Cloudflare incident on August 21, 2025

This post details a network congestion incident caused by a single customer's traffic surge overwhelming links between Cloudflare and AWS us-east-1. It explains the technical details of the congestion, including the role of BGP prefix withdrawals by AWS, saturated peering connections, and overloaded Data Center Interconnect (DCI) links. It also outlines short-term remediations like traffic deprioritization, expedited DCI upgrades, and improved BGP traffic engineering with AWS, as well as a long-term solution involving a per-customer traffic management system.