International Edition
Latest News
Technology

Microsoft 365 Global Outage Caused by Latent Bug, Traffic Surge and Cascading Failure

Microsoft 365 services experienced a global disruption lasting 23 hours and 50 minutes, beginning on Nov. 12 at 16:20 UTC and concluding on Nov. 13 at 16:10 UTC. The outage, which impacted millions of people worldwide, stemmed from…

Microsoft AI chip

Microsoft 365 services experienced a global disruption lasting 23 hours and 50 minutes, beginning on Nov. 12 at 16:20 UTC and concluding on Nov. 13 at 16:10 UTC.

The outage, which impacted millions of people worldwide, stemmed from a combination of a dormant software defect, a surge in network traffic, and a failure of system safeguards.

The failure began when a software defect, introduced in a July 2026 update, encountered specific load conditions on newer networking gateway servers. Although the code had been deployed for months, it remained dormant until a traffic shift occurred.

Approximately 15 hours before the outage, Teams switched from an alternate gateway back to the primary gateway, which also supports Exchange Online. This move increased the load on the primary gateway, which was already operating at reduced capacity due to an unrelated issue from the previous day.

“A latent defect caused servers to become overwhelmed and unresponsive under certain load conditions,” the company said in its preliminary post-incident report. Microsoft acknowledged that its testing environment did not accurately reflect the infrastructure used in production. While the company typically conducts load testing, it focused on hardware with fewer processor cores, failing to identify the contention issues that surfaced on newer, higher-capacity servers.

As the gateway servers became unresponsive, the system’s traffic management protocols attempted to redirect users to healthy endpoints. This caused a cascading failure, as the redirected load overwhelmed the remaining functional servers. At the peak of the incident, more than 80% of the networking gateway servers were unresponsive.

Recovery efforts were further complicated by the nature of the defect. Because the underlying issue was embedded in the code, standard automated recovery processes and rollbacks were ineffective. Engineers were forced to manually disconnect servers in batches, apply a hotfix, and restart them to restore service incrementally.

The disruption affected a wide range of critical infrastructure, including the Philippine Supreme Court, Singapore’s court systems, and Japan’s Government Solution Service, which government employees use for daily work. U.S. House staff said the outage ended a day of legislative work. Law enforcement agencies had trouble processing bail payments, and Richemont delayed its quarterly results. British minister Luke Pollard urged “greater resilience” against technology failures.

Microsoft 365 Global Outage Caused by Latent Bug, Traffic Surge and Cascading Failure
Photo: computerworld.com

Technical Failure Causes

Factor Contribution to Outage
Software Defect Introduced in July 2026; triggered by high-core server loads.
Traffic Surge Teams moved to the primary gateway 15 hours before the failure.
Reduced Capacity Nearly 14% of gateway server capacity was offline due to prior issues.

Frequently Asked Questions

Why did standard rollback procedures fail?
The defect was a latent bug present in the system for several months. Because the problematic code had been part of the infrastructure since July, standard “flighting” and feature-flag rollbacks could not revert the environment to a stable state.

Was the outage the result of a cyberattack?
Microsoft quickly ruled out malicious activity. “a technical issue not related to cyber or AI,” Microsoft said in a post on X.

Microsoft was founded in 1975 to market BASIC interpreters for the Altair 8800 microcomputer. Since those origins, the organization has grown into the world’s largest software company by revenue and a prominent leader in cloud computing, artificial intelligence, and video gaming.

Known for its influence on the rise of personal computing through the Windows operating system and the Microsoft 365 suite, the company has evolved significantly from its early days in the computer programming business to become one of the most valuable public companies globally.

What is the status of the affected services?
All impacted Microsoft 365 services, including Teams, Exchange Online, OneDrive, and SharePoint, were fully restored by the morning of Nov. 14.

About the author: Technology Desk

Technology Desk at Archynewsy.