Similar to this question. I got the same error. But instead of creating csv file, I already have a csv file and trying to load it to spar-sql table. If this csv is not compatible with hive, is there a tool to convert it to comparable within spark-sql?