Particle.news

Microsoft 365 Outage Linked to Latent July Fix That Caused Global Cascade

Microsoft deployed a repair update that stabilized services and set 20 November for a final incident report.

Overview

  • Microsoft says a software change rolled out in July caused lock contention on newer gateway servers, and a traffic shift during U.S. peak hours on Thursday, 12 November triggered an automated rerouting cascade that left more than 80% of affected servers unresponsive within about two hours.
  • Engineers built and began deploying a repair update on the morning of 13 November that restored the Windows code path and restarted servers in controlled groups, with Teams, OneDrive and SharePoint diverted to alternate infrastructure while Exchange restoration took longer because of its scale.
  • The outage lasted into 13 November and affected services in roughly 74 countries, disrupting business, courts, schools and parts of health care and prompting analysts to estimate global productivity losses in the billions of dollars.
  • Microsoft has apologized, published a preliminary post‑incident report and said the cause was internal change rather than an attack; it faces SLA service‑credit claims from business customers while private users generally lack SLA entitlements.
  • The incident has renewed calls for greater digital resilience, including maintaining secondary mail and messaging, following 3‑2‑1 backup rules, rehearsing offline recovery plans and exploring regional or open‑source alternatives to reduce single‑provider risk.