LIVE
Quad Leaders Meet in Manila Today in First Summit in Two Years  •  United Airlines Hit by Systemwide Outage  •  Ramayana Crosses 1,437 Crore Worldwide
Tech

Microsoft Blames July Update Bug for Global Microsoft 365 Outage

By Toijam Bimoljit Singh • 17 November 2026 • 5 min read

Building 92 on Microsoft's headquarters campus in Redmond, Washington, with trees and lawn in front of it

Photo: Coolcaesar, CC BY-SA 4.0, via Wikimedia Commons — Building 92 at Microsoft's headquarters campus in Redmond, Washington, photographed in May 2016.

TL;DR — the direct answer: Microsoft has said a software bug introduced in a July update caused last week's global Microsoft 365 outage, which knocked out Outlook, Teams, SharePoint and Copilot for more than a day. In a preliminary post-incident report published on Sunday, 15 November, the company said a latent defect in a fix written for its Microsoft 365 networking gateway servers sat undetected for more than three months, because its load testing never reproduced the conditions that set it off. The bug was triggered when Microsoft Teams automatically shifted its traffic onto the Exchange Online gateway in the early hours of Thursday, 12 November, at a moment when nearly 14 percent of gateway capacity was already offline for an unrelated reason. The result was a cascading failure that left more than 80 percent of gateway servers unresponsive. Microsoft has ruled out a cyberattack and any role for artificial intelligence, and says its final report is due on 20 November.

What happened: Microsoft published the preliminary report to customers in the Microsoft 365 admin centre on Sunday, 15 November, and Charles Lamanna, the company's executive vice president for Copilot, agents and platform, posted an abridged version on X on Monday. The report says the faulty change was a fix prepared specifically for the Microsoft 365 networking gateway servers, and that it went through safe deployment from July, saturating globally over several weeks with no issues observed. For more than three months the change was deployed worldwide without impact. The defect caused lock contention inside an internal Windows process, exhausting a server's processor and memory until it stopped responding. At 01:07 UTC on Thursday, 12 November, Microsoft's systems automatically moved Teams traffic from a second, independent gateway onto the one that also carries Exchange Online, a normal part of operations. At the same time nearly 14 percent of server capacity for the networking gateway was out of service because of an unrelated issue from the previous day. The increased load and the reduced capacity set off the latent bug, and the failure cascaded across the gateway.

Why it's trending: The outage was one of the biggest of the year, with the tracker Downdetector logging more than 7.4 million reports worldwide and users affected in at least 74 countries. Businesses, schools and hospitals lost email and chat for more than a day, and the explanation is striking: a bug that sat quietly in production for over three months, missed by the company's own testing, then brought down a core piece of Microsoft's cloud in a matter of hours. The admission that Microsoft's test environment did not match reality is the detail drawing the most attention, because it raises questions about how a company of Microsoft's size validates changes to infrastructure that millions of organisations depend on. The report also matters because it is unusually candid for a company that rarely explains its failures in this much detail.

Background / context: The outage began at 11:20 a.m. Eastern time on Thursday, 12 November, and Microsoft confirmed full recovery at 11:10 a.m. Eastern time on Friday, 13 November, with a final customer update at 04:15 UTC on Saturday, 14 November. Exchange Online, the email system behind Outlook, was the worst-hit service. We covered the outage as it unfolded and its restoration in two earlier articles. The networking gateway sits in front of many Microsoft 365 services, receiving users' connections before passing them on to services such as Exchange, so a failure there can take out several products at once. Microsoft said its normal rollback, flighting and feature flags could not quickly revert the change, because the defect had been live for months rather than hours. The company applied an update reverting the faulty change from 07:20 UTC on Friday, 13 November. Microsoft also said the defect is specific to its Microsoft 365 gateway infrastructure and does not affect broader Windows deployments.

What we know / what's unconfirmed: Confirmed: the July origin of the defect, the three-month quiet period, the Teams traffic shift at 01:07 UTC on 12 November, the 14 percent capacity shortfall from an unrelated issue, the lock contention mechanism, the fact that load testing missed the bug because laboratory hardware was weaker and the traffic mix differed from production, the scale of more than 7.4 million Downdetector reports, and that Microsoft has ruled out a cyberattack and any role for AI agents. Unconfirmed: the longer-term fixes, since the preliminary report does not yet spell out what Microsoft will change about its testing or its rollback procedures for latent defects; and whether the final report due on 20 November will refine any part of this account. The report also does not mention any loss of customer data.

What happens next: Microsoft says it will publish its final post-incident report, with a full list of repairs, by 20 November. Customers and administrators will be watching for what changes to testing, deployment and rollback the company commits to, since the preliminary report identifies the testing gap but does not yet say how it will be closed. Microsoft has apologised for the disruption, and the incident is likely to push other large cloud providers to re-examine their own testing of long-lived changes. For users, the practical effect is over: Outlook, Teams, SharePoint and Copilot are back, and the outage is now a case study in how a small, quiet bug can hide for months and then fail all at once.

Frequently asked questions

Q: What caused the Microsoft 365 outage?
A: A latent software bug introduced in a July update to Microsoft's networking gateway servers. The defect caused lock contention in an internal Windows process, exhausting processor and memory until servers stopped responding.

Q: Why did Microsoft's testing miss the bug?
A: The company's load testing used weaker laboratory hardware with fewer processor cores and a different traffic mix from production, so it never reproduced the conditions that set the bug off. The defect is worse on newer machines with more cores.

Q: Which services were affected?
A: Outlook, Teams, SharePoint and Copilot were disrupted, with Exchange Online, the email system behind Outlook, hit worst. Downdetector logged more than 7.4 million reports worldwide.

Q: How long did the outage last?
A: It began at 11:20 a.m. Eastern time on Thursday, 12 November, and Microsoft confirmed full recovery at 11:10 a.m. Eastern time on Friday, 13 November, with a final customer update on Saturday, 14 November.

Q: Was the outage a cyberattack?
A: No. Microsoft said it was not aware of any malicious activity, and ruled out a cyberattack and any role for AI agents. It was a technical failure, not a security incident.

Sources: Computerworld — Microsoft outage: a cascading failure that started with an undetected bug, TechCentral — Microsoft 365 outage: inside the 24 hours that broke email, and TweakTown — Microsoft's global Outlook outage caused by a months-old bug and Teams.

← Back home