backupninja Cookbook

The main highlights of this cookbook are the resources for managing backups. The following providers are currently implemented:

• backupninja_rdiff: backup action for rdiff-backup (incremental backups)
• backupninja_mysql: backup action for safe MySQL dumps

Requirements

Chef version 0.11+

Platforms

• Ubuntu

All other platforms are currently untested.

Resources/Providers

These resources aim to expose an abstraction layer for generating the ini-style configuration files necessary to configure a backup job in backupninja.

If you wish to install backupninja, add the included recipe[backupninja] to your runlist or wrapper cookbook.

backupninja

Generic backupninja resource. This resource contains all the generic configuration data available in all the backup job-specific resources. This resource is in charge of the file naming-scheme and the generation of the INI-file.

Actions

• :create: create the ini file
• :delete: removes the ini file

Attribute Parameters

• name: name attribute. Short descriptive name that will be used in the filename.
• priority
• description
• when_executed
• type: the extension of the INI-file indication which procedure for backup will be used.
• options: Hash of additional values to appear in the INI-file.

More information of about ordering, naming and scheduling can be found here.

The options hash will be heavily used by the job-specific resources to fill in their additional parameters.

Generation of these options depends on the class type of the hash value being used:

• Value is of type Array: A seperate "key = item" line will be generated for each item in the array.
• Value is of type Hash: the key will be used as a section [key] and the value should again be of type Hash. Everything in that hash will be generated as attributes inside this section.
• Everything else will simply be generated as "key = value"

backupninja_rdiff

Rdiff-specific job to make remote, incremental backups of the filesystem.

Actions

• :create: create the ini file
• :delete: removes the ini file

Attribute Parameters

The following attribute parameters are present alongside those in the backupninja resource:

The source section:
* source_label, String
* source_type, String
* source_keep, Integer
* source_include, Array
* source_exclude, Array

The dest (destination) section:
* dest_type, String
* dest_host, String
* dest_directory, String
* dest_user, String

More information about these attributes can be found here.

backupninja_mysql

MySQL-specific job to backup a mysql database using mysqldump.

Actions

• :create: create the ini file
• :delete: removes the ini file

Attribute Parameters

The following attribute parameters are present alongside those in the backupninja resource:

• databases, String
• backupdir, String
• hotcopy, String
• sqldump, String
• sqldumpoptions, String
• compress, String
• nodata, String

• vsname, String

• user, String

• dbusername, String

• dbpassword, String

• configfile, String

More information about these attributes can be found here.

Attributes

backupninja::default

Install and configure the backupninja daemon.

<table>
<tr>
<th>Key</th>
<th>Type</th>
<th>Description</th>
<th>Default</th>
</tr>
<tr>
<td><tt>['backupninja']['backupninja_package_name']</tt></td>
<td>String</td>
<td>Package name for backupninja</td>
<td><tt>'backupninja'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['rdiff_backup_package_name']</tt></td>
<td>String</td>
<td>Package name for rdiff-backup</td>
<td><tt>'rdiff-backup'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['loglevel']</tt></td>
<td>Integer</td>
<td>
Verbosity level for logging:
<ul>
<li>5: Debug and below</li>
<li>4: Information and below</li>
<li>3: Warnings and below</li>
<li>2: Errors and below</li>
<li>1: Fatal errors only</li>
</ul>
</td>
<td><tt>4</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['reportemail']</tt></td>
<td>String</td>
<td>Backup summaries will be emailed to this address</td>
<td><tt>'root'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['reportsuccess']</tt></td>
<td>String</td>
<td>When set to 'yes', a report email will be generated even
if there was no error.</td>
<td><tt>'yes'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['reportwarning']</tt></td>
<td>String</td>
<td>When set to 'yes', a report email will be generated even
if there was no error.</td>
<td><tt>'yes'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['logfile']</tt></td>
<td>String</td>
<td>All logs will be written to this file</td>
<td><tt>'/var/log/backupninja.log'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['configdirectory']</tt></td>
<td>String</td>
<td>Directory where all the backup configuration files live</td>
<td><tt>'/etc/backup.d'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['scriptdirectory']</tt></td>
<td>String</td>
<td>Where backupninja handler scripts are found</td>
<td><tt>/usr/share/backupninja'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['usecolors']</tt></td>
<td>String</td>
<td>Use colors in the log file</td>
<td><tt>'yes'</tt></td>
</tr>
</table>

backupninja::rdiff_backup_client

The recipe will install the rdiff-backup package and generates an ssh keypair for the root user. The generated public key will be saved as a node attribute (backupclient_pubkey) so the backup server knows which public ssh keys it should authorize.

<table>
<tr>
<th>Key</th>
<th>Type</th>
<th>Description</th>
<th>Default</th>
</tr>
<tr>
<td><tt>['backupninja']['rdiff_backup_package_name']</tt></td>
<td>String</td>
<td>Package name for rdiff-backup</td>
<td><tt>'rdiff-backup'</tt></td>
</tr>
</table>

backupninja::rdiff_backup_server

Use this recipe in the run list of the server that should receive the backups from the client.

Basically, it will create a backup user and allow him access to the backup location. An .ssh/authorized_keys will be generated based on the backupclient_pubkey attribute found on all nodes known to the Chef Server to allow all backup clients access to send their backups.

<table>
<tr>
<th>Key</th>
<th>Type</th>
<th>Description</th>
<th>Default</th>
</tr>
<tr>
<td><tt>['backupninja']['rdiff_backup_server_user']</tt></td>
<td>String</td>
<td>User used to connect to the backup server.</td>
<td><tt>'backupuser'</tt></td>
</tr>
<tr>
<td><tt>['backupninja']['rdiff_backup_server_directory']</tt></td>
<td>String</td>
<td>Directory where backups will be kept</td>
<td><tt>'/backups'</tt></td>
</tr>
</table>

Usage

backupninja::default

Installs and configures a backupninja daemon on the machine. No backup tasks are added.

e.g.
Just include backupninja in your node's run_list:

{
  "name":"my_node",
  "run_list": [
    "recipe[backupninja]"
  ]
}

To Do

• Make job-specific resources more DRY-er
• Implement other job processes
• Change all 'yes' attribute parameters from strings to booleans.
• Ensure only one of three allowed authorization techniques in the MySQL job can be configured in a resource.

Contributing

1. Fork the repository on Github
2. Create a named feature branch (like add_component_x)
3. Write your change
4. Write tests for your change (if applicable)
5. Run the tests, ensuring they all pass
6. Submit a Pull Request using Github

License and Authors

Authors: Kevin Van Wilder kevin@van-wilder.be