Digital
Microsoft blames faulty July update and traffic surge for global Outlook and Teams outage
Microsoft says a dormant bug in a July update to Microsoft 365 gateway servers, combined with a traffic surge and lost server capacity, took Outlook, Teams and Copilot down in more than 70 countries on Thursday, 12 November.
AI text based on reports by LaSexta, N-tv, europa press and other publishers. Published . How it is made
For membersThe first 30 seconds for everyone. Synthetic voice (AI).
In brief
Microsoft cites a defect in a Microsoft 365 gateway update rolled out from July, a Teams failover about 15 hours before the outage, and nearly 14% of server capacity out of service.
Microsoft says more than 80% of the relevant machines stopped responding on Thursday evening, 12 November, during US peak hours, and the failure spread to more than 70 countries.
Outlook was not restored until Saturday, 14 November, because restarted servers were overwhelmed by queued requests, so Microsoft built a special repair update and restarted servers in groups.
UK and Canadian hospitals and clinics switched to paper records in places, and Jaume Roura of Fecavem said Catalan car dealerships were paralysed for invoicing and vehicle records.
- July 2026Corrective update for Microsoft 365 gateway servers starts rolling out after passing safe deployment
- About 15 hours before the outageTeams fails over from its usual gateway to the one that also serves Exchange Online
- 12 NovOutage peaks Thursday evening, with more than 80% of relevant machines unresponsive
- 14 NovRepairs run into Saturday as restarted servers are overwhelmed, and servers are restarted in groups
Three causes behind the outage
Microsoft has published a preliminary account of the outage that knocked out Outlook, Teams and Copilot in more than 70 countries on Thursday, 12 November. Charles Lamanna, executive vice president for Copilot, Agents and Platforms, summarized the causes in a post on X and pointed to a preliminary post-incident report, available in the admin center under the references MO1550821 and EX1550543. The company blames three factors that combined into a chain reaction. The first was a latent defect in a corrective update for the network gateway servers that front Microsoft 365, which began rolling out in July after passing Microsoft's safe deployment process.
How the defect stayed hidden
For more than three months the update caused no visible problems, and testing suggested the servers were sound. Microsoft's tests had not checked how the servers behaved under heavy traffic, and the bug was especially harmful on newer, more powerful hardware that the test process did not use. A sudden rise in network load then made the gateways far more likely to hit the defect. About 15 hours before the outage began, Teams failed over from its usual network gateway to the one that also serves Exchange Online, which delivers Outlook email. That switch is a standard resilience measure, but it raised the load on that gateway and changed its traffic patterns.
Peak and spread
The second factor was a separate technical problem that had already taken nearly 14% of server capacity out of service, with Microsoft having scheduled those servers to return through routine processes. Combined with the load surge, this pushed the systems into the dormant defect during US peak hours. Microsoft says more than 80% of the relevant machines stopped responding at the height of the disruption on Thursday evening. Automatic traffic control then redirected requests to the remaining servers, which triggered the same defect there, and within less than two hours the failure had reached nearly the whole system. Digital protective barriers between world regions also failed, which let the outage spread around the globe.
- Server capacity out of service before the surge (nearly 14%)
- 14 %
- Relevant machines unresponsive at peak (more than 80%)
- 80 %
Data from the article text
Recovery
Teams, OneDrive and other services could be moved to alternative routes fairly quickly, but Outlook could not, because its email system was heavily loaded. Repairs stretched into Saturday, 14 November, since freshly restarted servers were immediately buried under queued requests and collapsed again. Microsoft eventually developed a special repair update and restarted the servers gradually in groups.
Impact on hospitals, offices and dealerships
Companies, courts, schools and public authorities were cut off for hours from email, calendars and online documents, and the German Federal Office for Information Security (BSI) received reports of the disruption. In Britain and Canada, hospitals and medical practices sometimes switched to paper records and postponed some appointments. In Catalonia, Jaume Roura, president of the vehicle distribution employers' group Fecavem, said the outage left dealerships paralysed, with invoicing and vehicle intake records affected. Staff fell back on phone calls and WhatsApp, while logistics and transport needed reworking at a cost in efficiency. Car factories were unaffected because they run different IT systems.
What comes next
Lamanna acknowledged the impact on customers in his post. Microsoft's account is preliminary, and the fuller detail is set out in the post-incident report accessible through the admin center.
Ask about this text
Sources and context
- LaSextaHow the Microsoft service outage affected the automotive worldAsí afectó la caída de varios servicios de Microsoft al mundo del motor · original in Spanish
- N-tvThe day: Cause for massive Microsoft outage knownDer Tag: Ursache für massiven Ausfall bei Microsoft bekannt · original in German
- europa pressMicrosoft explains the three causes of the massive Outlook outage that...Microsoft explica las tres causas de la caída masiva de Outlook que... · original in Spanish
- der StandardMajor Microsoft outage leads to 15 injured at PhantasialandMicrosoft-Großausfall führt zu 15 Verletzten im Phantasialand · original in German
- Places
- United States
- Organizations
- Microsoft
Pollar is free and carries no ads in the text. Its members keep it going.
Support Pollar