Failed Requests App Insights
Use this query in Application Insights (workspace-based) to inspect failed HTTP requests and operation context.
Data Source
| Table | Schema Note |
requests | App Insights table. Requires telemetry pipeline configured for the app. |
Query Pipeline
flowchart LR
A[Filter by cloud role name] --> B[Filter failed requests] --> C[Project operation metadata] --> D[Sort by time]
Query
let AppName = "my-container-app";
requests
| where cloud_RoleName == AppName
| where success == false
| project timestamp, name, resultCode, duration, operation_Id
| order by timestamp desc
Example Output
| timestamp | name | resultCode | duration | operation_Id |
| 2026-04-04T11:45:14.221Z | GET /api/items | 500 | 00:00:00.1820000 | 9f7a7d9d0bb84f0b |
| 2026-04-04T11:45:12.918Z | GET /health/dependency | 503 | 00:00:01.0040000 | f0e5946f613c4a49 |
| 2026-04-04T11:45:10.603Z | POST /api/orders | 504 | 00:00:02.0000000 | 4dbfc25be8c74999 |
Interpretation Notes
operation_Id links failed request to traces and exceptions. resultCode helps split ingress/proxy errors from app exceptions. - Normal pattern: low failed-request ratio aligned with SLO.
Limitations
- Requires App Insights instrumentation and sampling awareness.
- High sampling can hide low-frequency failures.
See Also