What I have tried so far:

@flow(
    name="Create Experiment",
    task_runner=DaskTaskRunner(
        cluster_class="dask_jobqueue.SLURMCluster",
        cluster_kwargs={
            ...
            "job_script_prologue": [
                String.load("log-slurm-job").value,
            ],
        },
    ),
)
def my_flow():

But this will not build a deployment because the load was not

awaited

. What would be the correct way to use blocks for flow-configurations?

I wonder if it has something to do with using DaskTaskRunner

I can wrap

String.load("test").value

into a `async`function and await the result inside it. Then the deployment works, but when I run the deployment it hangs forever (waited for 10min and then killed it).