Unicorn + RUnit Rails Not Killing Old Master

[Graham Christensen <graham@grahamc.com>]

Hi,

I'm deploying Unicorn on a Rails application with RUnit. Technically 
I'm using Chef's deployment tools, if any of you are familiar with it 
(https://github.com/opscode-cookbooks/application_ruby) but to be clear 
they aren't doing anything magical, so this is purely an issue with 
RUnit and Unicorn.

The TL;DR of the following post, which has lots of pastes (linking to 
pastebins for easier reading:) when I run /etc/init.d/myzippykid 
restart, the old unicorn master fails to die, causing the new master to 
be stuck in a loop waiting to own port 8080. The new master never takes 
over, as the old master never dies. When I manually kill the master, 
the new one takes over just fine.

What am I missing in my restart process to make the old master exit?

--

More details:

When I deploy a new version of software, the unicorn master fails to 
restart. Here is an example of my ps auxfg output after I run 
`/etc/init.d/myzippykid restart` (which is symlinked to /usr/bin/sv): 
https://gist.github.com/grahamc/35715c1a2c9717e461e3

When I run `killall unicorn`, the ruby script under runsvdir gets 
replaced with the unicorn worker. This is because the ruby script is 
attempting to bind to port 8080, which is already taken up by the 
unicorn master process. As soon as the old master is out of the way, 
its capable of binding.

Here is my unicorn configuration at /etc/unicorn/myzippykid.rb: 
https://gist.github.com/grahamc/08aaed97c9b39cf68f01

Here is my /etc/sv/myzippykid/run script which is how runit runs my 
unicorn daemon:  https://gist.github.com/grahamc/7ca7e3942d19f60d339a

my bundle_wrapper.sh simply inits the environment with the proper ruby 
and path: https://gist.github.com/grahamc/9760001e0e20c915d9e9


--

The contents of all my pastes for those of you without internet at time 
of reading:

https://gist.github.com/grahamc/35715c1a2c9717e461e3:

runsvdir -P /etc/service log: 
............................................................................................................................................. 

 \_ runsv myzippykid
     \_ svlogd -tt ./main
     \_ /bin/bash /opt/myzippykid/bundle_wrapper.sh exec unicorn -E 
staging -c /etc/unicorn/myzippykid.rb
         \_ ruby 
/opt/myzippykid/deployment/releases/6328251212178494a6c82b11ec518dc4d1962f9c/vendor/bundle/ruby/2.0.0/bin/unicorn 
-E staging -c /etc/unicorn/myzippykid.rb
unicorn master -E staging -c /etc/unicorn/myzippykid.rb
 \_ unicorn worker[0] -E staging -c /etc/unicorn/myzippykid.rb




https://gist.github.com/grahamc/08aaed97c9b39cf68f01:

# What ports/sockets to listen on, and what options for them.
listen "8080", :tcp_nodelay => true, :backlog => 100

working_directory '/opt/myzippykid/deployment/current'

# What the timeout for killing busy workers is, in seconds
timeout 60

# Whether the app should be pre-loaded
preload_app false

# How many worker processes
worker_processes 1



# What to do before we fork a worker
before_fork do |server, worker|
  sleep 1
end

https://gist.github.com/grahamc/7ca7e3942d19f60d339a:

#!/bin/bash

cd /opt/myzippykid/deployment/current

exec 2>&1
exec /usr/bin/chpst \
  -u myzippykid:myzippykid \
  /opt/myzippykid/bundle_wrapper.sh exec \
  unicorn \
  -E staging \
  -c /etc/unicorn/myzippykid.rb

https://gist.github.com/grahamc/9760001e0e20c915d9e9:

#!/bin/bash

source /etc/profile.d/rbenv.sh

bundle $@


Thank you for your help,
-- Graham Christensen


_______________________________________________
Unicorn mailing list - mongrel-unicorn@rubyforge.org
http://rubyforge.org/mailman/listinfo/mongrel-unicorn
Do not quote signatures (like this one) or top post when replying