A Microsoft 365 outage that began on November 12, 2026 at 16:20 UTC disrupted Outlook, Exchange Online, Teams and Copilot worldwide for nearly 24 hours, ending at 16:10 UTC on November 13. Microsoft’s preliminary Post Incident Report, published November 15, traces it to three problems that hit the networking gateway behind Microsoft 365 at the same time: a software defect that had sat undetected since a July 2026 update, a traffic surge from a Teams failover, and reduced server capacity on the gateway fleet.

What broke, and for how long

  • Services affected. Outlook, Exchange Online, Microsoft Teams and Copilot were disrupted worldwide. Reporting on the incident tracks it under two Microsoft 365 admin center incident IDs: EX1550543 for the Exchange Online portion and MO1550821 for the wider Microsoft 365 impact.
  • Duration. Microsoft’s preliminary report gives an incident window from 16:20 UTC on November 12 to 16:10 UTC on November 13, 2026 — just under 24 hours. Some coverage of Microsoft’s own posts about the incident describes a longer span from first notice to final restoration confirmation; the report’s own start and end times are the figures used here.
  • Scope. Microsoft describes this as a global disruption, not limited to one region or data center.

What Microsoft says caused it

Microsoft’s preliminary report describes three factors converging on the same networking-gateway infrastructure:

  • A dormant defect from a July update. A software bug introduced in a July 2026 change to the Microsoft 365 networking gateway went undetected for months because the conditions that triggered it had not occurred since the change shipped.
  • A Teams failover. Microsoft describes the failover itself as a routine part of its normal resiliency operations, not a mistake on its own. It pushed a surge of traffic onto the gateway layer.
  • Reduced server capacity. The gateway fleet was already running with less spare capacity than usual when the failover hit, so the combination of more load and the latent bug left a large share of gateway servers unable to process requests.

Microsoft states plainly that the outage was not caused by a cyberattack and was not related to any AI system malfunctioning.

What is still open

Microsoft’s report is explicitly preliminary. The company has said a final Post Incident Report will follow once the investigation is complete; at the time of writing, only the preliminary version was available, and figures such as the exact share of affected gateway capacity differ slightly between secondary reports summarizing it. A final report could adjust the timeline or add detail, the way Microsoft’s final review of a September Azure gateway outage named more affected services than its own preliminary version had.

This is the same class of failure our guide to reading a post-incident report describes: a shared piece of infrastructure, in this case a networking gateway, failing in a way that reaches several products at once. It is also another entry in the pattern covered in why one cloud outage can break unrelated apps — Outlook, Teams and Copilot are different products that depend on the same gateway layer. For more confirmed provider incidents with their causes and durations, see our cloud outage tracker.