Microsoft restores all services, blames network upgrade

Microsoft apologized on Friday, Nov. 13, for a global outage that disrupted Outlook, Teams and Copilot for most of two days. The company said the problem was not a cyberattack and was not caused by artificial intelligence. Executive Charles Lamanna blamed "cascading failures + thundering herds at our network perimeter," which he said made recovery harder than usual. By Saturday morning, Microsoft's service health page showed "All products are operational," but the company said it would keep monitoring for an extended period because of how severe the incident had been.abc10.com+1

What Microsoft says went wrong

"This was not a cyberattack or an issue caused by AI tooling," Lamanna, executive vice president of Copilot, Agents and Platform, wrote on X. He said a post-mortem with a step-by-step timeline and "many repair items" would be published soon. In a separate post, Microsoft said "the incident was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change." The company said it had fixed the underlying issue.investors.com+1

Earlier updates described a chain reaction. Microsoft said the heavy load was not caused by more user traffic. Instead, a number of servers had "unexpectedly" become unavailable, which put strain on the rest of its infrastructure. Some servers became unstable again when traffic returned, so Microsoft changed its authentication and traffic-routing settings. The company said it brought services back gradually and checked stability at each step. That took longer but was meant to keep the restored services stable.usatoday.com+2

A global disruption

Problems started around midday Eastern time on Thursday, Nov. 12. Downdetector recorded about 130,000 reports of Outlook problems at the peak, and disruptions were reported in 74 countries. Australian intelligence and police agencies said their work was affected, as did Singapore's tax authority. Some schools in Britain asked parents to phone instead of sending email. By midday Friday, Microsoft said it was restoring service gradually to the customers still affected.news.bloomberglaw.com+2

Financial fallout

Microsoft shares dipped in early trading on Friday and then recovered. Evercore ISI analyst Kirk Materne estimated that about 100 million paid Microsoft 365 seats were affected. He said service-credit claims could reach $100 million to $400 million, which he called "relatively benign" compared with a quarterly business of more than $90 billion. He added that it was "not a great look" ahead of Microsoft's Ignite conference next week.investors.com

On Saturday, Microsoft said it was still clearing the backlog of emails that built up during the outage. It said it will not consider the issue resolved until all of them are delivered.theweek.in