Timer triggered workflow run traces are saved to app insights, but when sending HTTP requests traceparent header has traceflag 00

[scaepz]

Hi,
logic apps automatically send traceparent headers in HTTP requests, and the last part of the traceparent is the traceflag, where -00 means "not recorded", and -01 means "recorded". Our API looks at this flag to decide whether or not to record the traces to App Insights. Always either dropping all telemetry or keeping all telemetry related to an operation is helpful since it keeps the span hierarchy nice and complete.

Our timer triggered logic apps workflow (standard plan) runs are all automatically recorded and I can find them in App Insights, but the workflows send a HTTP request to our API and automatically pass along the traceparent header which always has traceflag -00, leading to our system always deciding to drop the telemetry.

Is this a bug in logic apps? Is there anything I can do to influence the flag? I could send a custom header to the API informing it that the request came from a logic app and it can make sampling decision based on that, but that's not as neat as having the correct flag in the first place.