Expected behavior DBT should always run set spark.sql.sources.partitionOverwriteMode = STATIC as a pre-hook for liquid clustered model to ensure it will always avoid this kind of failure. This can be ...