Automatic Deployment of Rails Application by Capistrano

Previously we deployed our application manually in our Linux box. This was taking a lot of time and after every deploy we were facing some serious issues. Automatic deployment by Capistrano has made deployment life easy. Here i have shared my experience about deployment by Capistrano.

Requirements:
Install Capistrano in local machine by : gem install capistrano
Install subversion command line client in local machine
Install subversion in Linux box. I have done it by a rpm

Assumptions:

• You are deploying to a single server, at least initially.
• Your username is the same for both your subversion repository and your deployment machine.
• Your subversion repository is accessible from both your local host, and the host that you will be deploying to
• You have configured a public key to use for SSH access to your target host

Run Capistrano:

From command prompt go to your application directory then run .capify.This will create a capfile in the application root directory and deploy.rb in config folder.

Configure deploy.rb:

Now modify the deploy.rb file to deploy your application. Here is my deploy.rb

require 'mongrel_cluster/recipes‘

set :application, “your application name"
set :repository, “url of your repository"
set :scm_username, “name"
set :scm_password, “password"
set :rails_env, "production --trace"
set :deploy_to, "/var/www/rails/#{application}"

#ssh configuration
set :user, ‘user_name'
ssh_options[:port] = 22
ssh_options[:username] = ‘user_name‘
ssh_options[:host_key] = 'ssh-dss'

#set app, web, db role
role :app, "www.scrumpad.com"
role :web, "www.scrumpad.com"
role :db, "www.scrumpad.com", :primary => true

#configure for mongrel cluster
set :mongrel_servers, 3
set :mongrel_port, 6010
set :mongrel_environment, 'production'
set :mongrel_conf,"#{deploy_to}/current/config/mongrel_cluster.yml"

Now the time for deploying your application:

Go to your project directory and run cap deploy:setup Capistrano will now log into your server and try to set up a simple directory hierarchy. The basic layout is:
#{deploy_to}/
#{deploy_to}/releases
#{deploy_to}/shared
#{deploy_to}/shared/log
#{deploy_to}/shared/system
#{deploy_to}/shared/pids

Every time you deploy your application, a new directory is created under the releases directory, and then a symlink is created at “#{deploy_to}/current that points at the newr elease. This current symlink should be considered the root directory of the currently active release.

run cap -q deploy:check to check dependencies.

Cold deploy: The first time your application is not running. Things are “cold”. For this case, run cap deploy:cold

Subsequent Deploys: After the first deploy every time you want to deploy just
run cap deploy

Roll back your application to the previous release cap deploy:rollback

Problem I faced:
After deploying my application on a Linux box, I faced some permission related problem to run the application. To solve this I had to put the required file in a directory and modify my deploy.rb in the following way:

after "deploy:symlink", :copy_permitted_files
task :copy_permitted_files, :roles => :app do
run "cp -rpf /var/www/rails/Permitted/public/* /var/www/rails/# application}/current/public/"
run "cp -pf /var/www/rails/Permitted/database.yml /var/www/rails/#{application}/current/config/"
run "cp -rpf /var/www/rails/Permitted/log/* /var/www/rails/#{application}/current/log/"
end