@Marvin How can I read the result of the subflow run called using

run_deployment

. The subflow had persist_result on

Ah, the joys of dealing with subflows and persisting results. I can see you're having a grand time. The

run_deployment

function is used to run a pre-existing deployment as a subflow of the calling flow. This is similar to calling the flow object itself in a parent flow with something like

asyncio.gather

. The key difference is that when you use

run_deployment

, the subflows will run according to the infrastructure associated with that deployment. When you call the

subflow()

object itself, it will necessarily run on the calling (parent) flow’s infrastructure as a process[^1^]. Now, if you're dealing with an error like

Flow run encountered an exception. RuntimeError: The call get_task_call_return_value(task=<prefect.tasks.Task object at (some_address)>, flow_run_context=FlowRunContext(start_time=DateT...) is already done.

, it might be related to the version of Prefect you're using[^2^]. As for reading the result of the subflow run, if the subflow had

persist_result

on, the result should be stored and retrievable. However, the specifics might depend on your setup and the way you're calling and handling the subflow. I hope that helps. Or at least, I hope it doesn't make things worse. I'm here if you need more information, or if you just want to share your existential dread with someone who gets it. [^1^]: How can i run multiple subflows or child flows in parallel [^2^]: How can i run multiple subflows or child flows in parallel