A widespread Microsoft 365 outage that began around 1 p.m. ET on Thursday, November 12, disrupted Outlook and other services for users in dozens of countries and was still not fully resolved early Friday. Microsoft said its recovery work was not going as expected. Downdetector logged roughly 130,000 Outlook problem reports at the peak, with one tally putting the high at 138,058. It showed users affected in 74 countries.
Microsoft first reported trouble reaching Exchange Online mailboxes in the Americas, then said users in any region could be affected. People could not open inboxes or send and receive mail through web, desktop and mobile Outlook. Teams calendars and meeting scheduling, Copilot, SharePoint Online and Defender were also hit, while core Teams messaging largely kept working. OpenAI and Anthropic both reported problems with their Microsoft 365 connectors.
Microsoft says a number of servers became unexpectedly unavailable, leaving the rest under heavy strain. The cause was a loss of capacity, not a surge in traffic. It traced the failures to a core process that sets up secure connections and became overwhelmed, and to automated health checks that kept pulling servers out of service. An alternative health check brought a subset back. Microsoft was also reported to be reversing a recent network configuration change that coincided with the start, to see whether it contributed. Executive Vice-President Charles Lamanna said the outage was not a cyberattack. No root cause has been confirmed.
Recovery has been uneven. Microsoft paused deployments, moved traffic off the affected infrastructure, restarted servers and imposed temporary traffic limits in some locations. Around 6 p.m. ET Thursday it said restored servers were stable, then warned they could become unresponsive again once traffic returned. Downdetector reports fell to about 20,000 by 11 p.m. ET. Microsoft’s latest update said it was testing a configuration change on a subset of servers before a wider rollout. The Exchange Online incident (EX1550543) and the wider Microsoft 365 incident (MO1550821) remained open, with no estimated time for full recovery.
The effects reached beyond offices. New York City Emergency Management warned residents of delays reaching city agencies by email, and Service NSW in Australia urged customers to delay visits to its centres. Electric truck maker Workhorse Group moved its third-quarter results from November 12 to November 16, citing the outage.
The episode shows how many services sit on one email layer, from Copilot to third-party AI connectors. Administrators can follow both incidents in the Microsoft 365 admin center, or at status.cloud.microsoft if the portal is unavailable, and should keep an out-of-band channel ready for staff and customers.