Microsoft’s cloud-based services, including Microsoft 365, Outlook and Teams, suffered a major global outage on Thursday, disrupting business operations and communications for users across several countries.
At the height of the disruption, more than 130,000 users reported problems accessing Outlook on outage-tracking platform Downdetector, with complaints recorded in 74 countries.
The disruption, which began around 1 p.m. US Eastern Time on November 12, also affected some functions of SharePoint Online, Microsoft 365 Copilot and Defender XDR. However, the impact varied across services, with general messaging and online meetings on Teams largely remaining operational.
Microsoft attributed the incident to infrastructure problems involving Exchange Online, its cloud-based email platform. The company said some servers unexpectedly became unreachable, preventing automated health checks from running and placing additional pressure on functioning servers across multiple regions.
The resulting overload triggered widespread connection failures, particularly for Outlook users accessing their emails through web, desktop and mobile applications.
Microsoft disclosed that it had identified the affected infrastructure and was taking steps to prevent the disruption from spreading to additional servers. It also temporarily suspended new deployments and configuration changes globally while redistributing traffic and testing alternative health-check procedures.
By 5 a.m. Taiwan time on Friday, November 13, Microsoft said it had validated an alternative health-check method that restored some affected servers. The company also began restarting problematic servers and developing an automated recovery process.
Outage reports subsequently declined, with Downdetector recording approximately 31,000 Outlook complaints by 9:30 a.m. Taiwan time.
Microsoft’s executive vice-president of its Copilot business, Charles Lamanna, ruled out a cyberattack, saying there was no indication that malicious activity had caused the incident. He apologised to affected customers.
Although users of Amazon Web Services and Google also reported disruptions around the same period, Amazon maintained that its services were operating normally, describing reports linking its infrastructure to the incident as unrelated.
The outage also affected some artificial intelligence services. OpenAI reported connection disruptions and tool errors involving its Microsoft connector, while Anthropic said an upstream supplier problem had affected its Microsoft 365 connector.
Meanwhile, the disruption had consequences for businesses relying on Microsoft’s cloud infrastructure. Electric truck manufacturer Workhorse Group postponed the release of its third-quarter earnings, originally scheduled for November 12, to November 16, citing business disruption caused by the outage.
Microsoft 365 is a critical business platform used by organisations worldwide. The company’s January earnings report disclosed that it had more than 450 million paid commercial Microsoft 365 seats, underscoring the scale of businesses potentially exposed to disruptions affecting its services.
The incident has also renewed concerns about the operational risks associated with heavy dependence on major cloud service providers, particularly the ability of businesses to maintain communication and essential operations when critical infrastructure experiences widespread failures.
While no data loss or security breach was reported in connection with the incident, the disruption could raise questions about service reliability, customer confidence and Microsoft’s contractual service-level commitments.
Under its Exchange Online service-level agreement, Microsoft guarantees 99.9 per cent monthly availability for eligible customers, with service credits available when specified availability thresholds are breached.