A major component of the internet’s behind-the-scenes infrastructure experienced a global disruption on Tuesday, generating widespread error messages across numerous websites.

Cloudflare — a U.S.-based company that protects millions of sites from cyberattacks — faced an unspecified issue that temporarily prevented users from accessing some of its clients’ pages. Many site operators were also unable to open their performance dashboards. Platforms such as X and OpenAI reported increased outages that coincided with Cloudflare’s service problems, according to Downdetector.

The outage was first detected at 11:48 a.m. (London time). By 2:48 p.m., Cloudflare reported that a fix had been deployed: “We believe the incident is now resolved. We are continuing to monitor to ensure all services return to normal.”

A Cloudflare spokesperson apologized “to our customers and the internet in general for letting you down today,” adding that the company would learn from the event and strengthen its systems.

Despite the disruption, experts noted that the most feared internet threat remains a sudden cascading failure in the decades-old global protocols that keep the web functioning.

While working to solve the issue, Cloudflare temporarily disabled its encryption service Warp for London-based users, warning of connection failures in the region.

Cybersecurity expert Prof. Alan Woodward of the University of Surrey described Cloudflare as “the biggest company you’ve never heard of,” acting as a crucial “gatekeeper” that filters malicious traffic, blocks distributed denial-of-service attacks, and verifies whether users are human. According to Cloudflare, its tools protect websites, apps, APIs and AI workloads while improving performance.

The company later explained that the root cause of the outage was an automatically generated configuration file that had grown beyond its intended size. This triggered a crash in the software responsible for handling traffic across several Cloudflare services.

Cloudflare emphasized that there was no evidence of a cyberattack or malicious interference. Some minor service degradation is expected as traffic surges after the outage, but operations should fully normalize within hours.

The incident follows a widespread outage at Amazon Web Services less than a month ago, which brought down thousands of websites.

“We’re seeing how few companies hold up the backbone of the internet, so when one goes down, the impact becomes immediately obvious,” Woodward said. Although the exact cause is not fully known, he added that a cyberattack is unlikely, since systems of Cloudflare’s scale rarely rely on a single point of failure.