Question

I own a Rails application with Unicorn as Web server.

I deploy it by means of Capistrano.

Here my deploy.rb file:

require "bundler/capistrano"

server "91.121.11.100", :web, :app, :db, primary: true

set :application, "myapp"
set :user, "deployer"
set :deploy_to, "/home/#{user}/apps/#{application}"
#set :deploy_via, :remote_cache
set :use_sudo, false

set :scm, "git"
set :repository, "[e-mail]:therepository/#{application}.git"
set :branch, "master"

default_run_options[:pty] = true
ssh_options[:forward_agent] = true

#after "deploy", "deploy:cleanup" # keep only the last 5 releases

namespace :deploy do
  %w[start stop restart].each do |command|
    desc "#{command} unicorn server"
    task command, roles: :app, except: {no_release: true} do
      run "/etc/init.d/unicorn_#{application} #{command}"
    end
  end

  task :setup_config, roles: :app do
    sudo "ln -nfs #{current_path}/config/nginx.conf /etc/nginx/sites-enabled/#{application}"
    sudo "ln -nfs #{current_path}/config/unicorn_init.sh /etc/init.d/unicorn_#{application}"
    run "mkdir -p #{shared_path}/config"
    put File.read("config/database.yml"), "#{shared_path}/config/database.yml"
    puts "Now edit the config files in #{shared_path}."
  end
  after "deploy:setup", "deploy:setup_config"

  task :symlink_config, roles: :app do
    run "ln -nfs #{shared_path}/config/database.yml #{release_path}/config/database.yml"
  end
  after "deploy:finalize_update", "deploy:symlink_config"

  desc "Make sure local git is in sync with remote."
  task :check_revision, roles: :web do
    unless `git rev-parse HEAD` == `git rev-parse origin/master`
      puts "WARNING: HEAD is not the same as origin/master"
      puts "Run `git push` to sync changes."
      exit
    end
  end
  before "deploy", "deploy:check_revision"
end

Deployment well happens and current folder on server contains the updated files as expected.

But, something very strange that I don’t understand happens:

I’ve got this line in the beginning of my process:

logger = Logger.new "#{Rails.root}/log/web_agents.log"

and this error still appears:

No such file or directory - /home/deployer/apps/myapp/releases/20120612122610/log/web_agents.log

Why 20120612122610??? It’s an old release that I even deleted.

Why don’t Unicorn point to the last release ?

For testing, I even replace Rails.root by the hardcoded path to current one.

Still got the same error… I’ve killed, stop, force-stop Unicorn…doesn’t matter…

Any idea ?

I precise that I’m sure that process are launch using last updated files in the ‘current’ folder since when I delete one, process can’t work and many errors appear.

UPDATED

Here my config/unicorn.rb file:

root = "/home/deployer/apps/myapp/current"
working_directory root
pid "#{root}/tmp/pids/unicorn.pid"
stderr_path "#{root}/log/unicorn.log"
stdout_path "#{root}/log/unicorn.log"

listen "/tmp/unicorn.myapp.sock"
worker_processes 2
timeout 30

Answer 1

I found myself where the problem comes from.

In fact, there is a troubleshooting between Unicorn and Sidekiq.

Indeed, as I said in one comment above, process is launched by Sidekiq.

First, within deploy.rb, there must be this line:

require 'sidekiq/capistrano' 

This allows deployment process to restart Sidekiq gracefully.
See there:
https://github.com/mperham/sidekiq/wiki/Deployment

Secondly, within unicorn.rb, there must be this kind of block:

after_fork do |server, worker|
  Sidekiq.configure_client do |config|
    config.redis = { :size => 1 }
  end
end

For more information about it, see there:

https://github.com/mperham/sidekiq/wiki/Problems-and-Troubleshooting

And now, no more weird issue 🙂

A potential explanation:

Probably Sidekiq manages some caches that prevent it to be based on the last release…and that would be why a restart plus cleanly launches of Unicorn and Sidekiq were enough.