Configuration Version

Configuration versions are the mechanism by which Vagrant 1.1+ is able to remain backwards compatible with Vagrant 1.0.x Vagrantfiles, while introducing dramatically new features and configuration options.

If you run vagrant init today, the Vagrantfile will be in roughly the following format:

Vagrant.configure("2") do |config|
  # ...
end

The "2" in the first line above represents the version of the configuration object config that will be used for configuration for that block (the section between the do and the end). This object can be very different from version to version.

Currently, there are only two supported versions: "1" and "2". Version 1 represents the configuration from Vagrant 1.0.x. "2" represents the configuration for 1.1+ leading up to 2.0.x.

When loading Vagrantfiles, Vagrant uses the proper configuration object for each version, and properly merges them, just like any other configuration.

The important thing to understand as a general user of Vagrant is that within a single configuration section, only a single version can be used. You cannot use the new config.vm.provider configurations in a version 1 configuration section. Likewise, config.vm.forward_port will not work in a version 2 configuration section (it was renamed).

If you want, you can mix and match multiple configuration versions in the same Vagrantfile. This is useful if you found some useful configuration snippet or something that you want to use. Example:

Vagrant.configure("1") do |config|
  # v1 configs...
end

Vagrant.configure("2") do |config|
  # v2 configs...
end

What is Vagrant::Config.run? You may see this in Vagrantfiles. This was actually how Vagrant 1.0.x did configuration. In Vagrant 1.1+, this is synonymous with Vagrant.configure("1").

© 2010–2018 Mitchell Hashimoto
Licensed under the MPL 2.0 License.
https://www.vagrantup.com/docs/vagrantfile/version.html