Microsoft Says 365 Is Fully Restored After Outage of More Than 24 Hours, Promises Post-Mortem

Since Microsoft first flagged trouble just before 1 p.m. ET on Thursday, Nov. 12, the Microsoft 365 outage has gone from a recovery that kept stalling to a declared finish. On Saturday, Nov. 14, the company said it was done.

"We have completed our efforts to address the underlying issue of the incident and restored service to the remaining impacted customers and regions," Microsoft said in a post on X.

Reuters put the disruption at more than 24 hours. The Guardian put the stretch before recovery began, just before midday GMT Friday, at roughly 18 hours. Either way, it ranks among the longest in the company's history. It is also the second disruption to Microsoft's services this year, according to Reuters.

What Microsoft says went wrong

Microsoft's explanation is narrow. The company said the incident "was introduced by an issue during an earlier network capacity upgrade that surfaced" after a recent network change. Servers then could not cope.

The company said the failure hit features tied to Exchange Online, its cloud email and scheduling hub. Applications that do not connect to it kept working. Organizations that host Microsoft software on their own servers were somewhat spared.

Microsoft also said the outage was "not related to cyber" or artificial intelligence agents, and that it saw no indication of malicious activity. That is the company's own account of its own system. It has not yet been tested by any outside review.

Charles Lamanna, Microsoft's executive vice president of Copilot, Agents and Platform, said "technical challenges and failure modes" made recovery "more difficult than normal." He said a full post-mortem with a step-by-step timeline and a list of repair items would be published "shortly."

Who got hit

The reach was wide. Downdetector logged nearly 130,000 problem reports across 74 countries and territories, covering Outlook, Teams, Word, Excel, SharePoint, OneDrive and Copilot.

Governments in Japan, Singapore and Australia said their operations were affected. In the U.S., the House of Representatives and city governments including Jacksonville, Austin and Dallas reported problems. Municipalities in France did too.

In Britain, hospitals and GP practices were disrupted. Several police forces told people to phone in crimes instead of reporting them online. Some schools moved attendance reporting to phone calls, and National Rail warned that emailed tickets might not be accessible.

A trader told Reuters that foreign banks and financial institutions in Taiwan were hit especially hard because their data is emailed to overseas headquarters. Airlines warned customers of communication lags, according to public statements, and some companies were hindered from publishing earnings.

OpenAI said new connections through its Microsoft connectors might be broken. Anthropic said an upstream provider issue was affecting its Microsoft 365 connector.

More than 430 million people use Microsoft 365 apps, according to Microsoft's website. The company said in January it had more than 450 million paid commercial subscriptions.

The cost and the dependence question

A think-tank estimate cited by CNN put the cost of lost productivity at at least $7 billion. That is an estimate, not a measured loss, and Microsoft has not commented on it.

The larger argument is about concentration. Jeremy Singer, a professor at the University of Glasgow's School of Computing Science, said the episode "serves to emphasize the dependence we have built on one company's digital infrastructure, and may lead us to question whether we need more resilience in our IT ecosystem."

Walter Pasquarelli, an AI adviser to governments and a University of Cambridge researcher, called it "first and foremost a communication risk." He said government services and infrastructure rely heavily on these platforms. Some lawmakers outside the U.S. have argued the outage shows the consequences of overreliance on American tech.

Microsoft's side of that argument is an apology. "We are deeply sorry for the disruption this incident has caused our customers and partners," the company said Friday. "We know the trust they place in our services and take that responsibility extremely seriously."

Wall Street analyst Dan Ives of Yorkville Ives and Co. wrote on X that "there will be major questions for MSFT in this global outage."

Government agencies and public services that run on a single vendor's email and meeting tools are spending public money on that dependence. The outage showed police, hospitals and courts having to fall back on phones.

The next concrete step is Lamanna's post-mortem. It will show whether Microsoft's account of a capacity upgrade followed by a network change holds up, and what it plans to change so one bad configuration cannot take email down across 74 countries again.