Beaver

Beaver is a deployment management system, to manage and deploy all of your sites in one command. It's utilize Capistrano which is a Remote multi-server automation tool based on Ruby Rake.

A successful deployment with Beaver will result in a directory structure like the following:

site
├── current
├── releases
├── repo
└── shared

Each time you deploy,

The repo folder will be synced with the last version of the site from the remote repository.

A new directory will be created under the releases directory, and the new version of the site will be placed there.

The current folder is a symbolic link, and it will point to the new release directory whenever you will create a new release.

Installation

Windows 7

1. Clone the Beaver project into your favorite directory

2. Install RubyGems

RubyGems is a package manager for the Ruby libraries. Download RubyGems for windows, you can get it here.

Run it.

The installation will add ruby and gem environment variables into your system path.

If not just run

$ set PATH=C:\Ruby193\bin;%PATH%

3. Install Capistrano:

$ gem install capistrano -v 3.0.1

OSX

1. Clone the Beaver project

2. Install Capistrano:

$ sudo gem install capistrano -v 3.0.1

Folder structure

Cloning the Beaver project will create the following folder structure:

beaver/
│
├── Capfile
├── beaver.rb
├── deploy.rb
├── config/
│   └── deploy/
│     	 ├── examples/
│    	 │		├── sandbox.rb
│        │		├── qa.rb
│        │		├── staging.rb
│        │		└── production.rb
│		 │
│        └──.gitignore
│
├── sites/
│
└── lib

Configuration

In order to configure your servers and ssh keys, you will have to add environments config files under the C:\beaver\config\deploy folder...

Go to the examples folder:

$ cd C:\beaver\config\deploy\examples

Under that folder you can find server configurations, divided into environments files. Within each environment file, you will find a list, of all it's related severs, for example:

server 'ec2-50-19-45-59.compute-1.amazonaws.com', user: 'ubuntu', roles: %w{web app db}

The above, defines the server host, the user, and it's roles.

All you need to do is to copy the files that you need into the C:\yourfolder\config\deploy folder and declare the servers that will take part in the deployment process!

You can declare as many servers as you want, but, if all of the server share the same database, be sure to add the db role, only on the primary server(the top most server)

server 'ec2-50-19-45-59.compute-1.amazonaws.com', user: 'ubuntu', roles: %w{web app db}
server 'ec2-50-16-63-231.compute-1.amazonaws.com', user: 'ubuntu', roles: %w{web app}
server 'ec2-54-227-115-121.compute-1.amazonaws.com', user: 'ubuntu', roles: %w{web app}

Next, you have to configure your ssh keys.

Windows 7

1. Add your shh private key under C:\Users\USERNAME\.ssh

2. Within each environment config file replace the USERNAME with your user name.

set :ssh_options, {
	keys: ["#{ENV['USERNAME']}/.ssh/keyname.pem"],
	forward_agent: false,
	auth_methods: %w(publickey)
}

OSX

1. Add your shh private key under /home/user/.ssh/id_rsa

2. Within each environment config file replace the user with your user name.

set :ssh_options, {
	keys: %w(/home/user/.ssh/id_rsa),
	forward_agent: false,
	auth_methods: %w(password)
}

Usage

1. Go into the beaver project and run the beaver.rb script

$ cd C:\beaver
$ ruby beaver.rb -h

This will display all of the vireos options

Deploy

$ ruby beaver.rb -e sandbox -s example1.com,example2.com,example3.com

Running the beaver.rb with those options will deploy the sites: example1.com,example2.com,example3.com into all of the configured sandbox servers. The -s option takes an array of comma delimited sites (no spaces).

Rollback

$ ruby beaver.rb -e sandbox -s  example1.com,example2.com,example3.com -a rollback

Generic tasks

Beaver enable you to run tasks witch are not part of the deploy process.

$ ruby beaver.rb -e sandbox -s example1.com -a namespace:task

$ ruby beaver.rb -e sandbox -a namespace:task

Some more options

• -H, --hosts Optional, Let's you list all of the hostnames you wish to deploy to and it will create an environment file with the listed hostnames.

If you wish to query the AWS api in order to get a list of servers hostnames you wish to deploy to (useful in autoscale), you can pass the following params:

• --accesskey AWS config access key
• --secretaccess AWS config secret access key
• --region AWS region
• --filters AWS tags, ordered in key,value pairs

Defining the above params, will create a servers config file (with the name of the -e env).

Alternatively, you can create a this file on your own.