Outlook, Teams and Copilot have been hit by a global disruption affecting businesses and government institutions. Microsoft says its initial recovery measures have failed to deliver the expected results, forcing engineers to explore alternative ways to restore capacity

Microsoft’s global technology outage has entered a second day, with the company still struggling to restore services after its first recovery attempts failed to deliver the expected results. The disruption has affected Outlook, Teams, Copilot and other Microsoft 365 applications across 74 countries and territories, leaving businesses, government agencies and institutions facing problems with essential communications and online operations.

The scale of the incident has exposed a vulnerability extending well beyond Microsoft’s email platform. The company’s cloud-based infrastructure supports a vast network of business applications and public services, meaning that a failure affecting Exchange Online can have consequences across several interconnected products. The disruption has also reached third-party services that depend on Microsoft’s systems, including integrations used by OpenAI and Anthropic.

The Microsoft 365 outage began on Thursday, November 12, and quickly spread beyond the Americas, where the first problems were reported. Downdetector recorded nearly 130,000 reports of Outlook problems in the United States at the peak of the disruption. Reports also came from Europe, Asia, Africa and other parts of the world.

A server failure with consequences across the world

Microsoft initially attributed the connectivity problems to elevated workloads affecting parts of its infrastructure in multiple regions. Its subsequent investigation, however, revealed a more complicated situation: a significant number of servers had unexpectedly become unavailable, forcing the remaining infrastructure to handle additional traffic.

The company found that processes responsible for establishing secure connections on affected servers were becoming overwhelmed, causing the machines to stop processing requests. Some servers also failed automated health checks and became unavailable, further reducing the capacity of the network.

The result was a cascading problem. As Microsoft attempted to bring affected servers back online, some became unresponsive again once they began receiving traffic.

The company has been investigating the possibility that a recent configuration change contributed to the disruption. It has also started reversing a network configuration change that coincided with the beginning of the incident. Microsoft has not yet established a definitive root cause, but has said it has no indication that the outage was caused by malicious activity or a cyberattack.

The recovery effort has involved restarting affected servers, redirecting traffic through alternative infrastructure and restricting certain types of connections in specific locations to reduce pressure on the remaining systems.

Microsoft admits its recovery measures are not working

The most concerning development for customers came in Microsoft’s successive updates on X. After monitoring the changes introduced to improve service availability, the company acknowledged that the measures were not producing the expected results.

“We’ve monitored the status of the changes applied to improve serviceability and confirmed they’re not taking as expected,” Microsoft said.

The company added that it was exploring alternative approaches to increase available capacity.

“We’re identifying alternate workstreams that can be applied to increase capacity in the face of sustained traffic management imbalances,” the update said.

Microsoft had previously reported limited improvements in some regions after returning a portion of the affected servers to service and shifting connections to alternative routes. But the recovery remained uneven, with some restored servers struggling again when traffic resumed.

By Friday morning, November 13, the company was reporting positive results from its efforts to restart affected infrastructure and was monitoring server stability before extending the recovery measures. Full restoration, however, had not yet been confirmed, and Microsoft had not provided a definitive timetable for the return of all services.

Outlook, Teams and Copilot among the affected services

Although Outlook has generated the largest number of complaints, the disruption has extended across the wider Microsoft 365 ecosystem.

Exchange Online, the infrastructure underpinning Microsoft’s email services, has been particularly affected. Problems have also been reported with Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender XDR.

Not every feature has been equally affected. Microsoft has indicated that some Teams functions, including chats and online meetings, remained largely operational in certain regions, while features relying on Exchange Online, such as calendar functions and meeting scheduling, experienced problems.

The consequences have also reached beyond Microsoft’s own products. OpenAI reported problems with its Microsoft 365 connector, affecting new connections and the operation of certain tools. Anthropic similarly confirmed that its Microsoft 365 connector was experiencing failures involving email, calendars, search and other functions, with Claude for Outlook also affected.

The disruption therefore extended into the infrastructure used by companies developing some of the world’s most prominent artificial intelligence services.

Government agencies and institutions caught up in the outage

The geographical reach of the incident has been particularly striking. According to Downdetector, reports came from users in 74 countries and territories within approximately an hour of the initial disruption.

New Zealand was among the countries affected. Its Ministry of Foreign Affairs confirmed that its ability to receive emails had been compromised, with consequences extending to all New Zealand embassies, consulates and representative offices around the world.

Government operations were also affected elsewhere. In Japan, officials confirmed that the disruption had reached government IT infrastructure, while authorities in Australia reported problems across public administration, schools and other institutions. Some government departments had to activate alternative procedures to maintain communications and essential operations.

The impact has been particularly significant for organizations that rely heavily on Microsoft’s cloud-based services for internal communications, administrative functions and the exchange of information.

For many businesses and public institutions, the problem has not simply been an inaccessible email inbox. It has meant losing access, temporarily or partially, to several interconnected tools used to coordinate everyday operations.

Microsoft is continuing to work on restoring the affected infrastructure, while monitoring the stability of servers already brought back online. The company has yet to publish a complete account of what triggered the failure or provide a definitive timetable for the full restoration of its services.