Articles
The incident: a formula 1 car hit the side barrier just over 20 minutes before the race was about to start. The team sprang into action with an incredibly calm, orderly and speedy incident response to replace the damaged parts faster than they ever have before.
This article is a great analysis, and there’s also an excellent 8-minute video that I highly recommend. Listen to the way the sporting director and everyone else communicates so calmly. It’s a rare treat to get video footage of a production incident like this.
Chris Evans — incident.io
The underlying components become the cattle, and the services become the new Pet that you tend to with your utmost care.
Piyush Verma — Last9
AWS posted these example/template incident response playbooks for customers to use in their incident response process.
Aws
A list with descriptions of all DNS record types, even the obscure ones. Tag yourself, I’m HIP
.
Jan Schaumann
This one includes a useful set of questions to prompt you as you develop your incident response and classification process.
Hollie Whitehead — xMatters
The author of this article shows us how they communicate actively, perform incident retrospectives, and even discuss “near misses” and normal work in order to better learn how their system works — all skills that apply directly to SRE.
Jason Koppe — Learning From Incidents
Although the fundamental concepts of site reliability engineering are the same in any environment, SREs must adapt practices to different technologies, like microservices.
JJ Tang — Rootly
This one uses Akamai’s incident report from their July 22 major outage as a jumping-off point to discuss openness in incident reports. The text of Akamai’s incident report is included in full.
Geoff Huston — CircleID
Drawing from the “normalization of deviance” concept introduced in the Challenger disaster study [Diane Vaughan], this article explores the idea of studying your organization culture to catch problems early, rather than waiting to respond after they happen.
Stephen Scott
This episode of the StaffEng Podcast is an interview with Lorin Hochstein, whose writings I’ve featured here numerous times. My favorite part of this episode is when they talk about doing incident analysis for near misses. One of the hosts points out that it’s much easier for folks to talk about what happened, because there was no incident so they’re not worried about being blamed.
David Noël-Romas and Alex Kessinger– StaffEng Podcast
Outages
- Let’s Encrypt
- Snapchat
- Wikipedia
- To fact-check this one, I looked at their grafana dashboard. Neat!
- Netflix
- Venmo
- Blackboard Learn
- eBay