A library for creating machines and infrastructures idempotently in Chef.
A library for creating machines and infrastructures idempotently in Chef.
Chef Provisioning was officially end-of-life on August 31, 2019 and is no longer included with Chef DK. The Chef Provisioning source code and drivers have been moved into the chef-boneyard organization. If you are a current user of Chef Provisioning, please contact your Chef Customer Success Manager or Account Representative to review your options.
Please join us in Slack by signing up at http://community-slack.chef.io/ and joining #chef-provisioning
Chef Provisioning is a Cookbook and Recipe based approach for managing your infrastructure. Users can codify their infrastructure and use Chef to converge their infrastructure to the desired state. It has a plugin model (called Drivers) to manage different infrastructures, including AWS, Azure and Fog.
Chef Provisioning is maintained according to the Chef Maintenance Policy.
Due to the age and engagement level with different Drivers they have fallen into two categories - maintained and unmaintained.
Maintained drivers have a representative in the maintenance policy and get triaged / discussed during office hours. They are
| AWS | Fog |
|---|---|
Maintained by Chef-Partners:
Maintained, not by Chef:
Unmaintained Drivers are at risk of becoming stale. They are:
Chef Provisioning is a good fit for people who use or want to use the Chef ecosystem to manage their infrastructure. Infrastructure is represented as Chef Resources and managed in Chef Recipes and Cookbooks. These recipes run on a provisioner node and attempt to converge the infrastructure idempotently. The AWS driver is the most used, the most tested and receives the most updates.
Chef Provisioning and its Drivers are maintained by the open source community. They currently have a slow tempo for issue triage and pull request merging. Troubleshooting a bug may require stepping through the Ruby Chef codebase and engaging the community.
If you are new to Chef and Chef Provisioning, learning both can be daunting. If you are not familiar with Ruby as well, double the learning curve. Successful provisioning users know how they want to manage their cookbooks. Generally this is some CI/CD pipeline.
Chef is very flexible which allows customized provisioning solutions at the cost of high complexity. This makes it hard to document all the possible use cases. Patterns (for example, how to manage the provisioning node and run the provisioning recipes) are not well documented.
Chef Provisioning should be used to manage infrastructure (CRUD operations) and register Chef nodes with the server for the first time only. Other use cases (like scheduling chef runs, node inventory or reporting) are not supported.
These are the primary documents to help learn about using Provisioning and creating Provisioning drivers:
This video explains the basics of chef-provisioning (though provisioners are now called drivers). Slides (more up to date) are here.
| Date | Blog |
|---|---|
| 2014-12-15 | Using Chef Provisioning to Build Chef Server |
| 2014-11-12 | Chef Launches Policy-Based Provisioning |
| 2014-11-12 | Chef Provisioning: Infrastructure As Code |
| 2014-06-03 | machine_batch and parallelization |
| 2014-06-03 | Chef Provisioning, Configuration and Drivers |
| 2014-03-04 | Chef Metal 0.2: Overview - this is a pretty good overview (though dated). |
| 2013-12-20 | Chef Metal Alpha |
You can try out Chef Provisioning in many different flavors.
To give it a spin, install the latest ChefDK, Vagrant and VirtualBox. Run the following from the chef-provisioning/docs/examples directory:
export CHEF_DRIVER=vagrant
export VAGRANT_DEFAULT_PROVIDER=virtualbox
chef-client -z vagrant_linux.rb simple.rbThis will create two vagrant precise64 linux boxes, "mario" and "luigi1", in ~/machinetest, bootstrapped to an empty runlist. For Windows, you can replace myapp::linux with myapp::windows, but you'll need your own Windows vagrant box to do that (licensing!).
If you have an AWS account, you can spin up a machine there like this:
export CHEF_DRIVER=aws
chef-client -z simple.rbThis will create two linux boxes in the AWS account referenced by your default profile in ~/.aws/config (or your environment variables).
If you are on DigitalOcean and using the tugboat gem, you can do this:
export CHEF_DRIVER=fog:DigitalOcean
chef-client -z simple.rbIf you aren't using the tugboat gem, you can put driver and driver_options into your .chef/knife.rb file.
