A significant internet outage on Tuesday, 19 November 2025, brought numerous major websites including ChatGPT and X offline, leaving visitors confronted with perplexing error messages for approximately six hours. The disruption originated not from a malicious cyber attack, but from an internal technical failure at Cloudflare, a crucial web infrastructure company that helps ensure websites remain accessible to the public.
The Technical Glitch That Paralyzed the Web
Cloudflare has since revealed that the widespread failure stemmed from a problem with a single, critical file. The company utilises a 'feature file' to keep its global network updated with the latest security threats. However, a recent change in Cloudflare's systems caused its database to insert multiple duplicate entries into this file.
This error caused the file's size to rapidly double, exceeding its predetermined maximum limit. When this oversized file was distributed across all the machines in Cloudflare's extensive network, the software tasked with reading it failed. This cascade of failures meant that a host of seemingly unconnected websites, all reliant on Cloudflare's services, were suddenly knocked offline.
From Cyber Attack Fears to a Simple Solution
Initially, the severity of the situation led to speculation, including from within Cloudflare, that the company was under a major cyber attack. After thorough investigation, the team identified the internal, and relatively simple, root cause. The issue was fully resolved around six hours after it began, restoring service to all affected platforms.
In a public post explaining the incident, Cloudflare's Chief Executive, Matthew Prince, issued a sincere apology. 'We are sorry for the impact to our customers and to the Internet in general,' he stated. He emphasised that 'Given Cloudflare's importance in the Internet ecosystem any outage of any of our systems is unacceptable.'
Lessons from a Global Internet Failure
Prince expressed the team's profound disappointment, noting that the period during which their network could not route traffic was 'deeply painful to every member of our team. We know we let you down today.' The incident underscores the immense reliance the modern internet places on core infrastructure providers. A single configuration error at this level can have immediate and far-reaching consequences for global digital connectivity, highlighting the critical need for robust system safeguards and rapid response protocols.
