Microsoft 365 Outage Lasted Over 21 Hours, Hit Police, Hospitals, Airlines

A Microsoft 365 outage knocked out email, Teams chats and meetings for millions of users across four continents for more than 21 hours, with meaningful recovery only beginning at 11:20 GMT on Friday, according to data shared by the research firm Infoblox. Microsoft has ruled out a cyber-attack. It blames a configuration update that overloaded servers and took them offline.

The first surge of user reports appeared at about 6pm GMT on Thursday on the outage website Down Detector. By Friday morning Microsoft had traced most of the problems to Microsoft Exchange, the email service behind Outlook. Teams meetings and chats had already returned. By midday UK time, Microsoft reported "encouraging recovery progress in Europe", improvements in North America and Asia Pacific, and said Japan was "stable".

What caused the outage?

A configuration update caused some servers to become overloaded and go offline, Microsoft indicated. When engineers applied a fix, servers that came back online became unresponsive again as traffic returned. As some servers failed, traffic funneled onto the remaining machines, overloading those too. Microsoft responded by limiting traffic to affected servers in specific locations and rolling some servers back to an earlier software configuration.

The full picture will take time. "We don't know what actually went wrong, and that will take weeks to discover," said Steven Murdoch, a professor of security engineering at University College London. "The people who are working on it are too busy fixing to understand or report on the cause." Murdoch expects a proper postmortem on an outage of this size, but not for a few weeks.

He added that outages this large are usually "the result of some surprising interaction between systems. Each of the systems is normally quite well understood but interactions between them are complex and frequently the cause of large outages like this." In his words: "It's very hard to build a complex system without there being some risk of some failure."

Who was affected?

The impact was global and patchy — some users kept partial access, and recovery speeds varied by region. Affected organisations included:

• UK: police services, the Bank of England, ambulances, the NHS and schools. The NHS, which holds a five-year deal in England to use Microsoft 365, said a range of services had been disrupted.
• Australia: councils, state and federal government agencies, courts and schools, plus a spy agency.
• Europe: government agencies in the Netherlands, Belgium, Germany and France; one French municipality told residents to phone instead.
• Japan: government services and private companies. Shipping company Sagawa Express could not accept online reservations for some airport and railway station services; pharmaceutical company Taisho handled orders by phone; some employees switched back to faxes.
• India and Taiwan: airlines IndiGo and Akasa Air lost email support; Taiwanese security firms, life insurers and investment trust companies reported disrupted fund management services.

Even the US House of Representatives and Microsoft's own employees reported issues. The UK government said it was "working closely" with Microsoft and had "robust contingency plans in place to minimise disruption".

What does the outage say about cloud dependence?

UK politicians and experts said the incident exposed the world's dependence on cloud services run by three US giants — Amazon, Google and Microsoft. That unease has grown with each high-profile outage, and the Trump administration's hostility towards Europe has pushed some governments to consider alternatives to US software.

Mark Gregory, an associate professor at Australia's RMIT University, said organisations must thoroughly test updates before rolling them out to live networks, on top of extra security testing against threat actors whose AI tools can find vulnerabilities faster than ever. He also urged backup options such as Google's Gmail; others have suggested the free, open-source LibreOffice. In practice, many companies and government services clearly had no good backups, resorting to faxes and phones.

The Forrester analyst Dario Maisto said the outage "represents the perfect quasi-natural experiment for stress testing an organisation's – and even an entire continent's – digital sovereignty posture", forcing governments to consider what happens if mail services and key tools such as Copilot go dark. Microsoft says email queues are "significantly reduced and approaching normal levels", but a full accounting of the failure — and its political fallout in Europe — will take weeks.