--:--:-- ET
New York 62°F Partly cloudy Los Angeles 78°F Sunny Chicago 55°F Windy
Subscribe
U.S. Edition Est. 2026 Nov. 21, 2026

US News America

Independent reporting for a changing America

Breaking
Technology

Microsoft Blames Months-Old Windows Fix for 24-Hour Outage That Cut Off Outlook and Teams

Microsoft said a defect in a Windows fix sent to its gateway servers in July set off a cascading failure that left Outlook, Teams and Copilot out of reach for more than a day in over 70 countries. Services were fully restored on Saturday, and a final report is due around November 20.

A dark modern office at night with rows of desks and blank computer monitors, lit by desk lamps and city lights through large windows.

Microsoft has traced the global outage that cut off Outlook, Teams and Copilot for more than a day to a defect in a Windows fix sent to its gateway servers in July, according to a preliminary incident report dated November 15.

The company announced early Saturday that it had fully restored Microsoft 365 services after the outage, which lasted more than 24 hours and affected users in more than 70 countries. Microsoft said the problem was a technical issue introduced during a network capacity upgrade and not a cyberattack or an AI agent, and it apologized to customers, saying it was “deeply sorry” for the disruption.

The report, for incident EX1550543, says the defect sat inside a Windows fix built for the gateway servers that carry Microsoft 365 connections. It had been rolling out since July and ran for more than three months without trouble. Under certain conditions, it caused lock contention inside an internal Windows process, which exhausted CPU and memory on the server. Microsoft says the defect does not apply to general Windows deployments, so customer machines were never part of the problem.

A long corridor inside a data center with rows of tall server racks and small blinking indicator lights.

A cascade that spread across the fleet in under two hours

Two other conditions lined up with the defect. About 15 hours before the outage, Teams traffic had failed over to the primary gateway that also serves Exchange Online. Nearly 14 percent of the gateway servers were already out of service because of an unrelated issue the day before. Microsoft said the remaining capacity was enough on its own, but the three factors together pushed servers into the defect.

Once the first servers failed, their traffic moved to healthy ones, which then failed in turn. Microsoft describes a cascading failure from server to server, and says the defect was met across the fleet in less than two hours. At the worst point, more than 80 percent of gateway servers were unresponsive.

The outage ran from 16:20 UTC on November 12 to 16:10 UTC on November 13, about 23 hours and 50 minutes. Exchange Online took the hardest hit because, unlike Teams, OneDrive, SharePoint and the Office apps, it could not be moved off the faulty gateway path for reasons of scale, according to Microsoft. Mailboxes were unreachable and inbound mail queued for hours, and Microsoft stretched its delivery window so that delayed messages would not bounce back as undeliverable.

A laptop with a dark blank screen on a wooden desk beside a cup of coffee, in a home office with morning light.

Recovery came in stages. Microsoft applied a fix to part of the capacity at 07:20 UTC on November 13. Asia Pacific recovered around 11:00, Europe returned to healthy levels at 11:35, and global webmail availability was back to expected levels at 14:17. Customers were no longer actively affected from 16:10, and Microsoft issued its final customer updates on November 14 at 04:15.

Hospitals, courts and lawmakers felt the outage

The disruption reached beyond office workers. Britain’s National Health Service reported a disruptive impact, and British police forces had trouble receiving online crime reports, though emergency phone lines kept working. House staff members told Politico the outage had set back an entire day of legislative work.

A modern glass office building at dusk with lit windows, seen from a quiet street.

Microsoft lists five actions as completed: connection handling code was returned to a known good state, the email delivery window was extended, extra capacity was added for recovery, monitoring of traffic redirection and connection health was increased, and new guardrails were put on traffic distribution to stop a failure from cascading across the fleet again. A final report with the full list of repairs is due within five business days of full resolution, which points to around November 20.

Microsoft 365 is used by governments, businesses and institutions around the world, and the company has reported 450 million paid commercial seats. The outage was one of the most widespread service disruptions in recent years.

Reporting based on coverage by SMTPedia and Zotpaper.