Overview
- Microsoft said issues began around 16:00 UTC on Oct. 29, producing timeouts and access failures across Azure and downstream Microsoft 365 services, with user reports peaking at over 18,000 for Azure and nearly 11,700 for Microsoft 365 on Downdetector.
- Engineers blocked further Azure Front Door changes, moved the Azure Portal off AFD, rerouted traffic, and deployed a last known good configuration before recovering nodes and rebalancing traffic.
- By Wednesday evening, Microsoft reported services stabilizing with error rates and latency back to normal for most users, while customer configuration changes to AFD remained temporarily blocked and some intermittent failures persisted.
- The outage affected a broad range of platforms and organizations, including Outlook, Teams, Xbox, Minecraft, Copilot, and operations at Alaska Airlines, Vodafone, and Heathrow Airport.
- The disruption came days after a major AWS incident, sharpening concerns from analysts and industry watchers about concentration risk across a small number of hyperscale cloud providers.