LIVE: Following AWS’s Lead, Cloudflare Also Attempts “Turning the Internet Off and On Again”

Cloudflare announced a “global service disruption” Tuesday after a sudden spike in unusual traffic sent parts of the internet into a synchronized meltdown. Within minutes, websites, apps, and entire engineering teams collectively flatlined.

The most relatable moment of the outage came early:
An engineer opened Claude to ask what was happening: error.
Went to Twitter to complain: error.

They stared at the screen for a full ten seconds, closed the laptop, and went to bed.

Coworkers later described him as “the only one who made the right choice.”

Major platforms like ChatGPT and X sputtered, internal dashboards froze, and incident channels filled instantly with identical messages:
“Is this us?”
“No way this is us.”
“…wait is this us?”

One security analyst summed up the mood perfectly:

“Well, I guess we don’t have to patch anything today. Nothing works anyway.”

Some employees, unable to load anything online, reportedly opened local files, something historians believe hasn’t happened since the late 2000s.

Cloudflare says services are “recovering,” which every engineer instinctively translates to:

“It’s still on fire, but we’ve stopped adding new fires.”

Update 15:12 UTC — error reports have dropped significantly, though a few stragglers still linger. Engineers are cautiously reconnecting to the internet.

This is an ongoing story. Updates will resume once the internet remembers how to internet.