Microsoft outage today: Azure, Microsoft 365, Xbox and Minecraft hit by global DNS fault; services gradually recovering
A broad Microsoft outage disrupted Azure, Microsoft 365 (including Outlook and Teams), the Xbox network, and Minecraft on Wednesday, with issues peaking around late morning U.S. time and afternoon UTC. Microsoft attributed the incident to a DNS-related fault that impaired access to cloud control planes and sign-in flows. Engineers implemented mitigations and traffic reconfiguration through Microsoft’s edge to restore reachability; service availability has improved, but some regions may still see intermittent errors as caches and backlogs clear.
What’s down, what’s back
-
Azure Portal & management APIs: Many users were unable to load the portal or saw timeouts on subscriptions, resource groups, and policy views. Programmatic calls fared better as mitigations rolled out; latency and sporadic 401/403 errors may persist while authentication caches refill.
-
Microsoft 365: Outlook, Teams, SharePoint/OneDrive, and admin tools had sign-in failures and presence/meeting join problems. Functionality is returning in waves as DNS propagation stabilizes.
-
Consumer services: The Xbox network (multiplayer, store, sign-in) and Minecraft services saw login and session disruptions; recovery is underway but can lag in certain geographies and ISPs.
-
Azure Front Door & dependent apps: Some front-ended workloads experienced elevated error rates until routing was adjusted.
No evidence points to a security breach; the pattern matches a configuration/availability incident in global name resolution.
Timeline (approximate)
| Time | Event snapshot |
|---|---|
| ~15:30–16:30 UTC (11:30–12:30 ET) | Outage reports surge; portal sign-in and M365 access degrade globally. |
| ~17:00–18:00 UTC | Microsoft confirms a DNS issue impacting multiple services; mitigation begins, including traffic reconfiguration. |
| ~18:30–20:00 UTC | Gradual restoration across Azure Portal, Microsoft 365, Xbox/Minecraft; residual errors continue for some tenants/regions. |
| Remainder of day | Backlogs clear; admin centers and consumer sign-ins stabilize as DNS caches refresh. |
Exact experiences vary by tenant, region, and ISP DNS caching behavior.
Why a DNS fault can break so much at once
DNS is the phone book of the internet. If lookups fail or return stale routes, clients can’t find sign-in endpoints, API gateways, or workload front doors—even when the underlying compute is healthy. At hyperscale, misconfiguration or propagation hiccups ripple globally: some users fail immediately, others only after caches expire, creating a rolling wave of disruption.
What IT admins should do right now
-
Retry from multiple paths. If the Azure Portal is flaky, use CLI/PowerShell/SDK against management endpoints; many operations succeed once tokens refresh.
-
Flush and verify DNS. Clear local resolver caches; confirm your resolver sees updated records. Consider testing against a secondary resolver for comparison.
-
Stage rollouts. Delay non-urgent deployments, policy updates, or large-scale config changes until status indicators are green for your region.
-
Communicate end-user workarounds. Advise staff to rejoin meetings via web clients, retry mailbox access, and avoid repeated password resets that can lock accounts.
-
Monitor backlog clearance. Queued jobs (automation, pipelines, email delivery) may process more slowly; watch for duplicate triggers.
What gamers should know
-
Xbox: Party chat, sign-ins, store purchases, and multiplayer may bounce during DNS cache churn. Power-cycle the console or network gear to refresh DNS.
-
Minecraft: Authentication and realm access should return as identity endpoints stabilize; expect staggered recovery by region.
How Microsoft is unwinding the incident
The company says engineers reconfigured traffic at the edge and updated DNS to restore endpoint resolution, then monitored telemetry while caches refreshed across providers worldwide. As name records stabilize, dependent services recover automatically, though admin consoles and status tiles can lag behind real-world availability.
What to watch over the next 24 hours
-
Post-incident summary: Expect a preliminary root-cause explanation referencing DNS configuration/propagation and a timeline of mitigations.
-
Regional tail risk: A small subset of users behind resolvers with long TTLs may see lingering failures.
-
Service backlogs: Mail queues, pipeline runs, and telemetry ingestion catch up; transient throttling is possible.
Quick answers to common searches
-
Is Microsoft/Azure down? Earlier today, yes—now recovering; some features may still be spotty.
-
Azure status page shows nothing active—why? Status pages can lag during propagation; history and RCA typically post after stabilization.
-
Is Microsoft 365/Outlook/Teams affected? Yes; sign-ins and presence were impacted and are improving.
-
Is Xbox/Minecraft down? Both saw sign-in and session issues; recovery is in progress.
-
Do I need to change anything permanently? No general action is required. If you pinned custom DNS settings with long TTLs, consider reviewing resolver policies to reduce future exposure.
A global DNS misfire briefly knocked key Microsoft cloud and consumer services offline. Mitigations are in place, recovery is well underway, and the remainder of the day should see continuing stabilization. Admins should favor programmatic management, flush DNS where needed, and hold major changes until telemetry and status indicators are solid across their regions.
