Long running tasks on prefect-dask clusters

Are you running it all on WSL? This could be the issue indeed.

You could increase the timeout for the Postgres queries as shown here:

This issue may provide some additional pointers: