Capistrano-Nginx-Unicorn

Capistrano tasks for configuration and management nginx+unicorn combo for zero downtime deployments of Rails applications.

Provides capistrano tasks to:

• easily add application to nginx and reload it's configuration
• create unicorn init script for application, so it will be automatically started when OS restarts
• start/stop unicorn (also can be done using sudo service unicorn_<your_app> start/stop)
• reload unicorn using USR2 signal to load new application version with zero downtime
• creates logrotate record to rotate application logs

Provides several capistrano variables for easy customization. Also, for full customization, all configs can be copied to the application using generators.

Installation

Add this line to your application's Gemfile:

gem 'capistrano-nginx-unicorn', group: :development

And then execute:

$ bundle

Or install it yourself as:

$ gem install capistrano-nginx-unicorn

Usage

Add this line to your Capfile

require 'capistrano/nginx_unicorn'

You can check that new tasks are available (cap -T):

for nginx:

# add and enable application to nginx
cap nginx:setup

# upload ssl certificates
cap nginx:setup_ssl

# reload nginx configuration
cap nginx:reload

and for unicorn:

# create unicorn init script
cap unicorn:setup_initializer

# create unicorn configuration file
cap unicorn:setup_app_config

# start unicorn
cap unicorn:start

# stop unicorn
cap unicorn:stop

# reload unicorn with no downtime
# old workers will process new request until new master is fully loaded
# then old workers will be automatically killed and new workers will start processing requests
cap unicorn:restart

and shared:

# create logrotate record to rotate application logs
cap logrotate:setup

Hooks

nginx:reload, unicorn:restart are hooked to deploy:publishing

Customization

Using variables

You can customize nginx and unicorn configs using capistrano variables:

# path to customized templates (see below for details)
# default value: "config/deploy/templates"
set :templates_path, "config/deploy/templates"

# server name for nginx, default value: "localhost <application>.local"
# set this to your site name as it is visible from outside
# this will allow 1 nginx to serve several sites with different `server_name`
set :nginx_server_name, "example.com"

# path, where nginx pid file will be stored (used in logrotate recipe)
# default value: `"/run/nginx.pid"`
set :nginx_pid, "/run/nginx.pid"

# if set, nginx will be configured to 443 port and port 80 will be auto rewritten to 443
# also, on `nginx:setup`, paths to ssl certificate and key will be configured
# and certificate file and key will be copied to `/etc/ssl/certs` and `/etc/ssl/private/` directories
# default value: false
set :nginx_use_ssl, false

# if set, it will ask to upload certificates from a local path. Otherwise, it will expect
# the certificate and key defined in the next 2 variables to be already in the server.
set :nginx_upload_local_certificate, { true }

# remote file name of the certificate, only makes sense if `nginx_use_ssl` is set
# default value: `nginx_server_name + ".crt"`
set :nginx_ssl_certificate, "#{nginx_server_name}.crt"

# remote file name of the certificate, only makes sense if `nginx_use_ssl` is set
# default value: `nginx_server_name + ".key"`
set :nginx_ssl_certificate_key, "#{nginx_server_name}.key"

# nginx config file location
# centos users can set `/etc/nginx/conf.d`
# default value: `/etc/nginx/sites-available`
set :nginx_config_path, "/etc/nginx/sites-available"

# path, where unicorn pid file will be stored
# default value: `"#{current_path}/tmp/pids/unicorn.pid"`
set :unicorn_pid, "#{current_path}/tmp/pids/unicorn.pid"

# path, where unicorn config file will be stored
# default value: `"#{shared_path}/config/unicorn.rb"`
set :unicorn_config, "#{shared_path}/config/unicorn.rb"

# path, where unicorn log file will be stored
# default value: `"#{shared_path}/config/unicorn.rb"`
set :unicorn_log, "#{shared_path}/config/unicorn.rb"

# user name to run unicorn
# default value: `user` (user varibale defined in your `deploy.rb`)
set :unicorn_user, "user"

# number of unicorn workers
# default value: 2
set :unicorn_workers, 2

# local path to file with certificate, only makes sense if `nginx_use_ssl` is set
# this file will be copied to remote server
# default value: none (will be prompted if not set)
set :nginx_ssl_certificate_local_path, "/home/ivalkeen/ssl/myssl.cert"

# local path to file with certificate key, only makes sense if `nginx_use_ssl` is set
# this file will be copied to remote server
# default value: none (will be prompted if not set)
set :nginx_ssl_certificate_key_local_path, "/home/ivalkeen/ssl/myssl.key"

For example, of you site name is example.com and you want to use 4 unicorn workers, your deploy.rb will look like this:

set :nginx_server_name, "example.com"
set :unicorn_workers, 4

Template Customization

If you want to change default templates, you can generate them using rails generator

rails g capistrano:nginx_unicorn:config

This will copy default templates to config/deploy/templates directory, so you can customize them as you like, and capistrano tasks will use this templates instead of default.

You can also provide path, where to generate templates:

rails g capistrano:nginx_unicorn:config config/templates

TODO:

• add tests

Contributing

1. Fork it
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create new Pull Request