On the evening of February 20, 2026, millions of internet users across the globe found themselves unexpectedly cut off from some of the web’s most popular platforms. From food delivery to online betting, the digital world ground to a halt for several hours, all because of a technical meltdown at Cloudflare, a company that quietly powers a significant chunk of the internet’s infrastructure.
It all began shortly after 6:30 PM GMT (1:30 PM EST), when users in both the United States and the United Kingdom started reporting widespread outages on Downdetector, a site that tracks real-time problems and outages for various services. Bet365, Uber Eats, Workday, and a slew of other major websites—including Wikipedia, Steam, Sky Bet, and even Amazon Web Services—were suddenly inaccessible. The impact was immediate and far-reaching, with nearly 4,000 outage reports for Uber Eats alone logged within just ten minutes, according to Downdetector data cited by Hindustan Times and Express.
As the evening wore on, the numbers kept climbing. By 6:57 PM GMT, Bet365 had racked up 5,247 reported issues in the UK, while Cloudflare itself had more than 217 complaints by 7:02 PM. In the U.S., the story was much the same: Uber Eats, Workday, and Bet365 all saw hundreds to thousands of users unable to access their platforms. For many, the disruption wasn’t just a minor annoyance—it was a real headache, especially for those who rely on these services for work, food, or entertainment.
Social media quickly filled with frustrated posts. One hungry customer vented, “Imagine being hungry and Uber Eats app goes down.” Another quipped, “So my Uber Eats app just stopped working as my order is on its way.” The mood ranged from bemused to exasperated, with users questioning the reliability of services they had come to depend on—especially when orders and bets hung in the balance.
Behind the scenes, the culprit was soon identified: Cloudflare, a U.S.-based internet infrastructure company whose services are woven into the fabric of the modern web. Cloudflare operates as both a content delivery network (CDN) and a distributed domain name server (DNS). In plain English, that means it helps websites load faster, protects them from cyberattacks, and keeps them running smoothly even during traffic spikes. According to Express and iNews, about 20% of all websites worldwide use Cloudflare in some capacity—an astonishing figure that underscores just how central the company is to the internet’s day-to-day functioning.
When Cloudflare stumbles, the ripple effect is enormous. As Jake Moore, global cybersecurity adviser at ESET, explained to Express, “If a major provider like Cloudflare goes down for any reason, thousands of websites instantly become unreachable. The problems often lie with the fact we are using an old network to direct internet users around the world to websites but it simply highlights there is one huge single point of failure in this legacy design.”
Cloudflare itself was quick to acknowledge the trouble. In a statement posted to its website and shared widely by media outlets, the company said: “We are investigating issues with our services and/or network. Users may experience errors or timeouts reaching Cloudflare’s network or services.” The technical details were sparse, but the message was clear: something fundamental had gone awry, and it would take time to resolve.
Other affected companies also scrambled to reassure their customers. Bet365, a major online betting platform, posted on X (formerly Twitter): “We’re aware of an issue with our Website/App, and our Technical Team are working to resolve it as soon as possible. We apologise for the inconvenience.” The message was echoed by their tech support team, who reiterated that they were “working to resolve” the issue as quickly as possible.
For Uber Eats, the outage was particularly acute. Nearly 4,000 users reported problems by 6:40 PM, with many taking to social media to complain about missed deliveries and wasted subscriptions. “What’s the point of having Uber One when Uber Eats is so unreliable?” one user asked, referencing the company’s premium service. Another lamented, “We’re wasting our money!!” For those caught in the middle of an order, the outage was more than just a technical hiccup—it was dinner plans derailed.
By 8:07 PM GMT, Cloudflare’s troubles were still ongoing, with 3,998 people having filed complaints on Downdetector. The outage persisted for several hours, with some services gradually returning to normal as Cloudflare’s engineers worked behind the scenes. According to Hindustan Times, by 2:15 PM EST (7:15 PM GMT), Uber Eats, Workday, and Bet365 appeared to be functioning normally again, though the disruption had already left a mark on the evening for thousands.
Interestingly, this wasn’t Cloudflare’s first brush with such chaos. The company had suffered similar outages in November and December of 2025, each time casting a spotlight on the vulnerabilities of the internet’s underlying architecture. As Moore pointed out, “It simply highlights there is one huge single point of failure in this legacy design.” In other words, the internet’s remarkable speed and convenience are built atop a system with very real—and sometimes fragile—foundations.
So what exactly went wrong? While Cloudflare’s official statements were short on specifics, experts suggest that the interconnected nature of content delivery and DNS services means that even a minor technical hiccup can have outsized consequences. When the proxy service connecting users to websites fails, visitors are greeted with 500 or 5xx errors, effectively locking them out of vast swathes of the web. The fact that so many high-traffic platforms—from gaming giants like Steam and Dota 2 to essential business tools like Microsoft 365 and Workday—were affected underscores just how much rides on Cloudflare’s infrastructure.
For businesses, the fallout isn’t just about customer complaints. Every minute of downtime can mean lost revenue, damaged reputations, and a flood of support tickets. For end-users, it’s a reminder of just how reliant we’ve all become on a handful of behind-the-scenes players to keep the digital world humming along.
As the dust settles, the February 20 outage stands as a wake-up call for both companies and consumers. The internet may feel limitless and robust, but its backbone is more fragile than many realize. As Moore’s comments suggest, until the legacy design is overhauled—or at least diversified—the risk of another widespread blackout remains all too real.
For now, services are back online and life has returned to normal for most. But the disruption lingers in memory, a stark reminder that even the internet’s invisible giants can stumble—and when they do, the whole world feels it.