Particle.news

Microsoft 365 Services Restored After Global Outage

Microsoft says a dormant July software bug plus a recent network change and capacity shortfall triggered an automated cascade that left large parts of its infrastructure unresponsive.

Overview

  • A cascading failure began on Thursday evening when a latent bug in code introduced in July combined with a planned Teams switchover and about 14% reduced server capacity, causing more than 80% of the affected servers to stop responding.
  • Microsoft rolled out a special repair update and restarted servers in staged groups, stabilising services by Saturday while apologising to customers and saying the incident was not caused by hackers or artificial intelligence.
  • The outage disrupted core Microsoft 365 features including Outlook, Teams, OneDrive and Copilot across more than 70 countries and forced some hospitals, courts, schools and offices to use paper records or alternative communications.
  • A full independent or official post‑incident report is still pending, and impacted organisations face routine incident-reporting duties under rules like NIS2 even when the root cause lies with a vendor.
  • Experts and officials are pressing for stronger digital resilience by keeping tested fallbacks, using secondary providers, following the 3-2-1 backup rule, and considering staged adoption of open‑source or self‑hosted tools as longer-term alternatives.