During cap deploy:cold - command not found for /etc/init.d/unicorn

Question

I'm very close to having my first rails app live up on Linode VPS, but keep on getting a strange error message near the end of cap deploy:cold. I've been following railscasts 335 to deploy my Rails app to a VPS using nginx, Unicorn, PostgreSQL, rbenv and more (unfortunately for me from a Windows machine). I'm hosting on Linode Ubuntu 10.04 LTS Profile. Near the end of the deploy I get this error message:

  * ←[32m2013-04-24 13:08:13 executing `deploy:start'←[0m
  * ←[33mexecuting "sudo -p 'sudo password: ' /etc/init.d/unicorn_wheretoski start"←[0m
    servers: ["xxx.xx.xxx.242"]
    [xxx.xx.xxx.242] executing command
 ** [out :: xxx.xx.xxx.242]
 ** [out :: xxx.xx.xxx.242] sudo: /etc/init.d/unicorn_wheretoski: command not found
 ** [out :: xxx.xx.xxx.242]
    ←[2;37mcommand finished in 309ms←[0m
failed: "env PATH=$HOME/.rbenv/shims:$HOME/.rbenv/bin:$PATH sh -c 'sudo -p '\\''
sudo password: '\\'' /etc/init.d/unicorn_wheretoski start'" on xxx.xx.xxx.242

When I go to the server, it locates the file

:~/apps/wheretoski/current$ ls /etc/init.d/unicorn_wheretoski
/etc/init.d/unicorn_wheretoski

From deploy.rb

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

And from unicorn_init.sh

#!/bin/sh
set -e

# Feel free to change any of the following variables for your app:
TIMEOUT=${TIMEOUT-60}
APP_ROOT=/home/deployer/apps/wheretoski/current
PID=$APP_ROOT/tmp/pids/unicorn.pid
CMD="cd $APP_ROOT; bundle exec unicorn -D -c $APP_ROOT/config/unicorn.rb -E production"
AS_USER=deployer
set -u

OLD_PIN="$PID.oldbin"

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

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

run () {
  if [ "$(id -un)" = "$AS_USER" ]; then
    eval $1
  else
    su -c "$1" - $AS_USER
  fi
}

case "$1" in
start)
  sig 0 && echo >&2 "Already running" && exit 0
  run "$CMD"
  ;;

I then head over to the VPS and try to execute the various commands and I get an error when executing the following:

deployer@li543-242:~/apps/wheretoski/current$ bundle exec unicorn -D -c $/home/apps/wheretoski/current/config/unicorn.rb -E production
/home/deployer/.rbenv/versions/1.9.3-p125/lib/ruby/gems/1.9.1/gems/bundler-1.3.5/lib/bundler/rubygems_integration.rb:214:in `block in replace_gem': unicorn is not part of the bundle. Add it to Gemfile. (Gem::LoadError)
        from /home/deployer/.rbenv/versions/1.9.3-p125/bin/unicorn:22:in `<main>'

Here is what I get for echo $PATH on the VPS: /home/deployer/.rbenv/shims:/home/deployer/.rbenv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/home/deployer/.rbenv/versions/1.9.3-p125/bin

I have tried with both the unicorn gem under the production group and as part of the main gems, both have produced this same error message. When I open the Gemfile.lock in the current folder on the server Unicorn only shows up under the dependencies, not under the specs.

Thanks for any help!!

Yes, definitely ran cap deploy:setup prior to running cap deploy:cold — Mitch Nick

Mitch Nick Mitch Nick · Accepted Answer · 2013-05-01T19:31:56

Alright, there was a couple of issues here.

1 - I had different versions of bundler on my local machine and the server.

2 - Developing on a Windows machine. I had to put the unicorn gem under a production group in my gemfile, and for whatever reason the gemfile.lock was not created successfully as a result. Had a buddy with a mac pull my code, move unicorn to the main section of the gemfile, and bundle installed it. This created a good Gemfile.lock which is in use now on the server.

Not sure if this will be helpful to others or not, quite the weird error.

During cap deploy:cold - command not found for /etc/init.d/unicorn

1 Answers