I thought I’d follow up my completely impractical post on god with a practical one. I needed to write an upstart script for god and couldn’t find any examples out there. Here’s what I ended up doing.
# /etc/init/god.conf
start on runlevel [2345]
stop on runlevel [06]
setuid webkite
setgid webkite
respawn
respawn limit 10 60
env HOME=/home/webkite
exec bash -l -c 'cd /opt/node/apps/god && exec bundle exec god -c my.god.rb -l /opt/node/log/god.log -P /opt/node/pids/god.pid -D'
start on runlevel [2345]
stop on runlevel [06]
Run all the time, unless we go into single-user mode or shut down.
setuid webkite
setgid webkite
Run as an unprivileged user. Don’t run as root.
On the negative side: We can’t take advantage of the event driven conditions in god, such as ‘kill process if memory exceeds a half a gig’.
On the positive side: We don’t run as root. We don’t need a system rvm. And we don’t need to run rvm as root.
respawn
respawn limit 10 60
Have upstart respawn the process if it dies unexpectedly, but don’t let it go into death throes and overwhelm the server if it’s just plain broken.
env HOME=/home/webkite
We end up running a bash login shell to load rvm functions, but even
that assumes that you have a decent $HOME variable. We don’t
without this.
exec bash -l -c 'cd /opt/node/apps/god && exec bundle exec god -c my.god.rb -l /opt/node/log/god.log -P /opt/node/pids/god.pid -D'
Actually start god. This was the really tricky part.
We need bash -l -c so rvm works. rvm use won’t work in sh.
Upstart uses magic to track the process id. If you fork or
daemonize, this changes. Upstart provides the two options expect
fork and expect daemonize which works in most cases. Or so I’m
told. But we still lost the proper process id with god for
unknown reasons. So we needed to:
Use exec so bash doesn’t start its own process.
Specify -D (no-daemonize) even though we are daemonized,
so that god doesn’t fork on its own and upstart gets the
correct process id.
Hope this helps someone.