You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We’ve recently encountered an issue in our dev environment while running Azure Durable Functions. After some investigation, it seems to be directly related to the Datadog Application Services (AAS) extension that we are using to monitor our functions.
With the extension enabled, our Durable Function orchestrators exhibit errors related to non-deterministic behavior, which causes the orchestrator to fail.
Here is the error message we are seeing:
The orchestrator function completed on a non-orchestrator thread!
"@i":"3192603f",
"@l":"Error",
"@x":"System.InvalidOperationException: An invalid asynchronous invocation was detected. This can be caused by awaiting non-durable tasks in an orchestrator function's implementation or by middleware that invokes asynchronous code.
When we removed the Datadog extension, the Durable Functions performed as expected without any errors or failures. This leads us to believe that the extension might be introducing some asynchronous behavior or thread context switching that conflicts with the deterministic execution requirements of Durable Functions.
Are there any known workarounds or best practices to ensure the Datadog extension can work alongside Durable Functions without impacting their deterministic behavior?
Any configuration or adjustments we could make to minimize the interference?
Thanks in advance.
The text was updated successfully, but these errors were encountered:
Are there any known workarounds or best practices to ensure the Datadog extension can work alongside Durable Functions without impacting their deterministic behavior?
Unfortunately, at the moment we don't support Durable Functions so I don't think there are any workarounds for the async/thread issues here.
Submitting a feature request or a bug report for this will be the best route at the moment.
We’ve recently encountered an issue in our dev environment while running Azure Durable Functions. After some investigation, it seems to be directly related to the Datadog Application Services (AAS) extension that we are using to monitor our functions.
With the extension enabled, our Durable Function orchestrators exhibit errors related to non-deterministic behavior, which causes the orchestrator to fail.
Here is the error message we are seeing:
When we removed the Datadog extension, the Durable Functions performed as expected without any errors or failures. This leads us to believe that the extension might be introducing some asynchronous behavior or thread context switching that conflicts with the deterministic execution requirements of Durable Functions.
Are there any known workarounds or best practices to ensure the Datadog extension can work alongside Durable Functions without impacting their deterministic behavior?
Any configuration or adjustments we could make to minimize the interference?
Thanks in advance.
The text was updated successfully, but these errors were encountered: