Microsoft suffered a widespread outage on Thursday and Friday, as engineers scrambled to fix one of its biggest-ever server issues, causing disruption to millions of its customers.
Connectivity issues began shortly before 1 p.m. ET on Thursday and continued into Friday, with Outlook and Teams, along with multiple Microsoft 365 services such as Sharepoint and Copilot, affected. Numerous external services that connect with Microsoft products such as Outlook Calendar were also hit.
In an update shortly after 7 a.m. ET, Microsoft continued to warn that users may be unable to connect to multiple Microsoft 365 products, including Outlook on either web or desktop and many services that interact with email.
However, it said it was making "encouraging recovery progress in Europe," with service availability and email delivery improving for some users. It said service is being "gradually restored" in the Asia Pacific and that the situation in North America was "improving." Intermittent issues are expected as recovery efforts continue.
Email and related tools were unavailable for Microsoft users on Thursday afternoon and evening, throughout the Asian business day on Friday and across the European and early U.S. morning.
Engineering teams have been taking affected systems offline for fixes and bringing them back online in stages. They are prioritizing efforts to preserve emails delayed during the recovery process and reduce email queues and backlogs, extending the timeout window, whereby delayed emails continue to attempt delivery, to 48 hours to help protect messages.
Microsoft says it has not detected any malicious activity causing the outage, which it initially blamed on a traffic surge in part of its infrastructure and issues with a component responsible for processing server health.
The company said in its initial updates on Thursday that the connectivity problems related to infrastructure "elevated load across multiple regions," caused by a "number of servers being unexpectedly unavailable; straining the remaining infrastructure." Some servers were failing automated health checks, it said.
Microsoft shares were around 2% lower in premarket trade.
Governments and businesses in Asia-Pacific and Europe said they were monitoring the situation, which was causing widespread disruption.
The U.K.'s National Health Service said the outage was affecting some of its services but that patients should attend appointments as normal. Calls "may be subject to long waits" as a result of disruption to the organization, a hospital group serving South West London and the English county of Surrey said on X.
Earnings for Swiss luxury goods giant Richemont were slightly delayed on Friday due to the outage. Electric truck maker Workhorse Group, which was expected to report third-quarter results on Nov. 12, rescheduled the release to Nov. 16, citing business disruptions caused by the global Microsoft outage.
Air India said it had "temporarily rerouted" emails so it could continue to receive customer messages, adding that its flight operations were not affected.
Australian Home Affairs and Cyber Security Minister Tony Burke said in a press conference that Microsoft "believe they are dealing in some ways similar to what we saw with respect to CrowdStrike where there is an update that has gone awry," Reuters reported.
A fault with an update issued by cybersecurity company CrowdStrike caused one of the world's biggest IT system outages in 2024, hitting airlines, banks and more,
Microsoft reported in 2025 that Copilot, its collection of AI assistants and tools, had more than 100 million monthly active users.
Amazon Web Services and Google users also reported outages on the site on Thursday.
Amazon wrote on X that services were "operating normally" and an "unrelated disruption elsewhere on the internet is prompting some inaccurate news reports." The company referred users to its health dashboard for "accurate data."
At 4 p.m. ET Thursday, Microsoft said it had "validated an alternative service health check method" that brought some affected servers back online.
"We're now applying this change more broadly," Microsoft wrote. "We're also preparing automated recovery systems to continually detect and restart unresponsive servers, to maintain service stability while this change progresses."
