PANews reported on November 19th that, according to an analysis report on the large-scale outage on November 18th released on the Cloudflare official blog, this most severe outage since 2019 stemmed from an incorrect configuration file generated after database permission adjustments, causing an anomaly in the core proxy system and affecting multiple products including CDN, security services, Workers KV, Turnstile, and Access. The team initially misjudged it as a DDoS attack and ultimately fixed the issue by rolling back old files. All services were restored by 01:06 AM Beijing time on November 19th. Cloudflare stated that such outages are unacceptable and will accelerate system resilience upgrades.