Microsoft says July bug and server strain caused global outage


In its November 16 initial findings, Microsoft said increased gateway traffic and reduced server capacity triggered a defect from July, causing requests to pile up until servers stopped responding.
Situation: Microsoft’s November 2026 global outage
Microsoft says a months-old software defect, heavy traffic through a networking gateway and reduced server capacity combined to cause last week’s global outage. The company’s initial findings explain why servers stopped responding during peak US traffic and why recovery took longer than usual.
Microsoft executives said each problem might have been easy to resolve on its own, but the combination was harder to address, The Guardian said. Charles Lamanna, Microsoft’s executive vice president of Copilot, Agents, and Platform, explained the findings in a post on X.
Earlier: Microsoft resolves Exchange Online outage after network failure
Microsoft normally has two networking gateways, which connect systems and allow data to pass between them. Lamanna said one was unavailable for Microsoft Teams, overwhelming the primary gateway. More than 14% of server capacity was also unavailable because of an unrelated problem from the day before the outage. The earlier account had put the capacity unavailable at almost 14%.
Lamanna said the increased traffic and reduced capacity caused the gateway to encounter the July defect. Requests accumulated until servers stopped responding.

Microsoft marked the incident resolved on 13 November and attributed the failure to a network problem, not a cyberattack or superintelligence (SI) agents. It also warned that fraudulent notices used the outage to seek clicks.
As servers returned to service, waiting traffic immediately overwhelmed them and they became unresponsive again, Lamanna said. Microsoft’s automated recovery processes were not designed to restore service when servers were unresponsive on such a broad scale.
The defect’s age also made it harder to undo than a recent change. Lamanna said Microsoft’s usual rollback, staged-release and feature-control mechanisms could not quickly reverse it. Microsoft plans to publish a final report with further information.
3 min

Tom’s Hardware links the poisoning to a Microsoft outage that it says stopped safety emails from being sent. Local media cited by the outlet say 4 guests were taken to a hospital.
1 min

Microsoft said on 14 November that it had restored service to all remaining affected customers and regions after a worldwide outage that began on 12 November, and attributed it to an issue during an earlier network capacity upgrade.
23 min