A bug in Cloudflare's infrastructure shook the Internet
13:48, 19.11.2025
On November 18, a large-scale outage occurred, for which Cloudflare's CTO has already apologized. It is reported that the outage was caused by a hidden bug in the service, triggered by a standard configuration change. So the problem was caused by internal bugs and was not the result of a cyberattack.
The internal deterioration of services began at approximately 11:48 UTC when users from different countries were unable to access certain sites on Cloudflare's infrastructure. There were also outages in WARP and Access services.
Causes of the problem
The company determined that the bug was caused by the bot traffic protection module. During a standard configuration update, an error was activated in the system. This process triggered the degradation of internal network components, routes, and CDN layers, causing a massive effect. A patch was implemented approximately three hours later. After that, some services began to recover, but others needed more time.
Cloudflare handles a large percentage of web traffic, so malfunctions can lead to significant risks in the future. That is why using a centralized digital infrastructure is not the best option.
Cloudflare is preparing a full technical report that will outline the likely causes of the malfunctions and a plan to prevent them.