Writing an upstart config file for Unicorn

ruby-on-railsubuntu-10.04unicornupstart

I've had a good hunt and can't find a good example of an upstart script for Unicorn.

From what I've read however I think that upstart and unicorn might step on each others toes if I just try to set up a basic upstart config file given that unicorn manages its own workers etc…

Any tips or shortcuts before I dig in myself?

Best Answer

You may use this config, but my personal choice is to use BluePill instead. It will give you more flexibility

description "Starts unicorn as the user 'rails'"
start on filesystem
chdir /apps/rails/sample_app
exec sudo -u deployer -i 'rvm ree; unicorn_rails -c /apps/rails/sample_app/config/unicorn.rb -E production' 
respawn

Related Solutions

Can upstart handle symbolic links to config files

Upstart does not support symlinks because they might point to a file on a partition that is not loaded at boot time.

I have gotten around this in my own project by putting the conf files in /etc/init/myscripts and then binding that to a directory in my repository. mount --bind /etc/init/myscripts ~/code/repo/initscripts.

Add this to /etc/fstab and the binding will be persistent:

/etc/init/myscripts  /home/me/code/repo/initscripts      none    bind

This effectively gives you hard-linked directories. Upstart will treat the conf files as any other, because they are local to /etc/init. Your DVCS will also see them as local files in the repo, so it will also treat them as it would any other files stored there. Best of both worlds.

Using Upstart to manage Unicorn w/ rbenv + bundler binstubs w/ ruby-local-exec shebang

Indeed, a limitation of upstart is that it cannot track daemons that do what unicorn is doing.. that being fork/exec and exit their main process. Believe it or not, sshd does the same thing on SIGHUP, and if you look, /etc/init/ssh.conf makes sure sshd runs in the foreground. This is also one reason apache2 still uses an init.d script.

It sounds like gunicorn actually sort of daemonizes itself when receiving SIGUSR1 by forking and then exitting. This would be confusing for any of the process managers out there that try and keep a process alive.

I think you have two options. 1 is just to not use the SIGUSR1 and stop/start gunicorn when you need it.

The other option is to not use upstart's pid tracking, and just do this:

start on ..
stop on ..

pre-start exec gunicorn -D --pid-file=/run/gunicorn.pid
post-stop exec kill `cat /run/gunicorn.pid`

Not as sexy as pid tracking, but at least you won't have to write a whole init.d script.

(incidentally, this has nothing to do with the shebangs/execs. Both of those things work just like running a regular executable, so they wouldn't cause any extra forks).

Best Answer

Related Solutions

Can upstart handle symbolic links to config files

Using Upstart to manage Unicorn w/ rbenv + bundler binstubs w/ ruby-local-exec shebang

Related Topic