“The root cause was an internal configuration error and not the result of an attack or a BGP hijack.”
Cloudflare’s 1.1.1.1 DNS resolver went down globally for 62 minutes because a misconfiguration from June 6 silently linked the resolver’s IP prefixes to a pre-production service. That dormant landmine sat there for five weeks before a routine configuration change triggered it, yanking the resolver’s routes off the internet worldwide. Anyone using 1.1.1.1 as their DNS server lost access to everything. The postmortem blames legacy systems lacking staged deployment, which is a polite way of saying they pushed config changes globally without a canary and got exactly what you’d expect.