CODEWITHSUNDEEP
ruby-on-rails-4deploymentcapistranounicorn
lbar
lbar

Reputation: 63

Unicorn doesn't pick up changes with new deploy of Rails app (Capistrano, Nginx)

I'm new to deploying, so this is probably a rookie mistake, but here it goes.

I have a Rails 4 app that I'm deploying to a Linux server using a combination of Capistrano, Unicorn, and Nginx. The deploy script runs fine and the app is now reachable at the desired IP, so that's great. The thing is, a) Unicorn doesn't restart upon deployment (at least, the PIDs don't change) and b) not surprisingly, the new changes aren't reflected in the available app. I don't seem to be able to do anything other than completely stopping and restarting unicorn in order to refresh it. If I do this, then the changes are picked up, but this process is obviously not ideal.

Manually, if I run kill -s HUP $UNICORN_PID then the pids of the workers change but not the master, and changes aren't picked up (which, apparently they are supposed to be); using USR2 appears to have no effect on the current processes.

Here's the unicorn init script I'm using, based on suggestions from other stack overflow questions with similar problems:

set -e

USAGE="Usage: $0 <start|stop|restart|upgrade|rotate|force-stop>"

# app settings
USER="deploy"
APP_NAME="app_name"
APP_ROOT="/path/to/$APP_NAME"
ENV="production"

# environment settings
PATH="/home/$USER/.rbenv/shims:/home/$USER/.rbenv/bin:$PATH"
CMD="cd $APP_ROOT/current && bundle exec unicorn -c config/unicorn.rb -E $ENV -D"
PID="$APP_ROOT/shared/pids/unicorn.pid"
OLD_PID="$PID.oldbin"
TIMEOUT=${TIMEOUT-60}

# make sure the app exists
cd $APP_ROOT || exit 1

sig () {
  test -s "$PID" && kill -$1 `cat $PID`
}

oldsig () {
  test -s $OLD_PID && kill -$1 `cat $OLD_PID`
}

case $1 in
  start)
    sig 0 && echo >&2 "Already running" && exit 0
    echo "Starting $APP_NAME"
    su - $USER -c "$CMD"
    ;;
  stop)
    echo "Stopping $APP_NAME"
    sig QUIT && exit 0
    echo >&2 "Not running"
    ;;
  force-stop)
    echo "Force stopping $APP_NAME"
    sig TERM && exit 0
    echo >&2 "Not running"
    ;;
  restart|reload)
    sig HUP && echo "reloaded $APP_NAME" && exit 0
    echo >&2 "Couldn't reload, starting '$CMD' instead"
    run "$CMD"
    ;;
  upgrade)
    if sig USR2 && sleep 2 && sig 0 && oldsig QUIT
    then
      n=$TIMEOUT
      while test -s $OLD_PID && test $n -ge 0
      do
        printf '.' && sleep 1 && n=$(( $n - 1 ))
      done
      echo

    if test $n -lt 0 && test -s $OLD_PID
    then
      echo >&2 "$OLD_PID still exists after $TIMEOUT seconds"
      exit 1
    fi
    exit 0
  fi
  echo >&2 "Couldn't upgrade, starting '$CMD' instead"
  su - $USER -c "$CMD"
  ;;
  rotate)
    sig USR1 && echo rotated logs OK && exit 0
    echo >&2 "Couldn't rotate logs" && exit 1
    ;;
  *)
    echo >&2 $USAGE
    exit 1
    ;;
esac

Using this script, start and stop work as expected, but reload/restart do nothing (they print the expected output but don't change the running pids) and upgrade fails. According to the error log, it's because the first master is still running (ArgumentError: Already running on PID: $PID).

And here's my unicorn.rb:

app_path = File.expand_path("../..", __FILE__)
working_directory "#{app_path}"
pid               "#{app_path}/../../shared/pids/unicorn.pid"

# listen
listen "#{app_path}/../../shared/sockets/unicorn.sock", :backlog => 64

# logging
stderr_path "#{app_path}/../../shared/log/unicorn.stderr.log"
stdout_path "#{app_path}/../../shared/log/unicorn.stdout.log"

# workers
worker_processes 3

# use correct Gemfile on restarts
before_exec do |server|
  ENV['BUNDLE_GEMFILE'] = "#{working_directory}/Gemfile"
end

# preload
preload_app false

before_fork do |server, worker|
  old_pid = "#{app_path}/shared/pids/unicorn.pid.oldbin"
  if File.exists?(old_pid) && server.pid != old_pid
    begin
      Process.kill("QUIT", File.read(old_pid).to_i)
    rescue Errno::ENOENT, Errno::ESRCH
      # someone else did our job for us
    end
  end
end

after_fork do |server, worker|
  if defined?(ActiveRecord::Base)
    ActiveRecord::Base.establish_connection
  end
end

Any help is very much appreciated, thanks!

Upvotes: 1

Views: 1271

Answers (2)

Dave Morse
Dave Morse

Reputation: 757

The line

ENV="production"

looks extremely suspicious to me. I suspect that it wants to be

RAILS_ENV="production".

without this won't rails wake up not knowing which environment it is?

Upvotes: 0

Matt Brictson
Matt Brictson

Reputation: 11082

It is hard to say for certain, since I haven't encountered this particular issue before, but my hunch is that this is your problem:

app_path = File.expand_path("../..", __FILE__)
working_directory "#{app_path}"

Every time you deploy, Capistrano creates a new directory for your app at the location releases/<timestamp>. It then updates a current symlink to point at this latest release directory.

In your case, you may mistakenly be telling Unicorn to use releases/<timestamp> as its working_directory. (SSH to the server and check the contents of unicorn.rb to be certain.) Instead, what you should do is point to current. That way you don't have to stop and cold start unicorn to get it to see the new working directory.

# Since "current" is a symlink to the current release,
# Unicorn will always see the latest code.
working_directory "/var/www/my-app/current"

I suggest rewriting your unicorn.rb so that you aren't using relative paths. Instead hard-code the absolute paths to current and shared. It is OK to do this because those paths will remain the same for every release.

Upvotes: 1

Related Questions