HOW WHEN AWS WAS DOWN, WE WERE NOT

Summary from the TLDR Dev newsletter

Authress' auth solution remained operational during a recent AWS us-east-1 outage that impacted major companies. This was thanks to avoiding unreliable third-party components, implementing DNS failover with custom health checks, and using edge compute via CloudFront and Lambda@Edge. It also used incremental rollouts to reduce the impact of application failures, a web application firewall (WAF) for handling malicious requests, and prioritized customer support for issues outside of automated detection.