130,000 reports, no cyber-attack: Microsoft's mailboxes go dark, and the case for a second channel

On Thursday evening, European time, the dull machinery of office life stopped for a lot of people at once. Outage tracker Down Detector showed a spike in Microsoft 365 problems around 1pm New York time (19:00 CET), with nearly 130,000 reports covering Word, Excel, PowerPoint, Outlook and Teams, as well as SharePoint and OneDrive; the Copilot assistant was hit too¹. Disruption lasted more than five hours, and reports had fallen to about 13,000 five hours after the peak¹.

What Microsoft says happened. A portion of its infrastructure was “experiencing elevated load”, and a number of servers were “unexpectedly unavailable” after a problem caused them to fail automated checks^1,2. A spokesperson said the company “had no indication of a cyber-attack”¹. Late on Thursday Microsoft said restored servers had remained stable, that it was shifting traffic to alternate routes, and that users “may experience a gradual recovery” as their clients reconnect¹. It was also assessing whether reverting “the small number of recent configuration updates” it had identified could help¹. TechCrunch notes that Microsoft’s previous major incident, in September, was blamed on a “core authentication configuration” issue².

Why one fault spreads. Mark Gregory of RMIT University described the mechanism: if part of a pool of servers fails, every user crowds onto the rest, “and that overloads them. And then they go down”¹. The outage also broke connectors that OpenAI’s and Anthropic’s AI tools use to reach Microsoft 365¹. The more services lean on one another, the further a single misstep travels.

For you. Nothing here was a hack, and nobody needs to panic. But the lesson is the same one every payroll office learnt long ago: keep a second channel. A personal email address your colleagues know, a phone number for the people you must reach, and important documents saved locally rather than only in the cloud. Gregory put it simply: “it then becomes important to have an alternative”¹. And if a message arrives during an outage asking you to “re-verify” your account, treat it with extra suspicion; Microsoft’s own guidance was that recovery would come gradually as clients reconnect¹.