On November 12-13, 2026, Microsoft 365 services experienced a significant outage that was traced back to a hidden bug in Windows. This issue overwhelmed the servers responsible for managing cloud connections, leading to widespread service disruptions.
Initial Findings on the Outage
According to a preliminary incident report released on November 15, the problem originated from a software modification that inadvertently redirected traffic and diminished server capacity. This resulted in extensive service failures, with Exchange Online being the most affected. The outage persisted for nearly 24 hours, restricting many users from accessing their Exchange Online mailboxes and causing email delays.
The incident extended its impact to other Microsoft cloud services, as initially reported by Cybersecuritynews.com. This disruption highlighted the critical role of robust server management in cloud services.
Understanding the Hidden Windows Bug
The bug was linked to a fix intended specifically for Microsoft 365 networking gateway servers. Although this change was deployed globally in July and appeared stable for months, it did not concern general Windows deployments. The issue arose from competing tasks within an internal Windows process, known as lock contention, which exhausted processor and memory resources under specific traffic conditions.
Two main factors exposed the bug. First, a shift in Teams traffic to a primary gateway shared with Exchange Online occurred just hours before the outage. Second, an unrelated issue had already reduced server capacity by nearly 14%. Despite sufficient remaining capacity, these combined conditions triggered the bug.
Recovery Efforts and Future Outlook
Microsoft’s automated recovery systems inadvertently exacerbated the situation by transferring connections to functional endpoints, thereby spreading the fault across regions rapidly. At its peak, over 80% of servers were non-responsive. Efforts to contain the issue failed due to a secondary fault in boundary controls, leading to a global outage.
To address the problem, engineers developed an update that restored Windows connection handling to a stable state. They systematically removed large server groups from service, applied the update, and gradually reintroduced traffic. Additional capacity was added to prevent email delivery failures.
While the report is still preliminary, further details are expected on November 20. Microsoft reassured administrators that no tenant or device configuration changes were necessary, as the failure resided within Microsoft’s cloud infrastructure.
This incident underscores the importance of proactive monitoring and robust testing in cloud service management, emphasizing the need for comprehensive strategies to prevent similar occurrences in the future.
