Microsoft’s outage began shortly before 1 p.m. ET Thursday, as reports of Outlook problems surged. The company initially blamed “elevated load across multiple regions” and said it was testing a change to bypass a component linked to failed server health checks, while the cause and recovery timeline remained unclear.¹

The disruption quickly spread beyond email. Downdetector logged nearly 130,000 Outlook reports from U.S. users, with reports coming from 74 countries and territories within about an hour.² Outlook, Teams, SharePoint and Copilot were among the affected services, with some users also unable to reach other Microsoft 365 tools. Microsoft later said unexpectedly unavailable servers were straining the rest of its infrastructure; it was restarting servers and redistributing traffic.³

The outage carried into San Francisco’s city offices. Supervisor Bilal Mahmood said staff had shifted much of their communication to phone calls: “It’s definitely stalled certain things.” But the city’s Building Inspection Department said it could keep serving customers, while Public Works staff turned to texts, phone calls and in-person meetings. By evening, Microsoft reported “positive results” from server restarts, though recovery remained limited.⁴

Microsoft identified a more specific failure: a server process responsible for secure connections was becoming overwhelmed, and some servers became unresponsive again after returning to service. The company said it was limiting some traffic as it investigated, and stated it had “no evidence of malicious activity associated with this incident.”⁵

By Friday morning, Outlook was beginning to recover after about 18 hours of disruption. That eased the immediate pressure, but the slow restoration left a broader question: why did Microsoft’s own infrastructure prove so difficult to bring back online?⁶