Chef architecture has three main COMPONENTS, Chef server, Chef node, and Chef workstation. At the workstation, code for provisioning and managing the infrastructure is created and tested. This code is implemented as recipes and a collection of recipes is known as a cookbook. Once the cookbook is ready, it is pushed to the Chef server from the workstation. We use the “knife” command-line tool or Berkshelf” to upload the cookbook. Chef Server TYPICALLY acts as a hub for the infrastructure code. Chef node could be any system that is managed by Chef. It could be an ACTUAL or virtual or cloud machine.

Chef nodes communicate with the Chef server using an application called “chef-client”.Chef architecture works on pull configuration that means we don’t have to perform any ACTIONS on the Chef server to push the configurations to Chef nodes.“The chef-client” pulls down all the data needed for the Chef node to meet the desired criteria. This is also referred to as Chef run. During Chef run, a tool called OHAI collects the current status of the attributes of the node.