Microsoft said early on Saturday, November 14, that it had restored service to the remaining customers and regions affected by a global Microsoft 365 disruption. The outage began with Exchange Online on Thursday, November 12, and continued to affect Outlook, Teams calendar functions and other services into Friday. The company had not yet published the full post-incident report it promised.
A restoration notice closed a disruption that crossed two working days
Microsoft 365 Status said early on Saturday that it had completed its work on the underlying incident and restored service to the remaining impacted customers and regions. The notice, posted on X, ended the company's public description of an active global recovery effort; it did not include the full technical review that Microsoft said would follow.
Reuters reported that the outage had disrupted business and government communications for more than 24 hours. Downdetector said Outlook reports began climbing at about 17:25 UTC on Thursday, while its own reports reached 74 countries and territories in roughly an hour. Those are reports submitted to an outage tracker, not a count of people affected.
The disruption reached beyond office inboxes. The Associated Press reported effects on government agencies, courts, healthcare services, businesses, airlines and banks. Reuters described delays to police online reporting, health-service workarounds, airline customer communications and financial firms' operations. The consequences varied by organization and by the Microsoft service on which a task depended.
Sources: Service restored to the remaining impacted customers and regions · Microsoft says services fully restored after hourslong global outage · Microsoft restores service to Outlook and Teams after global outage · Microsoft outage reports reached 74 countries and territories in about an hour
Exchange Online was the common dependency beneath several products
Microsoft tracked the mailbox problem as Exchange Online incident EX1550543. A wider Microsoft 365 advisory, MO1550821, covered related service functions that depended on Exchange Online. The sequence matters: this was not evidence that every Microsoft application or every feature in each application stopped working at the same time.
Outlook users could have trouble reaching mailboxes through Outlook on the web, desktop connections, Exchange ActiveSync or REST-based connections, while email delivery could be delayed. In Teams, Microsoft identified problems with calendar-dependent tasks such as scheduling, invitations and editing meetings. It said most core functions, including joining calls and using group chats and channel messages, were not affected.
Other services experienced narrower knock-on effects. Microsoft listed Copilot Chat and access to work data, some Defender XDR portal pages, intermittent Intune enrollment and device-check-in scenarios, and selected SharePoint sharing or Exchange-dependent features. Direct access to files stored in SharePoint remained functional, according to the service-health description. The impact therefore followed service dependencies rather than a single uniform shutdown.
Sources: Microsoft 365 service health status: incidents MO1550821 and EX1550543 · Microsoft 365 incident updates for EX1550543 and MO1550821
Microsoft's explanation is still an interim account
Microsoft said the incident was introduced by a problem during an earlier network-capacity upgrade that surfaced after a more recent network change. In earlier updates, it said a number of servers had become unexpectedly unavailable, shifting load onto the infrastructure that remained in service. The company described this as its account of the incident, not as a completed public root-cause analysis.
During the response, Microsoft said affected servers were failing automated health checks and becoming unavailable, while the remaining infrastructure struggled with the redistributed traffic. Engineers tested configuration changes, restarted servers and moved traffic across its wider network. These operational updates explain why recovery required repeated validation, but they do not establish every technical step behind the original failure.
Microsoft said it had found no evidence of malicious activity and described the event as a technical issue not related to cyber activity or AI agents. That distinction matters: an availability failure can have serious consequences without being a cyberattack. Until the promised review is published, the capacity-upgrade explanation should remain attributed to Microsoft and treated as preliminary.
Sources: Service restored to the remaining impacted customers and regions · Microsoft says services fully restored after hourslong global outage · Microsoft 365 incident updates for EX1550543 and MO1550821
Recovery was staged, and restored access was not instant normality
Microsoft said it restored services gradually, checking stability at each stage because of the scale of demand. Its status updates described traffic being redistributed, unavailable servers being returned to service and recovery being assessed across different connection methods. A gradual restoration can take longer than switching a single component back on, particularly when the change itself must be shown to remain stable under load.
Email access and email delivery were not necessarily recovering at the same speed. During the incident, Microsoft said users might regain sign-in or mailbox access while queued messages were still being processed. It continued to monitor the accumulated email backlog as part of its recovery work. That distinction is important for organizations assessing whether messages were delayed, rejected or simply not yet delivered.
The Saturday notice said service had been restored to the remaining affected customers and regions. It should be read as a service-restoration statement, not as proof that the separate post-incident review had been completed or published. The status record and the promised report answer different questions: one concerns whether services are back; the other should explain why they failed and what will change.
Sources: Service restored to the remaining impacted customers and regions · Microsoft 365 incident updates for EX1550543 and MO1550821
Outage reports show the reach of the disruption, not its user count
Downdetector said reports about five Microsoft services peaked at 19,650 per minute on Thursday, compared with fewer than 15 per minute on an average Thursday. Reuters later reported more than 7.4 million Downdetector reports worldwide. Those figures measure incoming reports across services and time; they must not be converted into a number of unique affected users or customer accounts.
Microsoft says more than 430 million people use Microsoft 365 apps. That figure gives a sense of the platform's reach, but it is not a measure of how many people experienced this outage. The company has not provided a verified global count of affected users in the sources reviewed for this article.
Reuters called the disruption one of the longest in Microsoft's history and its second service disruption of the year. That is the news agency's characterization, not a ranking published by Microsoft. The more defensible point is that a failure centred on one cloud email and scheduling service can affect a wide set of products when they depend on the same underlying service.
Sources: Microsoft says services fully restored after hourslong global outage · Microsoft restores service to Outlook and Teams after global outage · Microsoft outage reports reached 74 countries and territories in about an hour · Advancing Microsoft 365: New capabilities and pricing update · Microsoft 365 incident updates for EX1550543 and MO1550821
The promised report is the next important test
As of November 14, Microsoft had marked the affected services as restored but had not published the full post-incident report. Its status update said the report would follow once the company's review was complete. The latest service-health material available at the article's information cut-off continued to describe the cause as an earlier network-capacity-upgrade issue that surfaced after a later network change.
The report will matter because the public explanation so far identifies a broad sequence but leaves technical questions unanswered: which component failed, why the earlier upgrade created a latent problem, how the later change exposed it, and which safeguards will prevent a recurrence. Those answers should come from Microsoft's review rather than from speculation about architecture that the company has not disclosed.
For organizations, the outage is a reminder to test continuity plans around shared dependencies, not just individual applications. A backup communication channel is useful only if staff can reach it when email, calendars or identity-linked services are impaired. The service has returned; whether the systems around it can keep essential work moving during the next disruption is a separate question.
Sources: Service restored to the remaining impacted customers and regions · Microsoft 365 service health status: incidents MO1550821 and EX1550543 · Microsoft 365 incident updates for EX1550543 and MO1550821