Skip to main content

Morning Overview

A worldwide Microsoft outage disrupted Outlook, Teams and Copilot

Email stopped arriving for Outlook users around the world on Thursday, Nov. 12, and by the afternoon Teams calendars, SharePoint sharing and Microsoft 365 Copilot were struggling too. Microsoft opened incident MO1550821 with a start time of 1:26 p.m. ET and warned that the connectivity problem could reach users in any region. By Friday, Nov. 13, the company had apologized while the backlog of delayed mail was still draining.

Read next on Morning Overview: 11 engines built to run well past 200,000 miles

Administrators spent much of two days refreshing the same service health page. The incident list grew from the five services named at the start to a long tail of dependent products, and Microsoft revised its own explanation of the cause more than once before it settled on one.

Five services named at the start, a longer list by Friday

BleepingComputer reported that the incident, which began Thursday at 1:26 p.m. ET, listed Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender XDR as affected. Exchange Online carried the worst of it: mailbox access failed in Outlook on the web, desktop and mobile, and over the REST API, and outbound and inbound mail was delayed.

The damage to the other services was narrower than the list suggests. In Teams, calendar scheduling, meeting invitations and edits, and channel creation or deletion failed, while core chat and meetings largely kept working, according to a timeline compiled by On-Site Technology. SharePoint trouble was limited to features that depend on Exchange, so files stored directly in the service still opened.

Defender XDR mattered for a different reason. A security analyst reading Threataft’s analysis would have seen alert visibility drop and incident queues stop updating, which is a bad moment to lose the console that watches for attacks. Microsoft later added Purview, Intune, Planner, To Do, Power BI, Fabric, Windows 365 and Universal Print to the affected set.

How Microsoft’s explanation changed

The first public account blamed “elevated load.” Microsoft then corrected itself, saying the strain was not caused by more traffic and that a number of servers had become unexpectedly unavailable, leaving the remaining infrastructure to carry too much. A community megathread on Windows Forum collected those updates, including the company’s statement that it had found no recent change correlated with the start of the outage.

By Thursday evening the working theory was that a process on each server, the one that sets up secure connections, had become overwhelmed and stopped handling requests, which meant the health checks removing those servers were right to fail them. Servers Microsoft returned to service around 6:06 p.m. ET became unresponsive again when traffic came back at 7:08 p.m., a setback that pushed recovery into Friday.

On Friday, Microsoft gave its clearest account. It said the incident started with an issue during an earlier network capacity upgrade that surfaced after a recent network change, and it said the outage was a technical issue unrelated to a cyberattack or to AI agents, as relayed by the Azure Blog UK write-up. A full post-incident report is still pending.

Microsoft’s wording on scope shifted as the hours passed. At 1:07 p.m. ET it said the issue may affect Microsoft 365 service connectivity for users in any region, which is the basis for calling the disruption worldwide, and recovery later spread unevenly, reaching Asia Pacific, Europe and North America from about 6:01 a.m. ET on Friday.

Recovery overnight and the 48-hour mail window

Recovery came in stages. A patch to an upstream authentication and routing component improved stability at about 2:24 a.m. ET on Friday, and it rolled out region by region from roughly 3:09 a.m. At 4:03 a.m., Microsoft extended the email timeout window from 24 hours to 48 hours so delayed messages would not be discarded, though the company did not guarantee delivery of every message.

Downdetector gave a rough sense of public pain. Outlook complaints peaked near 130,000, with more than 38,000 for Microsoft 365, figures that count user reports rather than affected accounts. The same user-report data mentioned OneDrive and Azure, but Microsoft did not list Azure in the incident.

As of 3:02 p.m. ET on Friday, the On-Site Technology timeline described the outage as mostly recovered, with MO1550821 and the companion Exchange incident EX1550543 still open and a next update due by 7:00 p.m. ET. An earlier, separate Teams incident, TM1550488, had run from 9:45 a.m. to 10:53 a.m. ET on Thursday, and Microsoft did not say whether the two were connected.

Microsoft also addressed the question many customers asked first. A company executive said there was no indication of malicious activity, and the Friday explanation described a technical fault in Microsoft’s own network capacity work, not an attack and not a failure tied to AI agents.

What administrators were told to do

When the admin center itself is unreachable, Microsoft’s public status page at status.cloud.microsoft is the fallback for following incident IDs. Guidance gathered in the Azure Blog UK piece told administrators to reach staff through a channel outside Microsoft 365, to avoid reinstalling Office or rebuilding Outlook profiles for a fault on the service side, and to hold their own changes until recovery was confirmed.

The open question is the root cause. Microsoft has promised a post-incident report covering the cause and repair measures, and until it appears the explanation rests on the company’s rolling status posts, which shifted more than once in two days.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.


More from Morning Overview

Morning Overview is reader-supported. Some links in our articles are affiliate links, and we may earn a commission at no extra cost to you. As an Amazon Associate I earn from qualifying purchases. Full disclosure.