Think this is a dagster error? Not being provided an option to specify config yaml via the UI, and receive this error for this configured asset below.

# the io manager stuff properly sorted.
# name of asset will be name of function definition
@asset(
    required_resource_keys={"ml_mongo_warehouse"},
)  # , non_argument_deps={'ml_cxr_datalake'})
def cxr_collection_metadata(context) -> pd.core.frame.DataFrame:
    ....

ml_cxr_collection_configured = with_resources(
    [cxr_collection_metadata],
    resource_defs={"ml_mongo_warehouse": mongo_handler_resource},
)

Bit perplexed, using very similar patterns successfully elsewhere. Problem resolves itself if I just bypass the dagster resource tooling, but that'd be a shame.

A quick fix could be to provide the config using the

resource_config_by_key

argument of

with_resources

, although if you need to change per run, that won't help you unfortunately.