mongrel_cluster service is not starting the mongrel cluster

I ran into an issue setting up a service to start mongrel_cluster on boot. I followed an excellent set of instructions that I found here, but I ran into a couple problems.

Problem 1: service mongrel_cluster start gives no output and does not start the cluster

I looked through the init script and found that the script called mongrel_cluster_ctl to start the cluster. So I logged in as root and ran the command.

# sudo su
# mongrel_cluster_ctl start
The program 'mongrel_cluster_ctl' is currently not installed.  You can install it by typing:
apt-get install mongrel-cluster

Ah! update $PATH for root to include gems.

Issue the following command to get the path to mongrel_cluster_ctl

which mongrel_cluster_ctl

Then edit /etc/init.d/mongrel_cluster to add the full path to mongrel_cluster_ctl to the inti script

Now test.

# sudo su
# mongrel_cluster_ctl start
Starting all mongrel_clusters...
# ps -ef | grep mongrel
root       860   834  0 20:22 pts/2    00:00:00 grep --color=auto mongrel

Progress! … but still no cluster

Problem 2: mongrel_cluster_ctl does not start the cluster

After much searching I finally found a reference to adding cwd: to the cluster config file in /etc/mongrel_cluster. mongrel_rails start does not work unless you are in the root of your rails project. Adding that config changes your working directory before issuing the start command.

Bam! It works!

~$ sudo su
# mongrel_cluster_ctl start
Starting all mongrel_clusters...
# ps -ef | grep mongrel
root      1289     1 86 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1292     1 99 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1295     1 99 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1298  1247  0 20:36 pts/2    00:00:00 grep --color=auto mongrel
# service mongrel_cluster stop
Stopping all mongrel_clusters...
# ps -ef | grep mongrel
root      1278  1247  0 20:36 pts/2    00:00:00 grep --color=auto mongrel
# service mongrel_cluster start
Starting all mongrel_clusters...
# ps -ef | grep mongrel
root      1289     1 86 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1292     1 99 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1295     1 99 20:36 ?        00:00:01 /usr/bin/ruby1.8 /var/lib/gems/1.8/bin/mongrel_rails
root      1298  1247  0 20:36 pts/2    00:00:00 grep --color=auto mongrel

Problem 3: mongrel_cluster is not restarting during capistrano deploy

Since I had created a custom restart task in my deploy.rb (to restart passenger), capistrano was not restarting the mongrel servers. I had to add a new line in that task to force the mongrel server restart. Here is my deploy.rb

require 'mongrel_cluster/recipes'

set :application, "blog"
set :repository,  "<my repo>"
set :branch, "master"
set :deploy_to, "/var/www/blog"
set :mongrel_conf, "#{current_path}/config/mongrel_cluster.yml"
set :user, "root"
#set :use_sudo, true
#default_run_options[:pty] = true

set :scm, :git
# Or: `accurev`, `bzr`, `cvs`, `darcs`, `git`, `mercurial`, `perforce`, `subversion` or `none`

role :web, "<my server>"                          # Your HTTP server, Apache/etc
role :app, "<my server>"                         # This may be the same as your `Web` server
role :db,  "<my server>", :primary => true # This is where Rails migrations will run

# If you are using Passenger mod_rails uncomment this:
# if you're still using the script/reapear helper you will need
# these http://github.com/rails/irs_process_scripts

 namespace :deploy do
   task :start do ; end
   task :stop do ; end
   task :restart, :roles => :app, :except => { :no_release => true } do
     run "#{try_sudo} touch #{File.join(current_path,'tmp','restart.txt')}"
      run "cd #{release_path} && mongrel_rails cluster::restart"  # Restarts mongrel clusters
   end
   desc "Update the crontab file"  
   task :update_crontab, :roles => :db do  
     run "cd #{release_path} && whenever --update-crontab #{application}"  
   end
 end

  after "deploy:symlink", "deploy:update_crontab"  
  after "deploy", "deploy:migrate"