TECHNOLOGY

Microsoft 365 outage disrupts Outlook and Teams worldwide

Some regions briefly regained access as Microsoft worked to bring failed servers back into service.

A global Microsoft 365 outage continued into Friday on the U.S. East Coast, leaving some users unable to open Outlook mailboxes and disrupting features in Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR. Microsoft reported temporary improvement in some regions, but both its Exchange Online and broader Microsoft 365 incidents remained open at midnight, with no confirmed root cause or estimated time for full restoration. Microsoft incident records

Microsoft logged the Exchange Online incident at 11:17 a.m. EST Thursday and opened a broader Microsoft 365 incident at 1:26 p.m. Around 1 p.m., Downdetector recorded more than 130,000 U.S. reports of Outlook problems, according to Reuters. By 9:30 p.m., the figure was still above 25,000. Those figures reflect reports submitted to the tracker. The total number of people affected remains unknown. Microsoft incident records · Reuters

What stopped working

Exchange Online users could have trouble opening mailboxes through Outlook on the web, the desktop app or mobile clients. Email delivery could also be delayed. In Teams, Microsoft flagged problems with calendar access, invitations and meeting edits, while most core calls and chats remained available. SharePoint users could still open files stored directly in the service, though features that rely on Exchange Online could fail. Microsoft incident records

Microsoft also warned that Copilot Chat could fail to load or respond, that some Defender XDR portal pages could be unavailable, and that Microsoft 365 Education features such as Assignments and Reading Progress could be affected. The disruption reached the U.S. House of Representatives and Japanese government IT systems. Electric truck maker Workhorse Group postponed its quarterly results, citing the outage. Microsoft incident records · Associated Press · Reuters

Why recovery kept slipping

Microsoft traced the initial strain to a loss of server capacity. A large number of servers failed automated health checks and dropped out of service, leaving the remaining machines to handle more requests. Engineers later identified a process that establishes secure connections and could become overwhelmed, causing affected servers to stop responding. Microsoft reviewed recent configuration changes as it investigated what triggered the failures. Microsoft incident records · PCMag reporting via AOL

Engineers restarted servers, tested a change to bypass a component involved in the health checks, and moved traffic away from affected infrastructure. Some servers appeared healthy after a restart but stopped responding when traffic returned. Microsoft tested bringing an isolated group of servers back online and feeding requests to them gradually. It also restricted certain types of traffic in some locations. Microsoft incident records

Near midnight, Microsoft said it was adding virtual-machine capacity after seeing an increase in requests between its own routing services and backend servers during recovery. Some machines remained stuck in a state that prevented a network change from taking effect. Earlier, Microsoft had attributed the initial overload to the loss of available servers and said it had found no indication of malicious activity. Microsoft incident records · The Guardian

Effects beyond Microsoft 365

The outage also reached tools that rely on Microsoft 365 data. Reuters reported that OpenAI investigated trouble with its Microsoft connectors and that Anthropic identified an upstream issue affecting its Microsoft 365 connector. The November incident followed a separate Azure outage in October that also disrupted Microsoft productivity services. Reuters

Where to follow recovery

Microsoft 365 administrators can follow EX1550543 for Exchange Online and MO1550821 for the wider disruption in the Service Health dashboard. At midnight EST, Microsoft had scheduled its next broader update for 1 a.m. EST and its next Exchange update for 1:30 a.m. EST. Readers can follow both incidents through Microsoft’s public status channels if the admin center is unavailable. Microsoft incident records · Microsoft service status · Microsoft 365 Status on X