Microsoft announced the full restoration of Microsoft 365 services after a global outage that lasted more than 24 hours. Issues with Outlook, Teams and Copilot have disrupted government agencies, stock traders and businesses around the world. “All products are operating as normal,” the company’s website says. Microsoft also confirmed that the emails that had accumulated during the outage have been processed.
The failure primarily affected the Exchange Online cloud platform and the functions of other Microsoft 365 services that depend on it. Several hundred thousand complaints were recorded on the Downdetector platform; Users from more than 70 countries reported problems. Employees of offices and government departments from Tokyo to London and New York had to switch to familiar means of communication: telephones and even faxes. Swiss holding company Richemont, whose companies produce luxury goods, and American transport manufacturer Workhorse were forced to postpone the publication of financial reports. The UK’s National Health Service has also faced challenges.
The prolonged outage has again raised questions about the reliability of Microsoft’s cloud services and its over-reliance on a single vendor. Microsoft 365 has more than 450 million paid commercial subscriptions, and its applications remain a critical part of the daily operations of businesses and government agencies. It is hardly possible to quickly abandon the company’s widely used products. At the same time, Microsoft shares remained almost unchanged in price.
According to preliminary data from Microsoft, the failure was caused by an error that was made earlier when expanding network bandwidth and appeared after a subsequent change in its configuration. Troubleshooting challenges arose: restored servers could once again become unresponsive after receiving load. Services were returned to service gradually, checking the stability of operation at each stage.
According to Bloomberg, after running on the servers, the Exchange processes that power Microsoft 365 email and calendars quickly exhausted available resources and became unresponsive. Engineers, in particular, checked whether the application was accessing databases instead of using data in local memory. A version of an error in the configuration of the Azure cloud network was also considered. The company has not yet disclosed the final technical reasons and promised to publish a detailed report after the investigation is completed. No signs of foul play were found; The problem is also not related to AI and agents, Microsoft assured.
Massive reports of problems began to arrive on the evening of November 12. The incident was reminiscent of last year’s massive AWS outage, which lasted 15 hours and affected hundreds of sites, and the CrowdStrike software solutions incident in 2024 that crippled organizations around the world.
This time, the glitch prevented even notification of employees about layoffs. PVH, which owns the Calvin Klein and Tommy Hilfiger brands, announced on November 12 its intention to cut about 4% of its workforce due to restructuring, but was unable to conduct some of the planned conversations with the laid-off workers, leaving them in the dark about their status. And the sheriff’s office in Colorado’s Arapahoe County was unable to process bail payments via email – people were asked to come and post bail in person.
If you notice an error, select it with the mouse and press CTRL+ENTER.
