Skip to content

Instantly share code, notes, and snippets.

@fnichol
Created April 9, 2014 14:56
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save fnichol/10279877 to your computer and use it in GitHub Desktop.
Save fnichol/10279877 to your computer and use it in GitHub Desktop.
ChefConf 2014 Talk: Test Kitchen: One Year Later and the Future

Test Kitchen: One Year Later and the Future

Where

ChefConf 2014 in San Francisco

When

Wednesday, April 16, 2014 at 14:20 (Ballroom A in Future Chef track)

Abstract

After one year of active development and real world use, Test Kitchen 1.0 was released on December 1, 2013. And this is only the beginning. In this session we will cover an introduction to Test Kitchen, its ecosystem, some common (and crazy) usage patterns in the wild, and ways to extend and bend the framework to your will. If you want a good insight into where Test Kitchen is going, then this talk is for you.

@tjsoftworks
Copy link

Fletcher,

I saw your post requesting: What do I want to see in your kitchen-ci talk.... You asked for it..... Here goes...

I discovered kitchen-ci tutorial..... It is fantastic! It does not hide or dumb down most things and does a great job of focusing on learning. There is lot of magic in kitchen but the tutorial exposes most of the right things to learn about the greater ENV of workflows and tool chains of infrastrucuture-as-code and application injection into a particular infrastructure....

I was actually going thru the learnchef tutorial again.... and it was failing because of my workstation environment didn't match some sort of expectation of the builder of learnchef. In the quest to get that under control again, I ended up on the Test Kitchen tutorial page where I ended up learning way more about lots of things ( even Chef ) than on the learnchef website ..... This caused me to ponder again: Why is chef and friends is so hard to keep stable - rvm amd friends, rbenv and friends, homebrew and friends, vagrant, vagrant plug-in family, etc.... etc. not to mention getting of the focus of a project .... like an app...

Omnibus installers seemed like the way to go last year for both Chef and Vagrant but I'm thinking that this just ends up in a frozing the build state of an important piece with everything else changing around them until something breaks.

So ..... What is the key to keeping all these animals in a green pasture producing milk? What are your recommendations for those of us who love the idea of Chef and friends but pick the wrong ENV. variables, tools and toolchains? What ENV do you use? ( and why ) I am of course using ENV to represent the bigger picture which I think of as the "ideal" ENV management tools and the ENV itself. A specific example for me is rvm vs rbenv. I ended up switch from rvm to rbenv under homebrew because I got really tired of pondering: "What is broken now and is it important to fix before moving on to real work".

So .... I'd like to hear your recommendation around 0) Platform - Mac? and OSX? 1) Homebrew, 2) Homebrew tap Kasks, 3) rbenv or rvm and perhaps even 4) chef install ( omnibus vs. gem install ) and then of course Kitchen CIs role and where the boundary of Kitchen CI is and should be....

Sorry about the length of this post. Perhaps it will help someone tune the talk and or someone's listening to better the Chef ecology.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment