@etorres182 I have examples of using CSV, ORC and parquet files as basis for external tables from HDFS and using them into’ real’ big data files (managed if you like). This is basically what the DB loader does under the hood.
Also if you are interested in KNIME and Hive you might want to check out this workflow:
Concerning the error could you provide us with a log file in debug mode when this happens. With big data an knime and ? Cloudera in an enterprise environment often it might be about permissions.
Also right before the upload a Cache node sometimes smoothes proceedings.