A major Microsoft 365 outage disrupted Outlook, Teams and other cloud services on Thursday, November 12, 2026, leaving businesses and government agencies around the world struggling to access email and workplace tools. The disruption lasted for hours as Microsoft attempted to restore servers that had unexpectedly become unavailable.
The outage began around midday in the eastern United States and quickly generated thousands of complaints. At its peak, the outage-monitoring website Downdetector recorded more than 130,000 reports involving Outlook in the United States, making it one of the most widely reported Microsoft service disruptions of the year.
Microsoft confirmed that the problems primarily involved Exchange Online, the cloud-based email service supporting Outlook and other Microsoft 365 applications. The company also identified connectivity problems affecting several additional products, including Teams, SharePoint and Copilot.
Although some services began recovering during the evening, Microsoft continued investigating the underlying technical problems. Its initial attempts to restore affected infrastructure did not provide the expected improvement.
Outlook and Microsoft 365 Users Report Widespread Problems
The outage prevented many Outlook users from opening mailboxes, sending messages or receiving email. Some customers encountered connection errors when accessing Outlook through web browsers, desktop applications and mobile devices.
Exchange Online serves millions of business users who depend on Microsoft's cloud infrastructure for workplace communications. When the service becomes unavailable, organizations can lose access to important messages, calendars and automated systems connected to their email accounts.
Microsoft Teams also experienced disruptions, although the effects varied depending on which features customers were using. Services connected to Exchange Online, including some calendar and meeting functions, were particularly vulnerable.
Other affected Microsoft products included SharePoint Online, Microsoft Defender, Planner, Purview and Copilot. Not every feature of these services failed, but users encountered problems when applications depended on the affected infrastructure.
The widespread nature of the outage became apparent as complaints increased across North America and other regions. Reports continued into the evening, with businesses and public agencies struggling to maintain normal operations.
Microsoft Identifies Problems With Unavailable Servers
Microsoft initially suspected that unusually heavy infrastructure loads were contributing to connection failures across multiple regions. Engineers began investigating whether a sudden increase in requests or changes in network traffic had overwhelmed the affected systems.
Further investigation revealed that the additional load was not primarily caused by a general increase in customer traffic. Instead, a large number of servers had unexpectedly become unavailable.
That reduced the capacity of the remaining infrastructure, forcing functioning servers to handle more requests. The resulting pressure contributed to additional connection failures and made recovery more difficult.
Microsoft also discovered that affected servers were failing automated health checks. These checks normally help cloud platforms identify malfunctioning equipment and remove unhealthy servers from service.
The company began examining whether a component involved in processing those checks was contributing to the failures. Engineers also investigated recent configuration changes and unusually high processor usage.
As the investigation continued, Microsoft identified a core server process that was becoming overwhelmed. The problem caused affected servers to stop responding, creating another obstacle to restoring normal capacity.
Recovery Efforts Encounter Additional Problems
Microsoft attempted several approaches to restore service, including redistributing connections across its global infrastructure and restarting affected servers.
Engineers also developed an alternative method for checking server health. Initial tests allowed some previously unavailable systems to return to service, and Microsoft began applying the change more widely.
However, the improvements were not immediately reliable. Some servers that initially recovered became unresponsive again after customer traffic returned.
Microsoft temporarily restricted certain types of network traffic in selected locations to reduce pressure on the affected infrastructure. Engineers also began rebuilding some servers using earlier software configurations to determine whether the changes would improve stability.
These measures were intended to restore service gradually while preventing additional failures.
The extended recovery illustrated how a problem affecting shared cloud infrastructure can disrupt numerous applications at the same time. Even services that remain technically operational may become difficult to use when they depend on a failing email or authentication-related component.
Businesses and Government Agencies Affected
The outage affected organizations that depend on Microsoft 365 for daily operations. Workers reported difficulty accessing messages, scheduling meetings and retrieving information needed to perform their jobs.
Government operations were also disrupted. Reuters reported that agencies in Australia and the United States were among those affected, including the U.S. House of Representatives.
The disruption extended beyond Microsoft's own applications. Some services that connect to Microsoft 365 experienced problems because they could not reliably communicate with the affected systems.
OpenAI and Anthropic reported issues involving Microsoft 365 connectors during the outage. These integrations allow users to connect certain workplace information and applications to other software platforms.
The problems demonstrated how a technical failure within one major cloud provider can affect outside companies without those companies experiencing a separate failure in their own infrastructure.
Businesses increasingly depend on connected software systems for communication, document management, security and other essential tasks. Interruptions in one shared service can therefore affect multiple business functions.
AWS and Google Users Also Reported Problems
Reports of technical difficulties involving Amazon Web Services and Google products appeared on outage-monitoring websites during the same period.
However, there was no confirmed evidence that those reports were caused by Microsoft's infrastructure problems. AWS, Google and Microsoft operate separate cloud platforms, and simultaneous user complaints do not necessarily indicate a common failure.
Downdetector collects reports submitted by users and other sources. Its figures can help identify unusual increases in complaints, but they do not independently establish the number of affected customers or explain what caused an outage.
The roughly 130,000 Outlook reports represented reported problems rather than a verified count of individual users who lost service.
The distinction is important because a single outage can generate multiple complaints from one user, while other affected customers may never submit reports.
For that reason, the confirmed incident centered on Microsoft 365 and its connected services, rather than a verified failure spanning all three cloud providers.
No Evidence of a Cyberattack
The scale of the disruption raised questions about whether malicious activity had contributed to the problems.
However, officials reported that Microsoft did not believe the outage resulted from a cyberattack. The company's investigation instead focused on server availability, automated health checks and configuration-related problems.
Australia's cyber security minister indicated that the incident appeared more consistent with a technical failure than a hostile intrusion.
Microsoft had not established a final root cause during the initial recovery efforts. Engineers continued examining the relationship between server failures, network connections and recent software changes.
A technical outage can have serious consequences even when no security breach occurs. Organizations may be unable to communicate with customers, access important records or complete routine operations.
Cloud Dependence Creates Business Risks
The incident renewed attention to the dependence of businesses and public institutions on a relatively small number of major cloud providers.
Microsoft 365 is widely used for email, messaging, document sharing and workplace collaboration. Many organizations depend on these products for functions that once operated through separate internal systems.
Cloud services offer advantages, including centralized management, automatic updates and access from multiple locations. However, centralizing essential operations can also increase the impact of a large infrastructure failure.
Companies that rely on a single provider for email and communications may have limited alternatives during a prolonged disruption. Backup communication methods and independent access to critical information can help reduce that exposure.
The November 12 outage also demonstrated the difficulty of restoring complex cloud platforms. Returning servers to operation does not necessarily solve a problem if those systems fail again when normal traffic resumes.
Microsoft continued working to stabilize its infrastructure and restore reliable connections. The company will need to determine why so many servers became unavailable and what changes can prevent a similar failure from spreading across its services.