The Power and Creativity of Retrospectives and Repair Items
Speaker: Nida Farrukh, Switzerland
Outages suck. But at least we learn from them! One of the great things about outages is that they’re a compelling argument for change. And a compelling demonstration of what needs changing. And they even bring their own stats on what we stand to lose if we don’t change!
So a retrospective is a neat little package argument for doing the repair items it suggests. Don’t squander this opportunity on repair items that just paper over problems. Really dig in - you’ve been given a window into how your system functions (and mis-functions), which you can use to revisit prior architecture decisions and inform your future design.
Take advantage of the data that’s fallen into your lap! Champion incisive change, one repair item at a time!
Nida Farrukh is a Zurich-based singer and chocolate enthusiast who has been working as a Site Reliability Engineer for the last decade and more.
Formerly at Google, now at Microsoft, she makes her living figuring out all the ways systems can fail - ideally before they actually do. In chasing that ideal, she's specialised in automated monitoring systems; but because of that ideal's elusory nature, she can also claim extensive experience in incident response and follow-up.