Hey there! I kindly ask for your swarm knowledge on ClearML pipelines.
I'm trying to setup a simple pipeline with a Controller running on the service-queue and three tasks, that are added with

pipe.add_function_step(
        function=training,
        function_kwargs=dict(args=args),
        function_return=["model_id"],
        task_type=TaskTypes.training,
        name=f"train_{task_name}",
        tags=pipe.tags + ["train"],
        repo="./",
        docker=config.base_docker_image,
        packages="./requirements.txt"
    )

The controller task initialized correctly, but once starting the execution it fails with

Environment setup completed successfully
Starting Task Execution:
2023-07-06 08:58:06
ClearML results page: 
ClearML pipeline page: 
Traceback (most recent call last):
  File "train.py", line 152, in <module>
    training_pipeline(args=args)
  File "train.py", line 114, in training_pipeline
    pipe.start()
  File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 895, in start
    self._start(
  File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1394, in _start
    self._prepare_pipeline(step_task_completed_callback, step_task_created_callback)
  File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1428, in _prepare_pipeline
    if not self._verify():
  File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1637, in _verify
    self._verify_node(node)
  File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1653, in _verify_node
    raise ValueError("Node '{}', base_task_id is empty".format(node.name))
ValueError: Node 'train_detection_2023-07-06_may-close-body-1688626563', base_task_id is empty

Can anyone explain this error? What is a base_task_id , why do I need it and where to set it?