Microsoft has revealed that more than 80% of the servers supporting its networking gateway became unresponsive at the peak of last week’s global outage, which disrupted Outlook and several Microsoft 365 services across the world. The company attributed the incident to a previously undetected software defect introduced in July 2026, which remained unnoticed for more than three months before a combination of technical issues triggered a widespread failure.

The outage affected businesses, airlines, financial institutions and millions of users who depend on Microsoft’s cloud-based services for everyday operations. According to Charles Lamanna, Microsoft’s Executive Vice President of Copilot, Agents and Platform, three factors combined to trigger the disruption: a hidden software bug, increased network traffic following a Microsoft Teams gateway change, and reduced server capacity. The company has apologised to customers and promised a detailed post-incident review.

Microsoft Identifies Three Reasons Behind Global Outage

Microsoft identified three separate technical factors that combined to trigger the global disruption.

The first was a hidden software defect introduced during a networking gateway update in July. The change was designed to improve the company’s Microsoft 365 networking infrastructure but contained a flaw that caused servers to become overwhelmed and unresponsive under certain conditions.

The software update remained active across Microsoft’s global infrastructure for more than three months without causing any known problems.

According to Lamanna, Microsoft’s load testing failed to identify the defect because the tests focused on older, weaker hardware. The problem was more severe on newer servers equipped with a greater number of processing cores.

The second factor involved a change in Microsoft Teams’ network routing. Approximately 15 hours before the outage, Teams traffic was shifted from an alternative gateway to the primary gateway that also supports Exchange Online.

This increased the workload and changed traffic patterns across the primary gateway.

The third factor was reduced server capacity. Nearly 14% of the gateway’s server capacity was already unavailable because of an unrelated technical issue that had occurred the previous day.

Together, these three conditions exposed the previously undetected software defect and triggered a cascading failure.

More Than 80% of Servers Became Unresponsive

The combination of increased traffic and reduced server capacity caused Microsoft’s networking gateway to experience severe performance problems.

As servers started losing connectivity, Microsoft’s traffic management systems automatically redirected users to other functioning endpoints.

This mechanism is normally designed to maintain service availability when individual servers experience hardware, software or network failures.

However, during this incident, redirecting traffic increased the workload on the remaining operational servers.

As more servers became unresponsive, traffic was distributed across an increasingly smaller pool of available infrastructure.

This created a cascading failure that spread across Microsoft’s server fleet in less than two hours.

At the lowest point, more than 80% of the servers behind the networking gateway were unresponsive, according to Lamanna.

The failure also affected safeguards that normally separate different sections of the networking gateway. A separate issue prevented these protections from containing the disruption, allowing the impact to spread across regions.

Outlook, Teams and Microsoft 365 Services Face Disruptions

The outage primarily affected Microsoft 365 services connected to Exchange Online, the company’s cloud-based email platform.

Users experienced problems accessing Outlook through desktop applications and web browsers. Email delivery was delayed, while some customers were unable to access their mailboxes.

Other Microsoft services, including Teams, OneDrive, SharePoint and Office applications, also experienced disruptions.

However, these services were able to move away from the affected gateway and recover earlier. Teams chats and meetings remained operational for much of the outage.

Exchange Online faced greater difficulties because it could not use the same alternative routing arrangement.

The disruption highlighted how several widely used business applications depend on shared cloud infrastructure.

Even when individual products operate separately, failures in common networking systems can affect multiple services simultaneously.

Why Microsoft’s Recovery Took Longer Than Expected

Microsoft’s recovery efforts faced additional challenges after the initial technical failure.

Normally, when servers become unavailable, automated recovery systems attempt to restore their operations and redirect traffic.

However, during this incident, servers returning to service were immediately overwhelmed by the large volume of waiting requests.

As a result, some recovered servers became unresponsive again.

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

Because the software defect had been introduced several months earlier, the company’s usual recovery methods could not quickly reverse the change.

Engineers eventually developed an update that restored the software code to its previous state.

The company then removed groups of servers from service, applied the update, restarted them and gradually brought them back online.

This process allowed Microsoft to restore services while managing the additional traffic accumulated during the disruption.

Indian Airlines and Financial Companies Also Affected

The global outage created operational challenges for businesses across multiple industries, including India.

Domestic airlines IndiGo, Air India and Akasa Air reported disruptions to their email-based customer support systems.

The problems affected their ability to receive and respond to some passenger queries.

Air India temporarily redirected its email services to ensure that customer communications continued reaching the airline.

Flight operations were not reported to have been affected by the email disruption.

Financial services companies also experienced difficulties accessing research reports and conducting client interactions.

The incident demonstrated how heavily businesses depend on cloud-based email and collaboration platforms for communication, customer support and daily operations.

At its peak, the outage generated more than 1.3 lakh reports of Outlook problems on Downdetector, with disruptions reported across approximately 72-74 countries.

Microsoft Promises Detailed Investigation Into Outage

Microsoft has apologised to customers affected by the disruption and acknowledged the impact on their business operations.

Lamanna said the company would provide a final post-incident review with additional technical details later in the week.

The review is expected to provide further information about the software defect, the failure of existing safeguards and the measures being introduced to prevent similar incidents.

The company will also need to evaluate why its testing procedures failed to identify the defect before it was deployed globally.

The incident raises important questions about software testing, server capacity planning, automated traffic management and disaster recovery systems.

For businesses relying on Microsoft 365, the outage also highlights the importance of maintaining alternative communication and operational arrangements.

The Bigger Picture

Microsoft’s latest outage demonstrates how a relatively small software defect can trigger widespread disruption when it interacts with increased traffic and reduced infrastructure capacity. The fact that more than 80% of the affected networking gateway servers became unresponsive highlights the risks associated with shared cloud infrastructure and interconnected enterprise services.

The incident also raises questions about the effectiveness of Microsoft’s software testing and disaster recovery systems. As businesses increasingly depend on cloud computing, email platforms and AI-powered workplace applications, maintaining service reliability is becoming just as important as expanding computing capabilities.

Looking Ahead

Microsoft is expected to publish a detailed post-incident review explaining the technical failures and the corrective measures being introduced. The company will likely focus on improving software testing across different hardware configurations, strengthening network isolation and ensuring that automated recovery systems can handle large-scale failures without triggering additional disruptions.

For enterprise customers, the incident serves as a reminder that even major cloud providers can experience widespread service interruptions. Businesses may increasingly look at backup communication systems, alternative infrastructure and stronger continuity plans to reduce their dependence on a single technology provider.

Get the day’s top stories in your inbox

One concise email. No spam, unsubscribe anytime.