gistfile1.txt

A rough outline of the talk:

- live demonstration of a simple cluster of machines (application, database, mail) provisioned by Chef (10 minutes)
- quick review of the reasons why Chef is better than Puppet, capistrano, etc. for managing servers (all ruby, spoke-and-hub rather than 'pushing commands') (5 minutes)
- code examples of the previously run chef 'cookbooks' that setup services (15 minutes)
- quick intro to 'ohai', the read-only system profiling tool that Chef uses to gather information about hosts (5 minutes)
- show how we use Chef server to automatically configure Nagios monitoring  (5-10 minutes)
- Q and A

Code used in the talk would be made available on Github prior to the talk.