CloudBootstrap

Bootstrapping scripts to quickly configure cloud servers

How to Use

This repository can be included as a Git submodule or packaged with your application, or downloaded as a tarball from GitHub. Both methods are detailed below.

Submodule

If included as a submodule or packaged with an application, the scripts/run script can be called directly. The configuration script location must be set in the ConfigFile environment variable (defaults to "github:JungleCatSoftware/CloudConfigs/example/config@master").

Tarball Download

The config scripts can be downloaded and launched by downloading and executing the bootstrap.sh script. As with the Submodule method, the ConfigFile environment variable will need to be set (defaults to "github:JungleCatSoftware/CloudConfigs/example/config@master"). The following is an example script that can be used or included into into your boot configs (such as AWS's Instance UserData):

#!/bin/bash
export ConfigFile='github:JungleCatSoftware/CloudConfigs/example/config@master'
curl -L https://raw.githubusercontent.com/JungleCatSoftware/CloudBootstrap/master/bootstrap.sh | /bin/bash

File Locators

The CloudBootstrap scripts make use of various file locators, which are shorthand for various types of files, such as modules from the Puppet Forge or GitHub. The following are all supported locator prefixes since version 0.0.1:

• forge:
• github:
• file:
• http://
• https://

Puppet Forge Locator

The Puppet Forge locator refers to a Puppet Module on the public Puppet Forge. This locator is only valid in the PUPPET_MODULES variable and cannot be used in any other context. It follows the form "forge:USER-MODULE(@VERSION)" where USER is the name of the account that uploaded the module, MODULE is the name of the module, and VERSION is an optional version to fetch (defaults to the latest). The following are all valid examples:

• "forge:puppetlabs-stdlib"
• "forge:jfryman-nginx@0.2.6"

GitHub Locator

The GitHub locator reference a GitHub repository or a specific file within a repository, optionally at a specific revision, tag, or commit. A GitHub locator follows the form "github:USER/REPO(/path/to/file)(@VERSION)" where USER is the account owning the repository, REPO is the name of the repository, followed by a (optional) path to a specific file in the repository, and VERSION is an optional version (commit, tag, etc) amd defaults to "master" if not provided. If no path to a file is provided, the whole repository will be downloaded as a tarball, otherwise, the indicated file will be downloaded only. The following are valid GitHub locators:

• "github:puppetlabs/puppetlabs-firewall@1.6.x"
• "github:camptocamp/puppet-openssl"
• "github:JungleCatSoftware/CloudConfigs/example/config"
• "github:JungleCatSoftware/CloudBootstrap/README.md@0.0.1"

File Locator

The File locator is a handy way to reference a file on the local filesystem. The file at the referenced path will be copied to /tmp and the new path returned. This is useful for development environments in Vagrant to copy config files from the /vagrant mountpoint.

The Config File

The config file, which is referenced by the ConfigFile envionment variable when calling either the bootstrap.sh or scripts/run is a simple container of variable definitions which are used by the scripts in this repository. The variables that are used are detailed below.

CONFIG_MANGER

This variable contains the name of the config manager to configure and install.

Valid values are:

• puppet

PUPPET_LIBRARIAN_FILE

This variable is a file locator to a librarian-puppet Puppetfile. If present, librarian-puppet will be installed, the will file downloaded, and then librarian-puppet install will be run to install the modules in the Puppetfile.

PUPPET_MODULES

This variable is an array of locators, including the special-case Puppet Forge ("forge:") locator. Each locator must either be a Forge module or a tarball (or a full GitHub repository) containing a Puppet module.

HIERA_FILE

The HIERA_FILE variable is a locator for a Hiera YAML file. This file will be downloaded and installed both to /etc/hiera.yaml and to /etc/puppet/hiera.yaml. It should be sure to reference the other Hiera data files referenced in the HIERA_DATA_FILES variable.

HIERA_DATA_FILES

This variable is an array locators and their destination file path in the form of LOCATOR|ABSOLUTE_FILE_PATH. For example:

github:JungleCatSoftware/CloudConfigs/example/hiera/example.yaml|/etc/puppet/hieradata/example.yaml

The locator should point to a file to download with a reference to where it should reside of the filesystem. The path should be absolute and should be within the hierchy defined in the hiera.yaml