Microsoft Azure challenges AWS for downtime crown

Steven

Microsoft Azure challenges AWS for downtime crown

Summary

Microsoft Azure experienced a global outage on 29 October 2025 beginning around 16:00 UTC. The fault affected Azure Front Door, Microsoft’s content delivery and edge service, and cascaded into many dependent services and customer platforms. Microsoft reported it reverted to a last-known-good configuration, temporarily blocked customer configuration changes and expected full service restoration by 23:20 UTC the same day.

The disruption hit a wide range of services and customers: airlines such as Alaska and Hawaiian reported website and check-in interruptions; Kubernetes package distribution (get.helm.sh) returned errors; Quebec’s Santé Québec suspended some patient access tools; and Microsoft’s own offerings including Outlook, Teams, Copilot and Xbox Live saw problems. Microsoft suggested customers consider redirecting traffic from Azure Front Door to origin servers via Azure Traffic Manager as a temporary failover.

Key Points

  • The outage began ~16:00 UTC on 29 October 2025 and affected Azure Front Door and many Azure-linked services.
  • Microsoft reverted to a last-known-good configuration and blocked customer config changes while mitigating.
  • High-profile customers were impacted: Alaska and Hawaiian Airlines, Helm (get.helm.sh), and Quebec health services among others.
  • Microsoft listed numerous affected services including App Service, Azure AD B2C, Azure SQL Database, Databricks, Azure Portal, and more.
  • Microsoft recommended using Azure Traffic Manager to redirect traffic from Front Door to customers’ own servers as a temporary failover.
  • The outage coincided with Microsoft reporting FY26 Q1 earnings, a quarter in which Azure revenue grew strongly year-on-year.
  • The event follows recent AWS outages, reviving debate about concentration risk in cloud platforms and prompting calls for resilience planning.

Context and relevance

This outage is significant because it demonstrates how a single CDN/edge service failure can ripple across many dependent platforms and consumer-facing applications. It arrives hot on the heels of recent AWS incidents, underscoring systemic risk from concentration in a few hyperscalers. Organisations that rely heavily on managed cloud edge and CDN services should review failover tactics, multi-region deployment and the feasibility of multicloud or on-prem fallback strategies.

Why should I read this

Short version: if you run services that sit behind Azure Front Door or depend on Azure-hosted components, this matters. Sites, check-in systems and developer tooling got hit — and that means real customer pain and potential revenue impact. Read on to see what Microsoft did, who got affected and quick steps you can apply now to reduce blast radius next time.

Source

Source: https://go.theregister.com/feed/www.theregister.com/2025/10/29/microsoft_azure_outage/