Systems returning to normal after global Cloudflare outage hits internet users

A massive global outage at web-infrastructure giant Cloudflare caused widespread disruption to major online platforms on Tuesday.

It knocked services such as X, ChatGPT, and numerous South African websites offline for hours before the company began restoring traffic late in the afternoon.

Cloudflare, whose network carries roughly a fifth of global web traffic and shields websites from traffic surges and cyberattacks,first detected “internal service degradation” at around 13:30 SAST.

The disruption rapidly cascaded across the internet, overwhelming websites and apps that rely on the company’s infrastructure to stay online.

“We saw a spike in unusual traffic to one of Cloudflare’s services. That caused some traffic passing through Cloudflare’s network to experience errors,” the company said in a statement.

“We are all hands on deck to make sure all traffic is served without errors.”

The outage affected everything from AI platforms to news websites. ChatGPT, Anthropic’s Claude, and X experienced failures ranging from blank feeds to an inability to post content.

Downdetector, the outage-tracking platform, was itself impacted, though it still recorded nearly 5,000 Cloudflare-related problem reports at the peak shortly after the disruption began.

In South Africa, several of the country’s largest online outlets, including BusinessTech, Daily Maverick, IOL, and MyBroadband, were knocked offline. Newcomer Newsday itself was not spared.

Streaming and radio services such as SABC+, Jacaranda FM, East Coast Radio, and Primedia+ (702 and CapeTalk) were also hit.

Cloudflare issued several updates as it worked to restore service. MyBroadband reported how at 15:09 SAST, it announced that engineers had identified the source of the problem and were rolling out a fix.

Just before 17:00, it reported progress restoring application services, and by 16:42 UTC it declared the issue resolved.

“A fix has been implemented and we believe the incident is now resolved. We are continuing to monitor for errors to ensure all services are back to normal,” the company said.

Cloudflare confirmed that the outages were triggered by a sudden and unexplained spike in traffic, though the underlying cause remains under investigation.

The company said it would continue posting updates on cloudflarestatus.com and provide a detailed post-incident analysis on its engineering blog once its review is complete.

The incident marks the latest in a string of high-impact internet disruptions. Just last month, an Amazon Web Services outage caused global turmoil, taking down thousands of popular websites and apps for hours.

Cloudflare’s shares fell about 5% in premarket trading following news of Tuesday’s outage.