How to get fix 'require' errors when setting up delayed_job on Elastic Beanstalk

Question

I'm having trouble getting delayed_jobs running on Elastic Beanstalk. I'm using the 64bit Amazon Linux 2014.03 v1.0.0 running Ruby 2.1 (Passenger Standalone) container.

This is my config script (delayed_job.config) ...

files:
  "/opt/elasticbeanstalk/hooks/appdeploy/post/99_restart_delayed_job.sh":
    mode: "000755"
    owner: root
    group: root
    encoding: plain
    content: |
      #!/usr/bin/env bash
      . /opt/elasticbeanstalk/support/envvars
      cd $EB_CONFIG_APP_CURRENT
      su -c "RAILS_ENV=production bin/delayed_job --pid-dir=$EB_CONFIG_APP_SUPPORT/pids restart" $EB_CONFIG_APP_USER

The 99_restart_delayed_job.sh script exists and runs ... but then I stumble into this error.

2014-10-02 15:28:32,332 [INFO] (17387 MainThread) [directoryHooksExecutor.py-29] [root directoryHooksExecutor info] Script succeeded.
2014-10-02 15:28:32,402 [INFO] (17448 MainThread) [directoryHooksExecutor.py-29] [root directoryHooksExecutor info] Executing directory: /opt/elasticbeanstalk/hooks/appdeploy/post/
2014-10-02 15:28:32,402 [INFO] (17448 MainThread) [directoryHooksExecutor.py-29] [root directoryHooksExecutor info] Executing script: /opt/elasticbeanstalk/hooks/appdeploy/post/99_restart_delayed_job.sh
/usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require': cannot load such file -- bundler/setup (LoadError)
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /var/app/current/config/boot.rb:4:in `<top (required)>'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /var/app/current/config/application.rb:1:in `<top (required)>'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /var/app/current/config/environment.rb:2:in `<top (required)>'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from /usr/share/ruby/vendor_ruby/2.0/rubygems/core_ext/kernel_require.rb:55:in `require'
    from bin/delayed_job:3:in `<main>'

2014-10-02 15:28:32,440 [ERROR] (17448 MainThread) [directoryHooksExecutor.py-33] [root directoryHooksExecutor error] Script /opt/elasticbeanstalk/hooks/appdeploy/post/99_restart_delayed_job.sh failed with returncode 1

I've already poured through this other thread on SO that showed me how to set up. My problem is that I don't know what's preventing the script from running without error.

If I SSH into the EC2 instance, I'm able to run this without error ...

RAILS_ENV=production bin/delayed_job --pid-dir=$EB_CONFIG_APP_SUPPORT/pids restart

While this asks me for a password ...

su -c "RAILS_ENV=production bin/delayed_job --pid-dir=$EB_CONFIG_APP_SUPPORT/pids restart" $EB_CONFIG_APP_USER

Which I can avoid by doing this ...

sudo su -c "RAILS_ENV=production bin/delayed_job --pid-dir=$EB_CONFIG_APP_SUPPORT/pids restart" $EB_CONFIG_APP_USER

See: 'How to automatically restart delayed_job when deploying a rails project on Amazon Elastic Beanstalk?'

Update 1: 2014-10-15

After applying the -l option with the change of directory passed in, I get this error ...

2014-10-15 06:17:28,673 [INFO] (4417 MainThread) [directoryHooksExecutor.py-29] [root directoryHooksExecutor info] Executing script: /opt/elasticbeanstalk/hooks/appdeploy/post/99_restart_delayed_job.sh
2014-10-15 06:17:30,374 [INFO] (4417 MainThread) [directoryHooksExecutor.py-29] [root directoryHooksExecutor info] Output from script: /opt/rubies/ruby-2.1.2/lib/ruby/gems/2.1.0/gems/daemons-1.1.9/lib/daemons/application.rb:393:in `kill': Operation not permitted (Errno::EPERM)
    from /opt/rubies/ruby-2.1.2/lib/ruby/gems/2.1.0/gems/daemons-1.1.9/lib/daemons/application.rb:393:in `stop'
    from /opt/rubies/ruby-2.1.2/lib/ruby/gems/2.1.0/gems/daemons-1.1.9/lib/daemons/application_group.rb:171:in `block (2 levels) in stop_all'

2014-10-15 06:17:30,374 [ERROR] (4417 MainThread) [directoryHooksExecutor.py-33] [root directoryHooksExecutor error] Script /opt/elasticbeanstalk/hooks/appdeploy/post/99_restart_delayed_job.sh failed with returncode 1

Update 2: 2014-10-15

Turns out the error above was caused by an existing pid created by root (while debugging I had started delayed_job manually) so the c2-user couldn't restart/kill it hence the error.

Answer 1

The problem, from what I can tell, is that the environment/path variables aren't being established when switching to the $EB_CONFIG_APP_USER linux user. I made three changes:

1. Add the -l option to the su command to simulate a full login of $EB_CONFIG_APP_USER.
2. As a side effect of the -l option, the change directory command must be brought into the -c option.
3. As a good measure, but maybe not necessary, include bundle exec to ensure that the proper gems are being used.

Here's my functioning content: area:

#!/usr/bin/env bash
. /opt/elasticbeanstalk/support/envvars
su -l -c "cd $EB_CONFIG_APP_CURRENT && RAILS_ENV=production bundle exec bin/delayed_job --pid-dir=$EB_CONFIG_APP_SUPPORT/pids restart" $EB_CONFIG_APP_USER