This will use your tugboat settings to create whatever sort of instance you normally create.
When you are done with the examples, run this to clean up:
chef-client -z destroy_all.rbChef Provisioning has two major abstractions: the machine resource, and drivers.
machine resourceYou declare what your machines do (recipes, tags, etc.) with the machine resource, the fundamental unit of Chef Provisioning. You will typically declare machine resources in a separate, OS/provisioning-independent file that declares the topology of your app--your machines and the recipes that will run on them.
The machine resources from the cluster.rb example are pretty straightforward. Here's a copy/paste:
# Database!
machine 'mario' do
recipe 'postgresql'
recipe 'mydb'
tag 'mydb_master'
end
num_webservers = 1
# Web servers!
1.upto(num_webservers) do |i|
machine "luigi#{i}" do
recipe 'apache'
recipe 'mywebapp'
end
endYou will notice the dynamic nature of the number of web servers. It's all code, your imagination is the limit :)
Drivers handle the real work of getting those abstract definitions into real, physical form. They handle the following tasks, idempotently (you can run the resource again and again and it will only create the machine once--though it may notice things are wrong and fix them!):
The driver API is separated out so that new drivers can be made with minimal effort (without having to rewrite ssh, tunneling, bootstrapping, and OS support). But to the user, they appear as a single thing, so that the machine acquisition can use its smarts to autodetect the other bits (transports, OS's, etc.).
Drivers save their data in the Chef node itself, so that they will be accessible to everyone who is using the Chef server to manage the nodes.
You can pass machine options that will be used by machine, machine_batch and machine_image to
configure the machine:
…This options hash can be supplied to either with_machine_options or directly into the machine_options
attribute.
Individual drivers will often add their own driver specific config. For example, AWS expects a :bootstrap_options hash at the same level as :convergence_options.
The ChefDK comes with a provisioner for Vagrant, an abstraction that covers VirtualBox, VMware and other Virtual Machine drivers. In docs/examples, you can run this to try it:
export CHEF_DRIVER=vagrant
export VAGRANT_DEFAULT_PROVIDER=virtualbox
chef-client -z vagrant_linux.rb simple.rbTo use with VMware, simply update the prior example to read export VAGRANT_DEFAULT_PROVIDER=vmware_fusion
This is a chef-client run, which runs multiple recipes. Chef Provisioning is nothing but resources you put in recipes.
The driver is specified on the command line. Drivers are URLs. You could use vagrant:~/vms or `fog:AWS:default:us-east-1' as driver URLs. More information here.
The vagrant_linux.rb recipe handles the physical specification of the machines and Vagrant box:
require 'chef/provisioning/vagrant_driver'
vagrant_box 'precise64' do
url 'http://files.vagrantup.com/precise64.box'
end
with_machine_options :vagrant_options => {
'vm.box' => 'precise64'
}require 'chef/provisioning/vagrant_driver' is how we bring in the vagrant_box resource.
vagrant_box makes sure a particular vagrant box exists, and lets you specify machine_options for things like port forwarding, OS definitions, and any other vagrant-isms.
Typically, you declare these in separate files from your machine resources. Chef Provisioning picks up the drivers and machine_options you have declared, and uses them to instantiate the machines you request. The actual machine definitions, in this case, are in simple.rb, and are generic--you could use them against Azure or EC2 as well:
machine 'mario' do
tag 'itsame'
endOther directives, like recipe 'apache', help you set run lists and other information about the machine.
By default, Chef Provisioning will put your boxes on the same Chef server you started chef-client with (in the case of -z, that's a local chef-zero server). Sometimes you want to put your boxes on different servers. There are a couple of ways to do that:
with_chef_local_server :chef_repo_path => '~/repo'with_chef_local_server is a generic directive that creates a chef-zero server pointed at the given repository. nodes, clients, data bags, and all data will be stored here on your provisioner machine if you do this.
You can use with_chef_server instead if you want to point at OSS, Hos
No open issues yet, or sync has not completed.