Overview
- Microsoft’s Exchange Online and connected Microsoft 365 services began failing on Thursday, Nov. 12, producing hundreds of thousands of user reports and outages across about 74 countries.
- Engineers traced the problem to servers that failed automated health checks and an elevated load that strained remaining machines while they tried traffic rerouting, restarts and other fixes.
- By Nov. 13 the company reported broad recovery and early stabilization though some regions and users still experienced errors, queued email backlogs and delayed deliveries.
- Microsoft says there is no indication of malicious activity, the company has fixed the underlying issue it ties to a prior network capacity upgrade, and it will publish a full post‑incident analysis.
- The outage disrupted governments, hospitals, courts and businesses, broke connectors used by AI firms such as OpenAI and Anthropic, and has renewed concerns about reliance on a small number of cloud providers.