Diagnostics in Durable Task SDKs

Your options for diagnosing issues with the Durable Task SDKs depend on the Azure compute you're using, like Azure Container Apps, an Azure Kubernetes Service cluster, or an Azure App Service app. We recommend enabling both Application Insights and the Durable Task Scheduler monitoring dashboard to get full visibility into orchestration status and failures.

Application Insights

Application Insights is the recommended way to monitor your apps running on the Durable Task SDKs. You can find and query these tracking events using the Application Insights Analytics tool in the Azure portal.

Each lifecycle event of an orchestration instance writes a tracking event to the traces collection in Application Insights. Use the custom dimensions on these events (like prop__instanceId, prop__name, and prop__runtimeStatus) to filter and query specific orchestrations.

Example queries

Once Application Insights is enabled, use these KQL queries in Logs to investigate orchestration behavior:

Find failed orchestrations:

traces
| where customDimensions.prop__runtimeStatus == "Failed"
| project timestamp, customDimensions.prop__instanceId, customDimensions.prop__name, message
| order by timestamp desc
| take 50

Orchestration duration by name (last 24 hours):

traces
| where timestamp > ago(24h)
| where customDimensions.prop__runtimeStatus == "Completed"
| extend duration = todatetime(customDimensions.prop__completedTime) - todatetime(customDimensions.prop__createdTime)
| summarize avg(duration), max(duration), count() by tostring(customDimensions.prop__name)

Distributed tracing with OpenTelemetry

The Durable Task SDKs support OpenTelemetry distributed tracing, which gives you end-to-end visibility across orchestrations, activities, and sub-orchestrations as correlated spans. You can export traces to any OpenTelemetry-compatible backend, including Application Insights, Jaeger, or Zipkin.

For setup instructions with code samples in all supported languages, see OpenTelemetry and distributed tracing with Durable Task Scheduler.

Durable Task Scheduler monitoring dashboard

The Durable Task Scheduler dashboard lets you monitor orchestration status, inspect inputs/outputs, view execution timelines, and manage orchestrations (start, pause, resume, terminate). It's available with both the local emulator and Azure-hosted schedulers.

Access the dashboard

• Local emulator: No authentication required. The dashboard is available at http://localhost:8082.
• Azure-hosted scheduler: Assign the Durable Task Data Contributor role to your identity, then access the dashboard via the task hub's endpoint URL in the Azure portal or at https://dashboard.durabletask.io/.

For full setup and usage instructions, see Debug and manage orchestrations using the Durable Task Scheduler dashboard.

Common issues

Next steps

• OpenTelemetry and distributed tracing with Durable Task Scheduler
• Debug and manage orchestrations using the Durable Task Scheduler dashboard
• Durable Task Scheduler overview