What’s so great about this incident write-up is the way that entrenched mental models hampered the incident response. There’s so much to learn here.
Ray Ashman — Mailchimp
The parallels between this and the Mailchimp article are striking.
This includes a review of the four golden signals and presents three areas to go further.
JJ Tang — Rootly
This one thoughtfully discusses why “root cause” is a flawed concept, approaching the idea from multiple directions.
Check it out, a new SRE conference! This one’s virtual and the CFP is open until October 1.
Robert Barron — IBM
To be clear, this article is about static dashboards that just contain pre-set graphs of specific metrics.
every dashboard is an answer to some long-forgotten question
Public incident posts give us useful insight into how companies analyze their incidents, but it’s important to remember that they’re almost never the same as internal incident write-ups.
John Allspaw — Adaptive Capacity Labs
In this incident from July 7, front-line routing hosts exceeded their file descriptor limits, causing requests to be delayed and dropped.
.io, assigned to the British Indian Ocean Territory is almost exclusively used by annoying startups for content completely unrelated to the islands.
Remember, it’s all fun and games until the random country you’ve attached your business to has an outage in their TLD DNS infrastructure.
If you’re curious about just what a columnar data store is like I was, this article is a good introduction.
Alex Vondrak — Honeycomb