Skip to content
This repository has been archived by the owner on Nov 10, 2019. It is now read-only.

Latest commit

 

History

History
228 lines (154 loc) · 8.89 KB

README.md

File metadata and controls

228 lines (154 loc) · 8.89 KB

⚠️ This project is no longer maintained. Thanks for your interest in capistrano-mb. I don't use this gem anymore and so I've chosen not to support it going forward. However, many of the lessons I've learned from building capistrano-mb I am now applying to a new project called tomo. Hope to see you there!


capistrano-mb

An opinionated Capistrano task library for deploying Rails apps from scratch on Ubuntu 16.04 or 18.04 LTS.

Gem Version

Capistrano is great for deploying Rails applications, but what about all the prerequisites, like Nginx and PostgreSQL? Do you have a firewall configured on your VPS? Have you installed the latest OS security updates? Is HTTPS working right?

The capistrano-mb gem adds a cap <stage> provision task to Capistrano that takes care of all that. Out of the box, provision will:

  • Install the latest postgresql, node.js, and nginx apt packages
  • Install all libraries needed to build Ruby
  • Lock down your VPS using ufw (a simple front-end to iptables)
  • Set up logrotated for your Rails logs
  • Schedule an automatic daily backup of your Rails database
  • Generate a self-signed SSL certificate if you need one
  • Set up ngnix with the latest SSL practices and integrate it with Unicorn for your Rails app
  • Create the deployer user and install an SSH public key
  • Install rbenv and use ruby-build to compile the version of Ruby required by your app (by inspecting your .ruby-version file)
  • And more!

The gem is named "capistrano-mb" because it is prescribes my (@mattbrictson) personal preferences for automating deployments of Rails projects. I've worked several years as a freelance developer juggling lots of Rails codebases, so its important for me to have a good, consistent server configuration. You'll notice that capistrano-mb is opinionated and strictly uses the following stack:

  • Ubuntu 16.04 or 18.04 LTS
  • PostgreSQL
  • Unicorn
  • Nginx
  • rbenv
  • dotenv

In addition, capistrano-mb changes many of Capistrano's defaults, including the deployment location, Bundler behavior, and SSH keep-alive settings. (See defaults.rake for details.)

Not quite to your liking? Consider forking the project to meet your needs.

Roadmap

I plan to continue maintaining this project for the benefit of deploying my own Rails apps for the foreseeable future. In practice, this means a new version or two per year. The behavior of capistrano-mb may change as I upgrade my apps to new versions of Rails. For example, at some point I might:

  • Replace Unicorn with Puma
  • Switch from dotenv to encrypted credentials
  • Add Let's Encrypt
  • Use a more robust database backup solution

Future changes in capistrano-mb are not guaranteed to have graceful migration paths, so I recommend pinning your Gemfile dependency to a specific version and upgrading with extreme care.

Quick start

Please note that this project requires Capistrano 3.x, which is a complete rewrite of Capistrano 2.x. The two major versions are not compatible.

1. Purchase an Ubuntu 16.04 or 18.04 VPS

To use capistrano-mb, you'll need a clean Ubuntu 16.04 or 18.04 server to deploy to. The only special requirement is that your public SSH key must be installed on the server for the root user.

Test that you can SSH to the server as root without being prompted for a password. If that works, capistrano-mb can take care of the rest. You're ready to proceed!

2. .ruby-version

capistrano-mb needs to know the version of Ruby that your app requires, so that it can install Ruby during the provisioning process. Place a .ruby-version file in the root of your project containing the desired version, like this:

2.5.0

If you are using rbenv, just run rbenv local 2.5.0 and it will create this file for you.

3. Gemfile

capistrano-mb makes certain assumptions about your Rails app, namely that it uses dotenv to manage Rails secrets via environment variables, and that it runs on top of PostgreSQL and unicorn. Make sure they are specified in the Gemfile:

gem "dotenv-rails", ">= 2.0.0"
gem "pg", "~> 0.18"
gem "unicorn"

Then for the capistrano-mb tools themselves, add these gems to the development group:

group :development do
  gem "capistrano-bundler", :require => false
  gem "capistrano-rails", :require => false
  gem "capistrano", "~> 3.10", :require => false
  gem "capistrano-mb", "~> 0.35.0", :require => false
end

And then execute:

$ bundle install

4. cap install

If your project doesn't yet have a Capfile, run cap install with the list of desired stages (environments). For simplicity, this installation guide will assume a single production stage:

bundle exec cap install STAGES=production

5. Capfile

Add these lines to the bottom of your app's Capfile (order is important!):

require "capistrano/bundler"
require "capistrano/rails"
require "capistrano/mb"

6. deploy.rb

Modify config/deploy.rb to set the specifics of your Rails app. At the minimum, you'll need to set these two options:

set :application, "my_app_name"
set :repo_url, "git@github.com:username/repository.git"

7. production.rb

Modify config/deploy/production.rb to specify the IP address of your production server. In this example, I have a single 1GB VPS (e.g. at DigitalOcean) that plays all the roles:

server "my.production.ip",
       :user => "deployer",
       :roles => %w[app backup cron db web]

Note that you must include the backup and cron roles if you want to make use of capistrano-mb's database backups and crontab features.

8. secrets.yml

Your Rails apps may have a config/secrets.yml file that specifies the Rails secret key. capistrano-mb configures dotenv to provide this secret in a RAILS_SECRET_KEY_BASE environment variable. You'll therefore need to modify secrets.yml as follows:

production:
  secret_key_base: <%= ENV["RAILS_SECRET_KEY_BASE"] %>

9. Provision and deploy!

Run capistrano-mb's provision task. This will ask you a few questions, install Ruby, PostgreSQL, Nginx, etc., and set everything up. The entire process takes about 10 minutes (mostly due to compiling Ruby from source).

bundle exec cap production provision

Once that's done, your app is now ready to deploy!

bundle exec cap production deploy

Advanced usage

Choosing which recipes to auto-run

Most of the capistrano-mb recipes are designed to run automatically as part of cap <stage> provision, for installing and setting up various bits of the Rails infrastructure, like nginx, unicorn, and postgres. Some recipes also contribute to the cap <stage> deploy process.

This auto-run behavior is fully under your control. In your deploy.rb, set :mb_recipes to an array of the desired recipes. If you don't want a recipe to execute as part of deploy/provision, simply omit it from the list.

The following list will suffice for most out-of-the-box Rails apps. The order of the list is not important.

set :mb_recipes, %w[
  aptitude
  bundler
  crontab
  dotenv
  logrotate
  migrate
  nginx
  postgresql
  rbenv
  seed
  ssl
  ufw
  unicorn
  user
  version
]

Even if you don't include a recipe in the auto-run list, you can still invoke the tasks of those recipes manually at your discretion. Run bundle exec cap -T to see the full list of tasks.

Configuration

Many of the recipes have default settings that can be overridden. Use your deploy.rb file to specify these overrides. Or, you can override per stage. Here is an example override:

set :mb_unicorn_workers, 8

For the full list of settings and their default values, refer to defaults.rake.

Further reading

Check out my rails-template project, which generates Rails applications with capistrano-mb pre-configured and ready to go.

History

This gem used to be called capistrano-fiftyfive. If you are upgrading from capistrano-fiftyfive, refer to the CHANGELOG entry for v0.22.0 for migration instructions.

As of 0.33.0, capistrano-mb no longer supports Ubuntu 12.04 or 14.04. If your server runs one of these older versions, use capistrano-mb 0.32.0.

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