I am using the below query in compute expression in dataflow:

case when Date_Taken is null then toDate(Date_Received, "yyyy-MM-dd") else toDate(Date_Taken, "yyyy-MM-dd") end



I am encountering the below error :

Something went wrong while executing the compute:SampleHeader node: exceeded max error limit: The number of failed rows exceeded the 10000 limit. Only the dataflow owner or users with the View All Data permission can download the log. (02K0r0000009ZcIEAU_03C0r000000HXN7EAO)

2 Answers 2


In your expression can you check for blank Date_Taken in addition to null or check in the dataset for the existence of values as you are expecting in Date_Taken. I am suspecting a mismatch there.


I used coalesce function as below and it worked for me :

  • Not clear about the "it worked for me" part of the solution. What coalesce did was only to replace null values (when encountered Date_Taken_sec_epoch) with value as toDate(Date_Received_sec_epoch). Is this ok per the logic you are trying to build using the computeExpression?
    – Mitesh
    Jan 16, 2020 at 8:09
  • Yes I wanted the first non -null value from Date_Taken and Date_Received . In the first case also, I was doing the same thing , I was storing Date_Recieved if Date_Taken was null else I was storing Date_Taken itself
    – Gurmeet
    Jan 16, 2020 at 8:13

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .