Table of Contents
Why Did Half the Internet Break? A Simple Guide to What Went Wrong with Cloudflare.
On November 18, 2025, a big piece of the internet seemed to just stop working. Many popular websites and online services suddenly became unreachable for people all over the world. If you tried to visit sites like X (formerly Twitter), ChatGPT, Spotify, or Canva, you likely saw an error message. This happened because a company called Cloudflare had a major problem. Cloudflare helps about one-fifth of all websites run smoothly and safely, so when it has an issue, the effects are felt far and wide.
What Went Wrong?
The problem started around 11:20 in the morning, Coordinated Universal Time (UTC). It was not a cyber attack. Instead, the outage was caused by a hidden software bug within Cloudflare’s own systems. This bug was accidentally triggered by a routine update. This internal error caused a chain reaction that disrupted services across Cloudflare’s global network. For several hours, users experienced widespread “Internal Server Error” messages, and even Cloudflare’s own dashboard for its customers was not working correctly.
Which Websites Were Affected?
The outage had a massive impact, affecting a wide range of online services that people use every day. The list of affected platforms included:
- Social Media: X (formerly Twitter) was down for many users.
- AI Tools: Popular services like ChatGPT were inaccessible.
- Business and Marketing: SEO tools like Ahrefs were not working, which stalled workflows for many marketing professionals.
- Entertainment: Streaming services like Spotify and online games such as League of Legends were also hit.
- Other Services: The disruption also affected e-commerce stores, news websites, and even outage tracking sites like DownDetector.
Fixing the Problem
Cloudflare’s engineers quickly began investigating the issue as soon as it was detected. The company posted regular updates on its status page to keep the public informed, though the page itself was sometimes hard to access due to the outage. The problem was officially identified around 1:09 PM UTC [, web-24]. A fix was put in place, and by about 2:42 PM UTC, services began to return to normal. The entire incident lasted for several hours, from the initial detection to full resolution. Cloudflare’s CEO, Matthew Prince, apologized for the disruption and promised a detailed explanation of what happened.
A Wake-Up Call for the Web
This event shows how much the internet relies on a small number of very large companies. When a major service like Cloudflare has a problem, it can take down a huge portion of the web with it. Experts point to this as a weakness in the internet’s structure. It raises important questions about the need for more diversity in web infrastructure to prevent such widespread disruptions in the future. The outage serves as a reminder that even the most reliable systems can fail and that having backup plans is crucial for any online business.