We Now Know Why the DNS Failed
AWS’s October 19, 2025 outage—which knocked DNS offline for over 24 hours—finally has a public Root Cause Analysis (https://aws.amazon.com/message/101925/). The breakdown started when a change in DynamoDB’s service discovery caused it to lose its DNS records, preventing EC2 instances from reaching critical endpoints.
That initial hiccup rippled outward: EC2 instances began throwing metadata-lookup errors, and Network Load Balancers, still pointing at stale targets, dropped traffic into a black hole. Together, these failures locked down large swaths of AWS for more than a day.
Watch on YouTube
Top comments (0)