Azure: Azure Application Gateway 502 — Backend Pool Empty
Azure Application Gateway returning 502 because all backends were removed from the pool during a scale-in event.
Pattern
AZURE_CLOUD
Expected: AZURE_APPGW_502
Severity
HIGH
Confidence
68%
Remediation
Auto-Heal
Test Results
Metric
Expected
Actual
Result
Pattern Recognition
AZURE_APPGW_502
AZURE_CLOUD
Severity Assessment
CRITICAL
HIGH
Incident Correlation
N/A
None
Cascade Escalation
N/A
No
Remediation
—
Auto-Heal — Corax resolves autonomously
Scenario Conditions
Azure Application Gateway v2. Backend pool 'api-pool' has 0 healthy targets. Scale-in removed all instances. Min instance count set to 0 in auto-scale.
Injected Error Messages (1)
Azure App Gateway 502 — backend pool 'api-pool' empty, auto-scale set min instances to 0, all backends removed during scale-in, 100% of requests returning 502
Neural Engine Root Cause Analysis
Azure cloud infrastructure event detected — an Azure resource may be failing, an App Service is unhealthy, Azure AD authentication is disrupted, or a Service Bus queue is backed up. Azure outages can cascade across dependent services and affect both cloud-hosted applications and hybrid on-premises integrations relying on Azure AD.
Remediation Plan
1. Check Azure Service Health (status.azure.com) for any active incidents in your region.
2. Review Azure Monitor alerts and resource health for the affected service.
3. For App Service issues, check the Kudu console for application logs and restart the app if needed.
4. For Azure AD issues, verify conditional access policies and check the Azure AD sign-in logs for failure reasons.
5. For Service Bus, check dead-letter queues and verify the sending/receiving applications are connected and processing messages.
Improvements Applied
Pattern classified as AZURE_CLOUD (expected AZURE_APPGW_502)