Microsoft has revealed that more than 80% of the servers supporting its networking gateway became unresponsive during last week’s global outage, which disrupted Outlook and several Microsoft 365 services. The incident, which began on November 12, affected businesses and users across multiple countries, leaving many unable to access email, manage calendars, or use certain workplace applications.

The company has attributed the disruption to a combination of three factors: a hidden software defect introduced in July, increased network traffic following a change involving Microsoft Teams, and reduced server capacity. According to Microsoft, these issues combined to trigger a cascading failure that spread across its networking infrastructure. The company has apologised to customers and is reviewing the incident to prevent similar disruptions.

What Happened During Microsoft’s Global Outage?

The outage affected Microsoft’s networking gateway, which sits in front of several Microsoft 365 services and helps direct user requests to the appropriate infrastructure.

At the peak of the incident, more than 80% of the servers behind this gateway were unresponsive. The disruption primarily affected Exchange Online, Microsoft’s cloud-based email platform, leaving Outlook users unable to access their mailboxes through web and desktop applications. Some emails were also delayed.

Other Microsoft services, including Teams, OneDrive, SharePoint and Office applications, experienced disruptions. However, some of these services were able to move to alternative infrastructure and recover earlier than Exchange Online.

Microsoft executive Charles Lamanna, who leads the company’s Copilot, Agents and Platform organisation, explained the incident in a public update on November 16. He apologised to customers and acknowledged the impact on their everyday business operations.

Three Factors Behind the Microsoft Outage

Microsoft identified three conditions that combined to trigger the failure.

1. A Hidden Software Defect Introduced in July

The first problem was a software defect that had remained undetected for more than three months.

The defect originated in a fix developed specifically for Microsoft’s Microsoft 365 networking gateway servers. Under certain workload conditions, it caused servers to become overwhelmed and stop responding.

The change began rolling out in July 2026 and was eventually deployed worldwide. However, it did not cause an immediate disruption.

Microsoft said its regular load testing failed to identify the problem because the tests used weaker hardware. The defect was more pronounced on newer machines with more processing cores, meaning the testing environment did not accurately reproduce the conditions that eventually triggered the failure.

2. Microsoft Teams Increased Traffic on the Primary Gateway

The second factor involved a change in how Microsoft Teams connected to the company’s networking infrastructure.

Approximately 15 hours before the outage, Teams switched from an alternative gateway to the primary gateway, which also supports Exchange Online.

This was part of Microsoft’s normal operational process. However, the change increased the workload on the primary gateway and altered its traffic patterns.

The additional demand made the infrastructure more vulnerable to the software defect that had been present since July.

3. Nearly 14% of Server Capacity Was Already Unavailable

The third factor was an unrelated infrastructure issue that had reduced the gateway’s available server capacity by nearly 14%.

Although this capacity reduction was not enough to cause an outage on its own, it left the network with fewer resources to handle the increased workload.

The combination of reduced capacity, additional traffic and the hidden software defect eventually pushed the gateway beyond its ability to operate reliably.

How the Failure Spread Across Microsoft’s Servers

The outage became significantly worse because of how Microsoft’s traffic management systems responded when individual servers stopped working.

Under normal circumstances, these systems automatically redirect requests away from unavailable servers and send them to healthy endpoints. This process is designed to maintain service availability when hardware, software or network problems occur.

However, during this incident, the redirection process unintentionally increased pressure on the remaining operational servers.

As more servers became unresponsive, their workloads were transferred to fewer available machines. Those machines then faced additional traffic and encountered the same software defect.

This created a cascading failure that spread across the gateway infrastructure.

Microsoft said the problem affected its server fleet in less than two hours. At the lowest point, more than 80% of the gateway servers were unresponsive.

The company also acknowledged that safeguards designed to separate different parts of the gateway infrastructure did not prevent the failure from spreading.

Why Did Microsoft’s Recovery Take So Long?

Restoring the affected servers proved more difficult than Microsoft initially expected.

One major challenge was the backlog of requests that accumulated during the outage. When engineers brought servers back online, those machines were immediately overwhelmed by waiting traffic and became unresponsive again.

Microsoft also said its automated recovery processes were not designed to handle a failure of this scale.

Another complication was the age of the software defect. Since the underlying change had been introduced months earlier and deployed worldwide, the company’s usual rollback procedures could not quickly reverse it.

Engineers eventually developed an update that returned the affected code to its previous state. They then restored the infrastructure in stages by taking groups of servers offline, applying the update, restarting them and gradually bringing them back into service.

Microsoft confirmed that services had fully recovered by November 13, although the incident lasted more than 24 hours.

Impact on Businesses and Indian Companies

The outage affected more than individual Microsoft users. Businesses that depend on Microsoft 365 for communication, customer support and daily operations also experienced disruption.

In India, airlines including IndiGo, Air India and Akasa Air reported problems with email-based customer support. The disruption affected their ability to receive or respond to some passenger queries.

Air India temporarily rerouted its email services to help ensure customer communications continued reaching the airline. Flight operations were not reported to have been affected.

Financial services firms also faced difficulties accessing research reports and conducting client interactions. The incident demonstrated how a disruption at a major cloud service provider can affect organisations across different industries.

The scale of the outage was also reflected in reports from Downdetector, which recorded a significant spike in complaints from users experiencing Outlook problems.

What Is Microsoft Doing to Prevent Another Outage?

Microsoft has apologised for the disruption and said it is working to improve the resilience of its networking infrastructure.

The company has described the incident’s contributing factors and its recovery process, but its investigation is continuing. A final post-incident review is expected to provide additional details.

The incident also raises questions about how software updates are tested before they are deployed across large-scale cloud infrastructure. Testing environments need to account for different hardware configurations, traffic patterns and workload conditions.

Microsoft’s experience shows that a software change can remain unnoticed for months before a particular combination of circumstances exposes the underlying defect.

The Bigger Picture

The outage highlights the risks associated with the concentration of business operations on a small number of major cloud and software providers. Microsoft 365 is deeply integrated into the daily operations of companies, government organisations and other institutions. When a critical part of its infrastructure becomes unavailable, the effects can spread far beyond Microsoft’s own products.

It also illustrates why redundancy and automatic traffic rerouting are not sufficient on their own to guarantee resilience. In this case, systems designed to protect availability contributed to the spread of the failure because the remaining servers could not handle the redirected traffic. Businesses therefore need contingency plans that allow essential communication and operations to continue even when a major technology provider experiences an outage.

Looking Ahead

Microsoft’s next challenge will be to identify why its testing processes failed to detect the software defect and ensure that similar issues cannot spread across its networking infrastructure. The company’s final post-incident review should provide further information about the technical causes, failed safeguards and changes planned to strengthen recovery systems.

For businesses, the incident is another reminder to assess their dependence on cloud platforms and prepare alternative communication and operational arrangements. As organisations increasingly rely on connected services for essential work, the ability to continue operating during a major technology outage is becoming an important part of business continuity planning.

Get the day’s top stories in your inbox

One concise email. No spam, unsubscribe anytime.