On November 18, 2025, starting at approximately 11:20 UTC, Cloudflare experienced a significant disruption in its network services, impacting many websites and platforms that rely on its infrastructure. This outage manifested as error pages and login failures for users trying to access affected services, causing widespread interruptions across the internet.
What Caused the Outage?
The root cause was traced to a change in database permissions that triggered the creation of an oversized configuration file used by Cloudflare’s Bot Management system. This file grew unexpectedly large, causing system errors that cascaded through Cloudflare’s core network proxy. The errors led to failed web requests returning server error codes (HTTP 5xx) and authentication issues across multiple services.
Scope and Impact
The outage affected a broad range of notable services and platforms, including ChatGPT, X (formerly Twitter), Shopify, Dropbox, and several public transit systems, such as New Jersey Transit and France’s SNCF railway. Users faced difficulties logging in, accessing websites, and utilizing essential digital services. Additionally, Cloudflare’s internal dashboard and control systems experienced degraded performance.
How Cloudflare Responded
Initially, Cloudflare’s team suspected a large-scale denial-of-service attack but quickly shifted focus once internal diagnostics pointed to the faulty configuration file. Engineers halted the propagation of the problematic file and restored the previous stable configuration by 14:30 UTC, gradually bringing services back online. By 17:06 UTC, Cloudflare had fully resolved the incident, with all systems functioning normally.
Preventative Measures Going Forward
Cloudflare acknowledged the severity of the event and committed to strengthening its system resilience. Planned improvements include enhanced validation of configuration data, global kill switches for critical features, and safeguards to prevent similar failures in core network modules.
Takeaway
This outage highlights the critical role of infrastructure providers in maintaining internet stability and the wide-reaching effects of issues within such centralized systems. Cloudflare’s swift response mitigated a potentially more prolonged disruption, and ongoing efforts aim to prevent similar incidents in the future.
