An Azure service that provides hosted, universal storage for Azure app configurations.
Use a step-by-step check along the telemetry pipeline to find where data stopped flowing.
- Verify ingestion endpoint and authentication
- If the Application Insights data collection endpoint is configured to use Microsoft Entra ID, and the application is still sending with only an instrumentation key/connection string, telemetry will be rejected and not appear in the portal even though the app is running.
- Confirm whether Microsoft Entra ID authentication was enabled on the Application Insights resource. If so, configure the app to authenticate accordingly as described in Enable Microsoft Entra ID (formerly Azure AD) authentication.
- Check for SDK/agent misconfiguration
- Missing telemetry can be caused by misconfigured SDK or agent, even if no intentional changes were made (for example, a deployment that changed environment variables or connection strings).
- Validate that the Application Insights SDK/agent is still enabled and using the correct connection string/ikey, and that the host is configured according to the language-specific guidance.
- Test connectivity from the host
- From the application host, send test telemetry to the ingestion endpoint using PowerShell or curl as described in the troubleshooting guidance.
- If these tests fail, investigate:
- DNS resolution of the ingestion endpoint.
- Firewalls or gateways blocking TCP connections to the ingestion endpoint.
- TLS version mismatch (endpoint requires TLS 1.2 while the app uses TLS 1.0/1.1).
- Azure Monitor Private Link or DNS overrides that may be resolving the ingestion endpoint to the wrong private IP.
- Check for backend rejection or throttling
- Even if usage is below the daily 100 GB limit, telemetry can still be dropped or throttled per-minute or due to service health issues.
- Use an HTTP inspection tool (for example, Fiddler) or SDK logs to confirm that telemetry uploads are returning HTTP 200 from the backend. If not, data may be rejected due to missing required fields, oversized fields, or other validation issues.
- Consider service health and query issues
- Telemetry can be delayed or missing in the portal if:
- The ingestion pipeline is degraded due to service health issues.
- Log Analytics has problems saving records (uncommon).
- The query API (
api.applicationinsights.io) is failing when retrieving data (uncommon).
- Check Azure Service Health for any incidents affecting Application Insights or Log Analytics in the region.
- If using App Service integration
- For App Service with codeless monitoring, confirm:
-
ApplicationInsightsAgent_EXTENSION_VERSIONis correctly set (~2or~3depending on OS/runtime). - The Application Insights extension status page shows the extension running and
IKeyExists = True. - The status file path exists and indicates the app is running.
-
- If the app also references Application Insights SDK packages, ensure Interop with Application Insights SDK is configured as required; otherwise, telemetry may be disabled.
If telemetry tests from the host succeed but nothing appears in the portal, and there are no service health issues, open a support case specifying Application Insights and include correlation IDs from failed or delayed requests so the ingestion pipeline can be investigated.
References: