Facebook and Instagram experienced severe issues around the world on Tuesday, with the services refusing login attempts and feeds stalling.
The outages were first reported at about 3.30pm GMT, and began to clear at about 5pm.
Unusually, the issues coincided with login problems on Google’s platform, suggesting a common cause between the outages at two tech conglomerates that largely control their own infrastructure.
Meta’s business status page reported a number of disruptions, including “major disruptions” for the group’s admin centre, as well as for Facebook login, the service that allows users to log in to third-party services using their Facebook details. That, in turn, has led to some reporting outages at a variety of other sites.
At 4pm GMT, Meta updated the entire status page to report “unknown” status for all services other than Messenger API for Instagram. Some Meta services, such as WhatsApp and the Facebook ads transparency page, were apparently still working. At 4.15pm, the Meta status page stopped working.
In his first post on X in a week, the Meta spokesperson Andy Stone said: “We’re aware people are having trouble accessing our services. We are working on this now.”
Google’s ads status page reported a disruption to the company’s Ad Manager beginning at about 3.30pm GMT. The company said it was investigating reports of other issues. But the company’s hiccups were more contained than those of Meta, with most consumer-facing services, including search and YouTube, still working as usual. Problems with the company’s login service, however, affected some business customers, including the Guardian.
A systemic internet issue is likely to be the root cause, with sporadic issues also being reported by users of sites including X and Microsoft’s Teams.
The outages are unlikely to be as severe as Facebook’s 2021 outage, when a configuration error in a little-known protocol called BGP led to the company accidentally deleting its own address from the systems that allow servers to talk to each other on the internet.
after newsletter promotion
Although that error was spotted immediately, it took several hours for the fix to be enacted and take effect – in part because the company’s engineers could no longer gain remote access to their own servers to fix the problem, nor could they use their corporate passes to get through the electronic locks to gain physical access.
Google has been contacted for comment.