OVER 132,000 OUTAGE REPORTS AS META GOES DOWN

Friday morning's Meta outage was brief but massive, hitting the company's three biggest consumer services at once and generating more than 132,000 user complaints on Facebook alone before the numbers began to fall. The outage started just before 7 a.m. Pacific Time and resolved broadly within about an hour, though pieces of the apps and sites stayed broken for longer.

The timing was unlucky for the West Coast (early risers) and brutal for much of the rest of the world, where people already well into their Friday were suddenly locked out of their main social feeds.

THE SCALE

Downdetector recorded the peak for Facebook at around 6:50 a.m. PT: 132,000 reports from users who could not log in, could not load the site, or saw errors. That number dropped to about 80,000 by 7:15 a.m. and then more quickly to 10,000 by 7:35 a.m., meaning the heaviest concentration of failures lasted less than an hour. Instagram's peak came four minutes later at 6:54 a.m., with about 9,500 reports. Messenger peaked above 16,000. WhatsApp saw a spike but it was tiny by comparison, fewer than 200 reports at its worst.

The numbers show that Facebook was the hardest hit by a wide margin, but Instagram and Messenger were affected at the same time, pointing to a backend issue common across Meta's infrastructure rather than a single-product glitch.

WHAT BROKE

The most common symptom was inability to log in at all. Users reported being kicked out of sessions on both web and mobile and then failing to reauthenticate. For those who did get in, features were missing or broken. The left sidebar on the desktop version of Facebook did not load for some users. The Stories tray on the mobile app displayed an error message instead of friend updates. One user opened the app to find a blank space where stories usually appear, with a system message in its place. Another reported that the desktop site worked fine but the mobile app was lagging.

These are the kinds of failures that suggest a partial service restoration: the authentication layer came back, but some downstream services (like the feed of story content or the sidebar component) were still timing out or returning errors.

THE RECOVERY

By 8 a.m. PT, most users could log in again. Downdetector's Facebook status page was back up and showing a steep downward curve. NordVPN, another outage tracking service, updated its page to reflect a partial outage rather than a full one, and at 8 a.m. reported only 702 users flagging problems.

Yet the recovery was not universal. Some users still reported being unable to access Facebook at all through the morning. Others got in but with limited access. The Instagram report count, after dropping from its peak, crept back up to 8,170 by 7:39 a.m., suggesting that while the main incident was over, secondary issues were still flaring for a subset of users. Facebook Messenger was described as working again, as was the Stories feature for those who could see it.

This is a common pattern for an outage that involves pushing a fix that covers the majority but leaves some edge cases unpatched.

WHAT META SAID (AND DIDN'T)

Meta did not issue a public statement about the outage during or immediately after it, and it has not explained what caused the failure. CNET has reached out to Meta for details.

WHAT THIS MEANS FOR USERS

For the vast majority of users, the outage was a short inconvenience. The platforms were effectively dark for less than an hour, and recovery was fast. But the incident is a reminder of how concentrated Meta's infrastructure is. A single failure can take down Facebook, Instagram, and Messenger at once, and any of those services going dark means a significant portion of global internet traffic goes silent for a while.

For now, the numbers tell the story. More than 132,000 users filed reports, and then they did not. That is the shape of a contained incident.