Azure outages ripple across multiple dependent services • The Register

Microsoft's Azure cloud platform suffered two significant outages within 48 hours, causing widespread disruption across multiple dependent services and highlighting the fragility of cloud service interdependencies.

Managed Identity Service Disruption

On February 3, 2026, Microsoft reported a major outage affecting Managed Identity for Azure resources in the East US and West US regions. The incident began at 0015 UTC and lasted nearly six hours until 0605 UTC, impacting users attempting to create, update, delete, or acquire tokens for Azure resources.

The Managed Identity service is designed to eliminate the burden of credential management from developers by providing an identity for applications to use when connecting to resources that support Azure Active Directory authentication. When this service fails, it creates a cascade effect across the entire Azure ecosystem.

Cascade of Dependent Service Failures

The Managed Identity outage had far-reaching consequences, impacting numerous Azure services:

• Azure Synapse Analytics - Data integration and analytics platform
• Azure Databricks - Apache Spark-based analytics service
• Azure Stream Analytics - Real-time analytics service
• Azure Kubernetes Service (AKS) - Managed Kubernetes container orchestration
• Microsoft Copilot Studio - AI development platform
• Azure Chaos Studio - Chaos engineering service
• Azure Database for PostgreSQL Flexible Servers - Managed PostgreSQL service
• Azure Container Apps - Serverless container hosting
• Azure AI Video Indexer - Video analysis service

This extensive list demonstrates how critical identity management has become in modern cloud architectures, where services are deeply interconnected and dependent on shared infrastructure components.

Virtual Machine Management Outage

Just one day prior, on February 2, Azure experienced another significant disruption affecting Virtual Machine management operations. This outage impacted service management operations including create, delete, update, scaling, start, and stop operations for virtual machines.

The VM management outage affected an equally impressive list of dependent services:

• Azure Arc Enabled Servers - Hybrid server management
• Azure Batch - Large-scale parallel computing service
• Azure Cache for Redis - Distributed caching service
• Azure Container Apps - Serverless container platform
• Azure DevOps (ADO) - Development collaboration platform
• Azure Kubernetes Service (AKS) - Container orchestration
• Azure Backup - Cloud backup service
• Azure Load Testing - Performance testing service
• Azure Firewall - Network security service
• Azure Search - Cloud search service
• Azure Virtual Machine Scale Sets (VMSS) - Auto-scaling VM groups
• GitHub - Code repository and CI/CD platform

Root Cause Analysis

Microsoft identified the root cause of the Virtual Machine management outage as a configuration change that unintentionally restricted public access to certain Microsoft-managed storage accounts used to host virtual machine extension packages. This incident underscores a critical challenge in cloud operations: even seemingly minor configuration changes can have catastrophic downstream effects when services are tightly coupled.

The company acknowledged the issue at 1946 UTC on February 2 but did not provide a timeline for mitigation, leaving customers in the dark about when services would be restored.

GitHub Actions Impact

GitHub, which is owned by Microsoft, experienced degraded performance for its Actions CI/CD service starting at 1903 UTC on February 2. The incident was not fully resolved until 0056 UTC on February 3, affecting developers who rely on automated workflows for their software development processes.

Industry Implications

These consecutive outages highlight several important lessons for cloud computing:

1. Service Interdependencies: Modern cloud platforms are built on complex webs of interdependent services. When one service fails, the ripple effects can be extensive and unpredictable.

2. Configuration Management: The VM management outage demonstrates how configuration changes, even when intended to improve security or performance, can break critical infrastructure if not properly tested.

3. Transparency and Communication: Microsoft's delayed acknowledgment and lack of mitigation timelines frustrated customers who needed to understand the scope and duration of service disruptions.

4. Testing Practices: The incidents suggest that Microsoft may need to enhance its testing procedures for configuration changes, particularly those affecting core infrastructure components.

Broader Context

These outages come amid Microsoft's ongoing efforts to expand its cloud services portfolio. Recent developments include the retirement of TLS 1.0 and 1.1 protocols, the launch of Azure HorizonDB to compete with distributed PostgreSQL solutions, and various regional service disruptions affecting Azure OpenAI Service.

For organizations relying on Azure services, these incidents serve as a reminder of the importance of:

• Implementing robust monitoring and alerting systems
• Designing applications with graceful degradation in mind
• Maintaining backup plans for critical services
• Diversifying cloud provider dependencies where possible

The Azure outages demonstrate that even the largest cloud providers are not immune to significant service disruptions, and that the complexity of modern cloud architectures can amplify the impact of seemingly minor issues.