Change Decimal Precision of all 'Double type' Columns in a Spark Dataframe

I have a Spark DataFrame, let's say 'df'.

I do the following simple aggregation on this DataFrame:

df.groupBy().sum()

Upon doing so, I get the following exception:

java.lang.IllegalArgumentException: requirement failed: Decimal precision 39 exceeds max precision 38

Is there any way I can fix this?

My guess is, if I can decrease the decimal precision of all the columns of double type in df, it would solve the problem.