Global network provider Cloudflare has published a statement via its official blog concerning the major six-hour disruption yesterday on 18 November 2025, which impacted websites and services worldwide. The incident began at 7:20pm local time and resulted in significant traffic failures across its network. Among those affected included our own website and forum, as well as social media platform X, OpenAI’s ChatGPT and many more.
Cloudflare says the problem was not caused by a cyber-attack or malicious activity. Instead, it was triggered by a change to its internal database permissions which caused an unexpected duplication in a configuration file used by its Bot Management system. That file then propagated across the network, exceeded built-in limits and led to failures in traffic routing.
The company revealed that core traffic had largely returned to normal by around 10:30pm, and full service restoration was achieved at 1:06am on 19 November 2025. “On behalf of the entire team at Cloudflare, I would like to apologise for the pain we caused the Internet today,” said company co-founder and CEO Matthew Prince in the post.
Cloudflare states it is taking various steps to prevent a recurrence. These include hardening the ingestion and distribution of configuration files, enabling global kill-switches for certain features, eliminating the risk of error reports overwhelming system resources, and reviewing the failure-modes of its core proxy modules.
The service also acknowledged that this was its most significant outage since 2019 and stressed that it was unacceptable for its network to stop routing at any point. Cloudflare notes that while most systems are back to normal, some customers may still notice minor residual effects as it completes its remaining mitigation and resilience work.
(Source: Cloudflare [official blog])
