The entire internet just crashed... again

Key Concepts

• Cloudflare Outage: A widespread internet disruption caused by a failure at Cloudflare, a major cloud infrastructure provider.
• DNS Hosting: The service that translates human-readable domain names (like google.com) into machine-readable IP addresses.
• DDoS Protection: Measures taken to defend against Distributed Denial of Service attacks, which aim to overwhelm a server with traffic.
• Bot Mitigation: Systems designed to identify and block automated bot traffic from accessing websites.
• Internal Service Degradation: Cloudflare's term for a widespread internal system failure.
• Configuration Change: A modification to the settings or parameters of a software system.
• Latent Bug: A flaw in software that is not immediately apparent and may only manifest under specific conditions.
• Configuration File: A file that stores settings and parameters for a software application.
• Centralized Infrastructure: The reliance on a few large providers for essential internet services, leading to significant impact when they fail.
• Sentry: A platform for application monitoring and error tracking, sponsoring the video and offering an AI code reviewer.

Cloudflare's Widespread Internet Outage

On November 19th, 2025, the internet experienced a significant disruption affecting millions of websites. This outage was caused by a failure at Cloudflare, a critical provider of DNS hosting, DDoS protection, and bot mitigation services. The incident began around 6:00 a.m. Eastern time, with Cloudflare reporting "internal service degradation."

Impact of the Outage

The outage had a broad impact, rendering numerous platforms inaccessible. This included:

• X (formerly Twitter): Users were unable to log in or access the platform.
• OpenAI's ChatGPT: The AI chatbot service was unavailable.
• Down Detector: Ironically, the website used to report outages was itself affected, displaying Cloudflare error pages.
• League of Legends: Players experienced connectivity issues with game servers.

The severity of the outage was such that even the Cloudflare status page struggled to load, reportedly displaying only basic HTML.

Root Cause Analysis

Contrary to initial speculation of a supply chain attack, a misconfigured DNS setting, or a conspiracy, Cloudflare's CTO stated that the issue stemmed from a "latent bug in a service underpinning our bot mitigation capability." This bug was triggered after a "routine configuration change."

The specific root cause was identified as a configuration file used to manage threat traffic. This file grew beyond its expected size, containing an excessive number of entries. This bloat caused a crash in the software system responsible for handling traffic across various Cloudflare services. Essentially, a system designed to protect the internet from threats became so overloaded by its own configuration that it caused a widespread failure.

Broader Implications: The Centralization of the Internet

This incident serves as a stark reminder that the internet is not a fully distributed, neutral system. Instead, it relies heavily on a few large, centralized infrastructure providers. When one of these providers experiences a catastrophic failure, the impact is massive and far-reaching.

Sponsor Spotlight: Sentry

The video highlights Sentry, a sponsor that offers solutions for preventing catastrophic failures in software. Sentry has launched a new AI code reviewer that scans pull requests for bugs before they reach production. This tool automatically analyzes code changes, provides detailed bug reports, suggests fixes, and offers AI prompts for further assistance. It can identify both minor issues like typos and logic errors, as well as more complex problems by leveraging existing error data captured by Sentry. Sentry is currently offering a free open beta at century.io/fireship.

Conclusion

The Cloudflare outage underscores the fragility of our current internet infrastructure, which is heavily dependent on a few key players. The incident, caused by an internal software bug triggered by an oversized configuration file, demonstrates how even systems designed for security can inadvertently lead to widespread disruption. This event reinforces the importance of robust error detection and prevention mechanisms, such as those offered by Sentry, for developers aiming to build more resilient applications.