Micah Zirn

I'm also observing the same CPU utilization and thrown exception symptoms. The thrown exceptions symptom I'm seeing in my setup is identical-- it's similarly being attributed to `_worker` at `thread.py:L#89`...

Hi @sanchda . Ah perhaps I had misread the recommendation in the ticket here https://github.com/DataDog/dd-trace-py/issues/9205 to address segmentation faults. I've re-enabled with `DD_PROFILING_STACK_V2_ENABLED=true`. The thrown exceptions are gone as you...

I don't believe I'm seeing the total thread count symptom. Thanks for the assistance here @sanchda!

The application is a FastAPI / uvicorn web server within a single process. It uses `asyncio` as the event loop. Each endpoint is an async function, and every request should...

That said, the service page on datadog for this service in our testing environment indicates a sum total of ~400 threads for the 3 running pods

@sanchda I wanted to add this clarification just in case I wasn't clear. In the details and screenshots above, we were already running with profiling enabled. At the 53 minute...