Microsoft's Outlook Outage Ran Nearly a Day as Its Fixes Fell Short

Microsoft traced the outage to an earlier network capacity upgrade and says it's holding delayed email for up to 48 hours.

Tokyo bond traders went back to phones and fax machines to close deals as Microsoft's Outlook outage stretched toward a full day. Ryutaro Kimura, a senior bond strategist at BNP Paribas Asset Management, told Bloomberg that desks were still getting trades done "by phone and fax," Time reported, and Tokyo International Air Cargo Terminal told customers to call or fax rather than email.

The trouble started Thursday at 1:26 p.m. Eastern, which is 10:26 a.m. in Los Angeles, according to Microsoft's incident timeline as BleepingComputer reported it. It hit Outlook and Exchange Online, Teams, SharePoint and Copilot. By midday Friday Eastern time, almost 24 hours in, Microsoft said email queues were nearly back to normal. It was still warning that some users would see spotty access and late mail.

Downdetector logged nearly 130,000 reports of Outlook problems, Time reported. Baltimore County said its staff couldn't send or receive email, though 911 kept working. In Britain, some National Health Service practices told patients to phone instead.

The outage dragged on because the fixes kept falling short. Microsoft first pointed to heavy load across several regions. Then it said on its service health page that traffic hadn't gone up at all: a large number of servers had unexpectedly dropped out and pushed their work onto the rest, XDA Developers reported. It froze all deployments and configuration changes.

By Thursday afternoon, Microsoft had found another way to run its server health checks. That brought one batch of servers back, and the company started applying it more broadly, XDA reported. It didn't fix the outage. Later Thursday, Microsoft said its recovery work hadn't brought the relief it expected, BleepingComputer reported.

The new diagnosis was that a process on each affected server that sets up secure connections was getting swamped until the server quit answering. Engineers started rolling back a recent network configuration change whose timing matched the start of the outage. By Thursday evening, Microsoft was reporting limited recovery. By Friday morning, some users could receive mail but still couldn't send it, XDA reported.

In a statement Time quoted, Microsoft said the outage was "not related to cyber or AI/agents." It blamed a problem from an earlier upgrade of its network capacity that showed up after a more recent network change. In Australia, the outage hit the intelligence agency ASIO and the Australian Federal Police. There, Home Affairs Minister Tony Burke compared it to CrowdStrike and said Microsoft believed it was dealing with "an update that has gone awry," according to Time.

The 2024 CrowdStrike meltdown, caused by a third-party security firm, grounded flights and took TV stations off the air. This time, by Microsoft's own account, the trouble started inside its network. Charles Lamanna, the Microsoft executive vice president over Copilot, agents and platform, said in a statement to BleepingComputer that the company regrets "the impact these outages had on those who depend on us" and is reviewing how to keep it from happening again.

For now, delayed email is still trickling into inboxes. Microsoft says it has stretched its email timeout window to 48 hours so those messages aren't dropped while it clears the backlog. The fax machines in Tokyo didn't need an extension.

Source: ktla.com.

Source: time.com, retrieved November 13, 2026. Other sources: bleepingcomputer.com, xda-developers.com.