On
November 18, 2025, Cloudflare suffered a major outage that disrupted traffic and caused “500 Internal Server Error” responses across many of its core services, affecting high-profile platforms like X (formerly Twitter), ChatGPT, Canva and others.
The root of the failure was not malicious: a routine configuration permission change triggered a latent bug in Cloudflare’s bot-management system, producing a “feature file” that grew well beyond its expected size and crashed traffic-routing software. Cloudflare rolled back to a stable version of the file, restored normal routing by around 14:30 UTC, and declared full recovery by 17:06 UTC. The company’s CTO publicly apologized, calling the incident unacceptable, and pledged to improve resilience by strengthening configuration ingestion and adding better fail-safes.
No comments:
Post a Comment