New Delhi | November 19, 2025 | SKY LINK TIMES
Cloudflare CEO Reveals Truth Behind Global Outage:
A massive global internet outage that disrupted platforms including X, ChatGPT, Canva, Discord, and dozens of major websites was caused by an internal configuration mistake, not a cyberattack, Cloudflare CEO Matthew Prince confirmed in a detailed postmortem.
The outage, which lasted nearly three hours, triggered widespread confusion as users speculated about hacking attempts or large-scale DDoS attacks. Cloudflare—one of the world’s largest internet infrastructure providers—later clarified that the root cause was entirely internal.
Table of Contents
Faulty Permission Change Triggered Chain Reaction
According to Prince, the issue began during a routine update involving permissions on a ClickHouse database cluster. The update was intended to improve internal access to data, but a faulty query inadvertently pulled in far more information than expected.
This created a bloated “feature file” used by Cloudflare’s Bot Management system. The file is refreshed across Cloudflare’s global network every five minutes. Once the file exceeded preset software limits, it caused routing software at the network edge to crash.
Network Entered a Five-Minute Crash-Reboot Cycle
The situation quickly worsened due to a mismatch in updated systems.
Only parts of the database cluster had applied the new update—meaning some nodes generated correct feature files while others produced corrupted ones. Every five minutes, Cloudflare’s network would receive one version or the other, causing:
Temporary recovery when the valid file propagated
Sudden failure when the corrupted file spread
This unstable cycle continued from 11:20 UTC to around 14:20 UTC, resulting in intermittent global outages.
Prince admitted the team initially misdiagnosed the symptoms as a massive DDoS attack because of the unusual traffic patterns triggered by the crash loops.
Also Read:https://skylinktimes.in/gati-shakti-vishwavidyalaya-drdo-sign-mou/
Engineers Contained the File and Restored Systems
Cloudflare engineers eventually halted the spread of the faulty file, replaced it with a known-good version, and manually restarted affected systems across the network.
By 17:06 UTC, Cloudflare declared that the issue was fully resolved, calling it the company’s most serious outage since 2019.
Prince publicly apologised for the disruption across thousands of businesses and millions of users who rely on Cloudflare’s infrastructure for content delivery, DNS, and security.
Cloudflare Announces New Safeguards
To prevent similar incidents, Cloudflare plans to introduce:
Stricter size limits on critical network files
A global kill switch to instantly stop propagation of unstable updates
Deeper architectural reviews to evaluate how core systems can fail
Improved monitoring to detect unusual file changes earlier
Prince emphasized that no malicious activity was involved.
“This was entirely our mistake,” he said, calling the event a critical lesson in infrastructure safety and resilience.
For More Info Stay Tuned: https://skylinktimes.in
