Microsoft said a global Microsoft 365 outage came from a network problem, Reseller News reported on 15 November 2026. A spokesperson said the fault began in an earlier network capacity upgrade and surfaced after a recent network change. Microsoft said it was not related to cyber or AI/agents.
Most affected services are back, Microsoft said, and it is still working with a small number of customers and regions. Reseller News said recovery took longer than first expected. Microsoft said that, given the scale of demand, it restored services gradually and checked stability at each stage. The company apologised. Microsoft has not said how the fault went unnoticed, how many customers were hit, how long the outage lasted, or when the review will appear. It has not named the exact mechanism.
Analysis, not from the sources: a network change can leave a hidden flaw, such as a wrong setting, that does nothing until a later change or heavier load sets it off. Microsoft has not said whether that happened here. On Microsoft's account, this was a change failure, not a security event. Buyers can plan for that: keep an offline fallback for email and calendar, check what your contract credits for downtime, and ask for the review's root cause.