir}/pids/puma.state"
    activate_control_app

    on_worker_boot do
      require "active_record"
      ActiveRecord::Base.connection.disconnect! rescue ActiveRecord::ConnectionNotEstablished
      ActiveRecord::Base.establish_connection(YAML.load_file("#{app_dir}/config/database.yml")[rails_env])
    end

Detail: after setting bind [...] you can just execute 

    $ bundle exec puma RAILS_ENV=production

Make sure that the directory shared exists and, within it, folders log and pids. Chmod them, otherwise, You'll face permission issues.

Another thing: can you show your nginx.conf file? If the user set in nginx.conf has no "rights" to read your app, You'll (also) face permission issues.

Tip: If you have problems with the empty folders from shared (log and pids) and git, put a .keep (empty file) inside both of them.

Another tip: you can set server_name (nginx directive) as _ (yes, an underscore) or localhost if you do not have a domain.

Almost forget: check if your vm has http and https traffic enabled.

  [1]: https://www.digitalocean.com/community/tutorials/how-to-deploy-a-rails-app-with-puma-and-nginx-on-ubuntu-14-04

来源：https://stackoverflow.com/questions/35152369/setting-up-a-rails-app-on-puma-rails-nginx-everything-is-running-but-nginx-se