A major Cloudflare outage on Tuesday triggered widespread service failures across some of the world’s most heavily used platforms, including X (formerly Twitter), OpenAI’s ChatGPT, Spotify and Canva, among others, leaving millions of users unable to log in, refresh feeds or access paid features for more than an hour.
The disruption began shortly after 11:20 UTC, according to Cloudflare, when the company detected a “spike in unusual traffic” hitting one of its core services. The surge caused traffic routed through Cloudflare’s global network to throw widespread 500 errors, taking down sites that depend on the company’s infrastructure to stay online. At the peak of the outage, even Downdetector, a commonly used monitoring site, stopped functioning because it also runs on Cloudflare’s systems.
Cloudflare, which handles roughly a fifth of global web traffic, acknowledged an “internal service degradation” and said some services had been temporarily disabled while engineers worked to contain the issue. “We are focused on restoring service,” the company said in its first advisory, issued nearly 90 minutes after user complaints began mounting. In a later update, it confirmed that Access and WARP services were recovering, though error rates remained elevated during remediation.
Platforms or websites hit by this outage included X (formerly Twitter), OpenAI/ChatGPT, Canva, Spotify, Archive of Our Own, DePop, Perplexity, Gemini, League of Legends, Ikea, Truth Social, Indeed, Character.ai, Zoom and Grindr, among others. Users across regions reported being unable to sign in, refresh timelines or load content, with some services briefly returning only to fail again.
The outage highlighted the key role Cloudflare plays in keeping the modern internet functioning. The U.S.-based company operates a vast network across more than 330 cities in over 120 countries, shielding websites from traffic overloads and cyberattacks while accelerating content delivery. Its systems process an average of 81 million HTTP requests per second and more than 42 million DNS requests per second — volumes that make any performance disturbance ripple across the digital ecosystem. As the outrage unfolded, Cloudflare’s shares dipped about 5 percent in pre-market trading.
The incident arrives weeks after an AWS disruption caused chaos across thousands of apps, renewing concerns about the risks of centralizing critical internet infrastructure in the hands of a few major providers. While Cloudflare reported that its services were largely stabilizing by late evening, it warned that some customers could continue to see elevated error rates as systems fully recovered.
Share your thoughts with us on X/Twitter or Instagram (@lyricalmuseblog), and visit Lyrical Muse for more real-time tech and culture stories.
Leave a Reply