Cloudflare’s network faltered for hours on Thursday, sidelining prominent websites and slowing internet access for millions. The disruption hit hardest in the United Kingdom, where users flooded social media with complaints about inaccessible sites and sluggish loading times. Reports poured in from the United States and other regions, painting a picture of global digital gridlock.

Users trying to reach Microsoft services, Outlook email and Uber’s platform encountered error pages or endless waits. Dozens of other sites reliant on Cloudflare’s infrastructure buckled under the failure. ‘Pages wouldn’t load, apps crashed—total chaos,’ one UK customer posted online.

Cloudflare, which shields and accelerates traffic for millions of websites, saw its systems glitch without warning. The company acknowledged the issue on its status page, pinning it on a technical fault in its core routing services. Engineers scrambled to reroute traffic and patch the problem.

By late afternoon UTC, Cloudflare reported partial recovery, with most services back online. Full restoration stretched into the evening, according to the company’s updates. Officials at the firm declined to specify the root cause but promised a detailed post-mortem.

The outage exposed the fragility of the modern web. Cloudflare processes a hefty slice of global internet traffic—enough that a single hiccup ripples far. Experts tracking the event pointed to past incidents, like a 2022 glitch that slowed Discord and League of Legends. This time, the impact felt broader, snaring enterprise tools and consumer apps alike.

UK regulators took note quickly. The National Cyber Security Centre urged affected organizations to monitor for fallout. In the US, Federal Trade Commission watchers stayed silent so far, though businesses tallied losses from stalled operations.

Cloudflare customers, from small blogs to Fortune 500 giants, depend on its defenses against attacks and its boosts for speed. Thursday’s mess reminded everyone of that reliance. One analyst called it a ‘wake-up call’ for diversifying infrastructure providers.

As services stabilized, user frustration lingered. Downdetector charts spiked with reports peaking near 100,000 in the UK alone. Cloudflare tweeted assurances: teams worked around the clock to prevent repeats.

The episode unfolded amid steady web traffic growth. No links surfaced to cyberattacks—officials described it as an internal glitch. Still, it disrupted remote workers checking Outlook, riders summoning Uber and developers on Microsoft platforms.

Cloudflare’s swift response averted a longer crisis. By evening, dashboards showed green across most regions. The company pledged transparency in a follow-up report, expected within days.