Facebook, Instagram, and Threads are back from a major outage: Here's what happened

If you were getting errors when logging into Meta's apps, it wasn't just you.
Written by Artie Beaty, Contributing Writer
Screenshot by Alyson Windsor/ ZDNET

Were you logged out of your Facebook, Messenger, Threads, or Instagram account on Tuesday morning? You're not alone.

At around 10 a.m., reports began surfacing of users getting completely logged out of their accounts for Meta's services. It quickly became apparent that this was a widespread issue, with seemingly nobody able to use Meta's services.

The real-time outage monitoring website DownDetector showed nearly 400,000 people reporting Facebook issues by 10:45 a.m. By noon, all services had returned. 

Also: Meta is deleting Oculus accounts and everything in them: How to save yours

During the outage, attempting to log in led to an error message stating "An Unknown Error Has Occurred," as did trying to reset your password. Even two-factor authentication was broken, with the code being sent, but a prompt to enter it never appearing.

A post from Meta communications spokesperson Andy Stone blamed "a technical issue" and added that the problem "caused people to have difficulty accessing some of our services." "We resolved the issue as quickly as possible for everyone who was impacted," he wrote, "and we apologize for any inconvenience."

Meta's status website, which tracks the condition of the company's services, showed "major disruptions" to tools like Ads Manager, Facebook & Instagram Shops, and Meta Business Suite, WhatsApp Business API, and Messenger API, but surprisingly indicated "no known issues" for Messenger, despite massive reports to the contrary. By noon, all services showed "recovering from an outage." 

Since all down platforms use Meta to log in, early speculation is that the cause could be tied to Meta identity and access management. No actual network failures or massive DDoS attacks have been reported (yet). A similar situation happened in 2021, caused by a faulty configuration change. 

We'll update with more information as it becomes available. 

Editorial standards