New Delhi: Microsoft is gradually restoring its cloud services following a global outage that disrupted Outlook, Teams-related functions and other Microsoft 365 products on Thursday. Recovery progressed across several regions on Friday, although some users continued experiencing delayed emails and intermittent connection failures.
In its latest service-health update, Microsoft reported improvements in email delivery and service availability across Europe, comparatively stable services in Japan and signs of recovery in North America. Recovery measures had already been completed across parts of its Asia-Pacific infrastructure, but the incident remained classified as service degradation.
Engineers continued restoring affected infrastructure and clearing accumulated messages. Microsoft also extended its email-delivery timeout window to 48 hours to help protect delayed correspondence while systems recovered.
Global outage disrupts governments, hospitals, airlines
According to the Associated Press, the disruption affected businesses and government operations from Asia to the United States.
Microsoft said services dependent on Exchange Online, including Outlook, calendars, some Teams scheduling functions and Copilot’s access to workplace information, experienced problems. However, core Teams messaging and online meetings largely remained operational.
The BBC reported nearly 140,000 user-submitted Outlook complaints across 74 countries at Thursday’s peak. AP separately cited approximately 130,000 reports, which subsequently declined. Neither figure represents the total number affected, while Microsoft 365 applications have more than 430 million users worldwide.
Singapore reported government email failures, the Philippines’ supreme court experienced operational disruptions, and Japanese authorities arranged alternative communications. Hong Kong businesses and banks also reported problems, while US manufacturer Workhorse Group postponed an earnings call.
In the United States, courts from Los Angeles to Texas warned of delayed communications. A Toronto healthcare network also reported email disruptions across several hospitals, although patient care remained unaffected, according to the BBC.
In Britain, schools, police forces, GP surgeries and pharmacies experienced communication difficulties. Some NHS administrative and prescription-related processes were disrupted, although essential clinical services continued in many locations.
In India, Air India temporarily rerouted customer emails, while IndiGo reported interruptions to email-based support. Both airlines confirmed that flight operations remained unaffected.
OpenAI and Anthropic also experienced problems connecting to Microsoft 365 services, highlighting the disruption’s impact on third-party AI platforms, as RNA Media reported earlier.
What went wrong with Microsoft’s infrastructure?
Microsoft initially attributed the disruption to elevated load on its infrastructure. Investigators subsequently found that numerous servers had unexpectedly become unavailable, forcing the remaining machines to handle additional traffic.
The increased load was not necessarily caused by unusually high customer demand. Instead, reduced server capacity placed greater pressure on functioning infrastructure, creating further instability.
Engineers identified an overwhelmed process responsible for establishing secure connections, which caused affected servers to stop responding. Microsoft’s initial recovery attempts proved unsuccessful, with some restarted servers becoming unresponsive again when traffic returned.
The company subsequently applied changes to authentication and routing components, restarted affected infrastructure and gradually reintroduced traffic. These measures produced improving stability, although Microsoft had not established a definitive root cause.
Microsoft said it had found no evidence of malicious activity and would publish a detailed post-incident report after recovery.
Australian home affairs minister, Tony Burke, said Microsoft had indicated that the disruption might resemble the July 2024 CrowdStrike outage, when a faulty software update disrupted airlines, hospitals and businesses worldwide. However, Microsoft has not confirmed that a software update caused the latest incident.
Why the outage raises wider concerns
Cybersecurity expert, Troy Hunt, told the BBC that Microsoft’s central role in digital infrastructure meant the disruption could affect much more than workplace communication, including appointments, online purchases and account verification.
Former software engineer, Gergely Orosz, described it as potentially one of the most disruptive business email outages, although the full extent of its impact remained uncertain.
The incident highlighted how dependence on interconnected services from a handful of technology providers can disrupt multiple business functions simultaneously.
While Microsoft works towards complete restoration, the disruption underscores the importance of alternative communication channels, secure backups and regularly tested business-continuity arrangements.