Chapter 1. Bad ’Form!
It happens all the time: an organization adopted Terraform and has been using it, fairly successfully, for some months or years, when suddenly, no one can push a change. The automation jobs that apply Terraform code are either failing or reporting unexpected changes to the configuration of running instances, or both. Pretty much all new features are blocked until the issue is resolved and the jobs are clean and green. You get your team(s) together on multihour long virtual calls to swarm on the issue (despite only two or three engineers really understanding the most complex portions of the codebase, but why should they be punished have all the fun?). Hours turn into days. Meanwhile, other meetings are called (primarily among nontechnical, unfamiliar parties) to discuss why and how and when and why (again) this happened. A week goes by. People have spent nights, weekends, and lunches trying to figure out what’s going on. People are fearful of losing their jobs. Then, suddenly, almost miraculously, the issue is resolved by a single line of configuration code that no one can remember being there in the first place (“Oh, yeah, here’s the commit where it was removed...”). Or worse, the problem seems to magically resolve itself.
No one wants to be in this scenario. We all want to think we’re immune. We overlook the warning signs and even the shouted alarms by that one pesky, annoying guy who keeps bringing up problems until it’s far too late: budgets are blown, ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access