Backup is a system utility for Linux and Mac OS X, distributed as a RubyGem, that allows you to easily perform backup operations. It provides an elegant DSL in Ruby for modeling your backups. Backup has built-in support for various databases, storage protocols/services, syncers, compressors, encryptors and notifiers which you can mix and match. It was built with modularity, extensibility and simplicity in mind.

To install the latest version, run:

$ [sudo] gem install backup

Do not add gem backup to an application's Gemfile

See Installation for more information about installing and updating your installation of Backup.

See Release Notes for changes in the latest version.

Backup supports Ruby versions 1.8.7, 1.9.2, 1.9.3 and 2.0.0.

Backup allows you to model your backup jobs using a Ruby DSL:

Backup::Model.new(:my_backup, 'Description for my_backup') do
  # ... components here ...
end

The :my_backup symbol is the model's trigger and used to perform the job:

$ backup perform --trigger my_backup

Backup's components are added to the backup model to define the actions to be performed.
All of Backup's components are fully documented in the Backup Wiki.
The following is brief overview of the components Backup provides:

Archives create basic tar archives. Both GNU and BSD tar are supported.
Databases create backups of one of the following supported databases:

• MySQL
• PostgreSQL
• MongoDB
• Redis
• Riak

Any number of Archives and Databases may be defined within a backup model.

Adding a Compressor to your backup will compress all the Archives and Database backups within your final archive package.
Gzip, Bzip2 and other similar compressors are supported.

Adding a Encryptor allows you to encrypt your final backup package.
Both OpenSSL and GPG are supported.

Your final backup package might look something like this:

$ gpg --decrypt my_backup.tar.gpg --outfile my_backup.tar
$ tar -tvf my_backup.tar
  my_backup/
  my_backup/archives/
  my_backup/archives/user_avatars.tar.gz
  my_backup/archives/log_files.tar.gz
  my_backup/databases/
  my_backup/databases/PostgreSQL.sql.gz
  my_backup/databases/Redis.rdb.gz

Once your final backup package is ready, you can use any number of the following Storages to store it:

• Amazon Simple Storage Service (S3)
• Rackspace Cloud Files (Mosso)
• Ninefold Cloud Storage
• Dropbox Web Service
• Remote Servers (Available Protocols: FTP, SFTP, SCP and RSync)
• Local Storage (including network mounted locations)

All of the above Storages (except RSync) support:

• Cycling to keep and rotate multiple copies of your stored backups.

• Splitter to break up a large backup package into smaller files.

When using the RSync Storage, once a full backup has been stored, subsequent backups only need to transmit the changed portions of the final archive to bring the remote copy up-to-date.

Syncers are processed after your final backup archive has been stored and allow you to perform file synchronization.

Backup includes two types of Syncers:

• RSync: Used to sync files locally, local-to-remote (Push), or remote-to-local (Pull).
• Cloud: Used to sync files to remote storage services like Amazon S3 and Rackspace.

A backup model may contain only Syncers as well.

Notifiers are used to send notifications upon successful and/or failed completion of your backup model.

Supported notification services include:

• Email (SMTP, Sendmail, Exim and File delivery)
• Twitter
• Campfire
• Presently
• Prowl
• Hipchat
• Pushover
• POST Request
• Nagios

Backup makes it easy to setup new backup model files with it's Generator command.

$ backup generate:model -t my_backup --archives --databases=postgresql,redis --compressors=gzip \
    --encryptors=gpg --storages=sftp,s3 --notifiers=mail,twitter

Simply generate a new model using the options you need, then update the configuration for each component using the Wiki documentation.

The following is an example of a what this Backup model might look like:

Backup::Model.new(:my_backup, 'Description for my_backup') do
  split_into_chunks_of 250

  archive :user_avatars do |archive|
    archive.add '/var/apps/my_sample_app/public/avatars'
  end

  archive :log_files do |archive|
    archive.add     '/var/apps/my_sample_app/logs'
    archive.exclude '/var/apps/my_sample_app/logs/exclude-this.log'
  end

  database PostgreSQL do |db|
    db.name               = "pg_db_name"
    db.username           = "username"
    db.password           = "password"
  end

  database Redis do |db|
    db.name               = "redis_db_name"
    db.path               = "/usr/local/var/db/redis"
    db.password           = "password"
    db.invoke_save        = true
  end

  compress_with Gzip

  encrypt_with GPG do |encryption|
    encryption.mode = :symmetric
    encryption.passphrase = 'my_password'
  end

  store_with SFTP do |server|
    server.username   = "my_username"
    server.password   = "my_password"
    server.ip         = "123.45.678.90"
    server.port       = 22
    server.path       = "~/backups/"
    server.keep       = 5
  end

  store_with S3 do |s3|
    s3.access_key_id     = "my_access_key_id"
    s3.secret_access_key = "my_secret_access_key"
    s3.region            = "us-east-1"
    s3.bucket            = "bucket-name"
    s3.path              = "/path/to/my/backups"
    s3.keep              = 10
  end

  notify_by Mail do |mail|
    mail.on_success           = false

    mail.from                 = "[email protected]"
    mail.to                   = "[email protected]"
    mail.address              = "smtp.gmail.com"
    mail.port                 = 587
    mail.user_name            = "[email protected]"
    mail.password             = "my_password"
    mail.authentication       = "plain"
    mail.encryption           = :starttls
  end

  notify_by Twitter do |tweet|
    tweet.consumer_key       = "my_consumer_key"
    tweet.consumer_secret    = "my_consumer_secret"
    tweet.oauth_token        = "my_oauth_token"
    tweet.oauth_token_secret = "my_oauth_token_secret"
  end
end

The Getting Started page provides a simple walk-through to familiarize you with setting up, configuring and running a backup job.

If you have any suggestions or problems, please submit an Issue or Pull Request using Backup's Issue Log.

If you find any errors or omissions in Backup's documentation Wiki, please feel free to edit it!

Backup has seen many improvements over the years thanks to it's Contributors, as well as those who have help discuss issues and improve the documentation, and looks forward to continuing to provide users with a reliable backup solution.

Copyright (c) 2009-2013 Michael van Rooijen ( @meskyanichi )
Released under the MIT License.