GitLab · January 31, 2017 · critical
When a Database Engineer Accidentally Deleted the Production Database
An engineer ran rm -rf on production. Five backup strategies all failed.
Read → 12 minWhere's the Postmortem?
Curated incident stories from the best engineering teams. Learn from failures you didn't have to live through.
No postmortems in this category yet.
Cloudflare · major
When Someone Else Hijacked 1.1.1.1
A Brazilian ISP announced a more-specific route for 1.1.1.1, hijacking DNS traffic across 300 networks in 70 countries.
Knight Capital · critical
How Dead Code Caused a $440 Million Loss in 45 Minutes
A botched deployment activated 9-year-old dead code that bought high and sold low for 45 minutes straight.
Heroku · critical
When a Stolen Token in an Archived Repo Compromised an Entire Platform
Roblox · critical
73 Hours Down: When Service Discovery Took Down an Entire Platform
Meta · critical
The Day Facebook Disappeared from the Internet
Fastly · critical
One Customer Setting Broke the Internet for an Hour
Slack · major
When Everyone Came Back from Holiday and Slack Couldn't Handle It
GitHub · critical
43 Seconds of Network Partition, 24 Hours of Degraded Service
Amazon Web Services · critical