Cloudflare suffered a software crash that disrupted services for about three hours, briefly taking down sites such as X, ChatGPT and Spotify. The company traced the failure to an auto-generated configuration file that became too large and crashed the traffic-handling software. Cloudflare's CTO publicly apologized and said there is no evidence the outage was caused by an attack. A detailed post-mortem will be published and the company has pledged improvements.
Cloudflare: Oversized Configuration File Caused Software Crash — X, ChatGPT and Others Disrupted
Cloudflare suffered a software crash that disrupted services for about three hours, briefly taking down sites such as X, ChatGPT and Spotify. The company traced the failure to an auto-generated configuration file that became too large and crashed the traffic-handling software. Cloudflare's CTO publicly apologized and said there is no evidence the outage was caused by an attack. A detailed post-mortem will be published and the company has pledged improvements.
05:30 AM, 11/19/2025Technology
Cloudflare experienced a software crash on Tuesday morning that disrupted access to major websites — including X, ChatGPT and Spotify — for roughly three hours. The company issued a public apology and said it found no evidence the outage was caused by an attack or malicious activity.
Cloudflare's engineers first observed "unusual traffic" to one of its services at 6:20 a.m. ET, and a status update about internal service degradation appeared about 30 minutes later. The outage persisted until approximately 9:30 a.m. ET, with some services returning intermittently before stabilizing.
Root cause and impact
After investigation, Cloudflare traced the incident to an automatically generated configuration file used to manage threat-related traffic. That file "grew beyond an expected size of entries and triggered a crash in the software system that handles traffic for a number of Cloudflare's services." The crash interrupted traffic handling across multiple customers and platforms that rely on Cloudflare's infrastructure.
"I won't mince words: earlier today we failed our customers and the broader internet," wrote Dane Knecht, Cloudflare's chief technology officer.
The outage affected millions of users temporarily, with some sites coming back online only to experience further interruptions before full recovery. Cloudflare said it will publish a detailed post-mortem on its blog explaining what went wrong and what steps it will take to prevent recurrence.
Context and industry implications
Cloudflare provides content delivery, security and traffic-management services to millions of websites. When a major provider experiences an outage, the effects can ripple widely because many sites and apps depend on a small number of shared infrastructure providers. This incident follows other high-profile cloud outages this year, including a recent Amazon Web Services disruption and earlier incidents affecting several cloud platforms.
Cloudflare noted that scheduled maintenance was underway at some data centers during the outage window, including locations in Atlanta and Los Angeles, but emphasized that its investigation found no indication of malicious activity. The company pledged to learn from the event and improve its systems and safeguards.