I am reading data from hbase using spark sql jdbc. one column has xml data. when xml size is small , I am able to read correct data in all columns. but if sizeof xml increases too much in a given row, some of the columns in dataframe becomes null for that row. xml is still coming correctly.