A Microsoft 365 outage that began on Thursday afternoon has knocked out email, meetings and cloud services for millions of people around the world, in one of the biggest failures of Microsoft’s workplace software in years.
Key Points
- Microsoft 365 outage — Outlook, Teams, OneDrive, SharePoint, Copilot and other services began failing from about 1:26 p.m. ET on Thursday, 12 November.
- Downdetector logged nearly 130,000 Outlook problem reports in the United States, and more than 43,000 across 72 countries by Friday morning.
- Microsoft says a large number of servers became unexpectedly unavailable, overloading the rest of its infrastructure. It says the failure is not a cyber attack.
- By Friday morning Microsoft reported early signs of stabilisation, but said capacity had not yet returned to normal.
The problems started late on Thursday afternoon in the United States and spread to every region. People found Outlook would not load, emails would not arrive, Teams could not book meetings, and files in OneDrive and SharePoint were unreachable. Microsoft’s own status page was slow or unreachable at times, so the company posted updates on X through its @MSFT365Status account, and tracked the incident in its admin centre as MO1550821, with a separate incident, EX1550543, for Exchange Online.
What went wrong
Microsoft’s explanation is that the elevated load was not driven by a surge in traffic. Instead, a number of its servers became unexpectedly unavailable, and the remaining infrastructure buckled under the extra work. The company said it had paused deployments, freed up memory, rerouted traffic across its infrastructure and begun rebooting servers. By 6:57 p.m. ET on Thursday it said it was seeing initial signs of recovery after bringing some affected servers back online, and on Friday morning it said traffic was being carefully reintroduced.
The company has been clear on one point: it says it is not aware of any malicious activity, and the outage is not a cyber attack. No account has been hacked and no data has been leaked because of the failure, according to Microsoft’s updates.
Who was affected
The outage reached far beyond office workers. In Britain, GP surgeries, pharmacies, schools, police forces, Parliament and thousands of small businesses reported problems, and personal Outlook.com and OneDrive accounts were hit as well as business ones. Some services kept working: Teams calls, meetings, chats and channel messages were largely unaffected, and SharePoint and OneDrive files remained available for many users, even while email was down.
Microsoft has extended the email timeout window to 48 hours, so messages sent during the outage should still arrive rather than bouncing back. It is the second big Microsoft 365 outage this year, after one in January that lasted nearly 10 hours in North America. Businesses that pay for Microsoft 365 may be able to claim service credits, but they will need to ask Microsoft, quoting the incident number, rather than waiting for a refund to arrive on its own.
For background on the AI tools caught up in the outage, see our report on OpenAI’s revenue hitting $10 billion in a quarter, and on Anthropic launching its faster, cheaper Claude models. More in our Technology section.
Sources: Dijitul, “Microsoft 365 Outage: What To Do Now”; Windows Latest, “Microsoft 365 outage: you may get money off your bill if you use it for work”.