ra10ke

Rake tasks related to R10K and Puppetfile.

Usage

Add the following line to your Gemfile:

gem 'ra10ke'

Add the following lines in your Rakefile:

require 'ra10ke'
Ra10ke::RakeTask.new

Configuration

You can configure the tasks in a block:

Ra10ke::RakeTask.new do |t|
  t.basedir = File.join(Dir.pwd, 'some_dir')
  t.moduledir = File.join(Dir.pwd, 'some_dir/strange_module_dir')
end

Available settings are:

Setting	Documentation
basedir	Base directory with the Puppetfile and modules directory (Default: Same directory as Rakefile)
moduledir	Directory to install the modules in (Default: 'modules' in basedir)
puppetfile_path	Directroy where the Puppetfile is (Default: basedir)
puppetfile_name	The Puppetfile name (Default: basedir/Puppetfile)
force	Overwrite locally changed files on install (Default: false)
purge	Purge unmanaged modules from the modulesdir (Default: false)

Rake tasks

r10k:syntax

Syntax check for the Puppetfile. Similar to the r10k puppetfile check command.

r10k:dependencies

This rake task goes through the modules that are declared in the Puppetfile, and prints outdated modules.

Supports:

Puppet Forge
Git (SHA-ref and Tagging)

Ignoring specific modules:

Under specific conditions you may not wish to report on specific modules being out of date, to ignore a module create .r10kignore file in the same directory as your Puppetfile.

r10k:solve_dependencies

Reads the Puppetfile in the current directory and uses the ruby 'solve' library to find missing and outdated dependencies based on their metadata.

The solver does not allow major version bumps according to SemVer by default. To allow major upgrades, call the rake task with any parameter.

The rake task will download git modules into the modules/ directory to access their metadata.json. It will also cache forge metadata in ̃$XDG_CACHE_DIR/ra10ke.metadata_cache in order to make subsequent runs faster.

r10k:install[path]

Reads the Puppetfile in the current directory and installs them under the path provided as an argument.

Limitations

It works only with modules from the Forge, and Git. SVN modules will be ignored.
Git support is explicitly SHA Ref and Tag supported. If tag is used it must follow v0.0.0 convention, other wise it will be ignored.
The version has to be specified explicitly. If it is omitted, or it is :latest, the module will be ignored.

r10k:validate[path]

The validate rake task will determine if the url is a valid url by connecting to the repository and verififying it actually exists and can be accessed. Additional if a branch, tag, or ref is specified in the Puppetfile the validate task will also verify that that branch/tag/ref exists in the remote repository.

If you have ever deployed r10k to production only to find out a tag or branch is missing this validate task will catch that issue.

A exit status of 0 is returned if there are no faults, while a 1 is returned if any module has a bad status.

Status emojis can be customized by setting the following environment variables.

Example

GOOD_EMOJI='👍'
BAD_EMOJI='😨'

NAME     | URL                                           | REF                            | STATUS
---------|-----------------------------------------------|--------------------------------|-------
splunk   | https://github.com/cudgel/splunk.git          | prod                           | 👍
r10k     | https://github.com/acidprime/r10k             | v3.1.1                         | 👍
gms      | https://github.com/npwalker/abrader-gms       | gitlab_disable_ssl_verify_s... | 👍
rbac     | https://github.com/puppetlabs/pltraining-rbac | 2f60e1789a721ce83f8df061e13... | 👍
acl      | https://github.com/dobbymoodge/puppet-acl.git | master                         | 👍
deploy   | https://github.com/cudgel/deploy.git          | master                         | 👍
dotfiles | https://github.com/cudgel/puppet-dotfiles.git | master                         | 👍
gitlab   | https://github.com/vshn/puppet-gitlab         | 00397b86dfb3487d9df768cbd36... | 👍

👍👍 Puppetfile looks good.👍👍

r10k:duplicates

This rake task parses the Puppetfile and looks for modules with duplicate declarations.

All found duplicates are reported along with their source and their version (if taken from the Forge) or their ref/tag/branch. (if taken from git)