Michael JohnsonThe Microsoft 365 outage, which began on November 12, affected access to Outlook and several other services, as well as Microsoft Graph calls involving email, calendars and contacts. According to Microsoft status updates, some servers unexpectedly became unavailable, shifting the load to the remaining infrastructure. As of November 13, Exchange Online and Microsoft 365 incidents were still listed as ongoing in a public mirror of Microsoft administrative notices.
The practical takeaway for organizations is to check more than just the email client. If an application or automation accesses Exchange data through Microsoft Graph, the outage may show up as a timeout or failure in a related operation—even if the application's own interface continues to open.
What Microsoft said about the cause
In its Exchange Online log, Microsoft said the increased load was not caused by an overall rise in traffic: many servers unexpectedly became unavailable, shifting the load to the remaining nodes. In a subsequent update, the company said a process on affected servers responsible for establishing secure connections was becoming overloaded, and the servers were ceasing to process requests.
Microsoft described several recovery measures: redirecting some traffic, restarting servers, limiting request types in certain locations, and checking the previous software configuration on some nodes. The notice also said deployments were paused and configuration changes made while Microsoft investigated a possible connection to recent changes. This describes the investigation's progress; the cited status update did not publish a definitive root-cause analysis.
Why an email outage affected other products
According to Microsoft's Exchange Online outage notices, affected ways of accessing mailboxes may have included Outlook on the web and desktop, Exchange ActiveSync, and REST connections. Email delivery could also be delayed. Microsoft separately warned that Graph requests for Exchange data—such as email, calendars and contacts—could time out.
As a result, features that rely on email or calendar data may have been affected even in products with a separate interface. Microsoft listed possible disruptions to Teams calendar scenarios, some SharePoint Online features, Copilot and administration tools. The company noted, however, that core Teams functions—calls, meetings, group chats and channel messages—generally remained available, while files stored directly in SharePoint remained accessible.
The scale of the disruption was also reflected in an Associated Press report on November 12, which described problems with Outlook, Teams and other Microsoft 365 services. AP reported, citing Downdetector, that the number of user reports about Outlook problems exceeded 100,000. That figure represents reports on an outage-tracking service, not the number of confirmed affected accounts.
What administrators and developers should check
If an application depends on Microsoft Graph, identify exactly which calls read email, calendars or contacts, and check how the system handles timeouts and temporary errors. For tasks that can be safely retried, limited retries with delays can help; operations with side effects need safeguards against repeat execution so that retrying a request does not create duplicates.
Administrators should check the status of specific services and incident IDs in their Microsoft 365 tenant's Service Health dashboard: access to the dashboard and affected features may vary between organizations. The status log indicates that the disruption did not affect all functions of related products equally. When investigating employee reports, it is useful to ask separately whether email is not arriving, mail will not open, or a specific operation is failing—for example, creating a meeting invitation or downloading a file with a protection label.
In the latest updates cited, dated November 13, Microsoft was continuing its recovery work and assessing infrastructure changes. For users, this meant that availability of one part of Microsoft 365 did not guarantee that features relying on Exchange Online were working. For teams supporting integrations, the key is to check dependencies on email data and handle temporary Graph errors correctly, rather than treating the outage as only an Outlook interface problem.
