Back to All Scenarios
PASSEDcloud / azure_app_service_outage
Azure App Service Outage — 502 Errors
An Azure App Service Plan hosting 3 production web apps starts returning 502 Bad Gateway errors after an Azure platform update. The apps intermittently crash with out-of-memory exceptions. Azure Status page shows degraded performance in East US 2 region.
Pattern
AZURE_CLOUD
Severity
CRITICAL
Confidence
85%
Remediation
Auto-Heal
Test Results
| Metric | Expected | Actual | Result |
|---|---|---|---|
| Pattern Recognition | AZURE_CLOUD | AZURE_CLOUD | |
| Severity Assessment | CRITICAL | CRITICAL | |
| Incident Correlation | Yes | 35 linked | |
| Cascade Escalation | Yes | Yes | |
| Remediation | — | Auto-Heal — Corax resolves autonomously |
Scenario Conditions
Azure App Service Plan P2v3. 3 web apps (customer portal, API, admin). East US 2 region. Azure platform update triggered. App Insights showing 40% error rate. Auto-scale maxed at 10 instances.
Injected Error Messages (3)
Azure App Service 502 Bad Gateway — customer-portal.azurewebsites.net returning HTTP 502, App Insights: System.OutOfMemoryException in w3wp.exe, platform issue detected in East US 2, Azure incident ID: VM9K-T80
API backend health check failing — HTTP 502 from api-backend.azurewebsites.net, Azure App Service instances restarting in loop, Application Insights exception rate: 847/min, availability: 58%
Azure App Service Plan at maximum scale — 10/10 instances running, all showing elevated memory (95%+), Azure Resource Health: Degraded, platform event: Host node maintenance in progress, mitigation ETA unknown
Neural Engine Root Cause Analysis
The Azure App Service is experiencing memory exhaustion (System.OutOfMemoryException in w3wp.exe) causing the application worker process to fail and resulting in 502 Bad Gateway responses. This appears to be compounded by a broader Azure platform issue in the East US 2 region (Azure incident ID: VM9K-T80). The 14 correlated incidents within the same timeframe strongly suggest this is part of a regional Azure infrastructure problem affecting multiple services simultaneously.
Remediation Plan
1. Restart the Azure App Service to clear memory issues and reinitialize the worker process 2. If restart fails or issue persists, scale up the App Service plan temporarily to provide more memory resources 3. Monitor Azure Service Health for updates on the East US 2 platform issue 4. Consider failover to secondary region if available 5. Implement application-level memory optimization if pattern continues after platform issue resolves
Tested: 2026-03-30Monitors: 3 | Incidents: 3Test ID: cmncjk6ig0258obqe8zt4g7fx