Incident Summary
Between 07:58 UTC on February 7, 2026 and 04:24 UTC on February 8, 2026, customers using Azure services in the West US region experienced intermittent service unavailability, timeouts, and higher than normal latency due to an unexpected power interruption affecting a single datacenter.
| Field | Details |
|---|---|
| Duration | 20 hours 26 minutes |
| Root Cause | Unexpected power interruption at West US datacenter |
| Impact | West US region only |
| Severity | High (localized but prolonged) |
Timeline
| Time (UTC) | Event |
|---|---|
| Feb 7, 07:58 | Power interruption begins at West US datacenter |
| Feb 7, 08:15 | Service health alerts triggered |
| Feb 7, 08:45 | Microsoft confirms power-related issue |
| Feb 7, 12:00 | Partial restoration of services |
| Feb 7, 18:00 | Backup power systems fully engaged |
| Feb 8, 02:00 | Primary power restored |
| Feb 8, 04:24 | Full service recovery confirmed |
Services Impacted
Azure Virtual Machines (West US)
- Intermittent connectivity failures
- Failed start/stop operations
- VM extensions timing out
Azure Storage (West US)
- Increased latency on blob/file operations
- Timeout errors for storage account access
- Queue and table service disruptions
Azure Functions (West US)
- Function execution failures
- Trigger processing delays
- Timeout errors
App Services (West US)
- Web apps returning 503 errors
- Deployment failures
- Slow response times
Impact Assessment
Regional Impact: Only customers with resources deployed in the West US region were affected. Multi-region deployments with failover to other regions experienced minimal disruption.
Business Impact:
- Web applications hosted in West US experienced downtime
- Storage-dependent workflows slowed or failed
- Function-based automation disrupted
- Increased support ticket volume
Microsoft's Response
- Power Restoration — Backup power systems engaged while primary power issues resolved
- Service Recovery — Rolling restart of affected services as power stabilized
- Customer Communication — Status updates via Azure Service Health Dashboard
- Datacenter Review — Investigation into power infrastructure resilience
Lessons Learned
For Azure Customers
- Geographic Redundancy — Deploy critical workloads across multiple regions
- Traffic Manager — Use Azure Traffic Manager or Front Door for automatic regional failover
- Monitoring — Implement multi-region health checks and alerting
- SLA Credits — File for service credits if you experienced SLA violations
For Cloud Providers
Physical infrastructure failures (power, cooling, network) remain a significant risk even in hyperscale cloud environments. Redundancy at the datacenter level is critical.
Current Status
Resolved: All West US services fully recovered as of 04:24 UTC on February 8, 2026.
Microsoft confirmed that primary power has been restored and backup systems are fully operational. Datacenter infrastructure is undergoing review to prevent recurrence.
Related Incidents
This power outage occurred just 4 days after the major Azure VM and Identity outage, marking the third significant Microsoft cloud outage in 2026.