Cloud · Microsoft
Azure Front Door and Azure CDN global connectivity and DNS outage
- Started
- Oct 29, 2025, 15:41 UTC
- Resolved
- Oct 30, 2025, 00:05 UTC
- Duration
- 8h 24m (from provider timestamps)
- Affected
- Azure Front Door, Azure CDN, Azure Portal, Azure App Service, Azure SQL Database, Azure Active Directory B2C, Microsoft 365, Microsoft Entra ID (some components), Microsoft Dynamics 365 and Power Platform, Microsoft Sentinel (Threat Intelligence), Microsoft Purview, support case creation
- Cause
- Final reviewA sequence of valid customer configuration changes made across two different control plane build versions produced incompatible configuration metadata that exposed a latent bug in the Azure Front Door data plane. The crash surfaced asynchronously, about five minutes later, so the configuration passed the protection system's health checks, reached most edge sites and was saved into the Last Known Good snapshot. Edge servers crashed worldwide, which also took down AFD's internal DNS service.
- Was it an attack?
- No, per the provider or investigators
- State
- resolved
Anatomy
1. Trigger
Configuration change
2. Already broken
Latent bug, Missing validation
3. What made it spread
Global configuration propagation, DNS dependency
4. Blast radius
global: Global
5. Cause category
Change or deployment
Labels follow the root-cause taxonomy. A person's change is recorded as the trigger and the missing safeguard as what was already broken; “human error” is never a cause here.
How the explanation changed
October 29, 2025 · Final review
“A specific sequence of customer configuration changes, performed across two different control plane build versions, resulted in incompatible customer configuration metadata being generated. These customer configuration changes themselves were valid and non-malicious – however they produced metadata that, when deployed to edge site servers, exposed a latent bug in the data plane.”
What the provider committed to change
- Fixed the control plane incompatibility and the data plane bug (completed)
- Synchronous processing of each customer configuration before production stages (completed)
- Additional rollout stages with extended bake time, including a pre-canary stage (completed)
- Decouple configuration processing from traffic-serving instances (est. January 2026)
- Backwards-compatibility validation across control plane builds (est. February 2026)
- Active-active fail-away for critical first-party services (completed)
- 'Micro cell' segmentation of the AFD data plane (est. June 2026)
- Data plane recovery within one hour (completed), target ~10 minutes (est. March 2026)
- Faster and broader Azure Service Health alerts; support-case failover
Attack or not
“These customer configuration changes themselves were valid and non-malicious”
Microsoft Azure, Post Incident Review YKYN-BWZ, October 29, 2025 (Azure status history data feed (page 2), PIR YKYN-BWZ)
Post-incident reviews
Other Microsoft incidents
- Microsoft 365 global outage: latent networking gateway defect cascades across the fleet (EX1550543 / MO1550821), November 12, 2026
- Azure West US loses connectivity to the WAN after an over-scoped break-fix repair, July 23, 2026
- Azure East US control plane failures, April 24, 2026
- Azure West US power event, February 7, 2026
Sources
- Azure status history: Post Incident Review (PIR), Azure Front Door, Connectivity issues across multiple regions, Tracking ID YKYN-BWZ (provider, read November 17, 2026)
- Microsoft Tech Community, Azure Networking Blog: AFD resilience post (2025-12-18) (provider, read November 17, 2026)
- Azure status history data feed (page 2), PIR YKYN-BWZ (provider, read November 17, 2026)
Last verified November 17, 2026. Spot an error? The methodology page says how corrections work.