Bug in SSAS Rolling up data?

I'm developing an Invoice cube and while I was debugging it I noticed some of the unit costs were too high, so I use the drillthrough to get the data behind it and the Row Count Measure (the one AS can add to your cube for you) had a 2 for one of the records.

All of the other records had a row count of 1, and it's my understanding that drillthrough shows the details of the data at the grain of the fact table.

I looked at the fact table and I had 2 distinct records that do indeed add up the the values represented in the erroneous record, so... Has anyone else had this? Is there a work around? Did I do something wrong?

Thanks in Advance,

Bryce

The drillthrough returns the most detail records in your cube schema aggregated at the grain of each of your dimensions, that's why you see a record count of 2 for some of the rows. What you will have to do is create a degenerate dimension off of your fact table which includes your primay key along with any other columns you want to see. You can then hide this dimension but include the columns in your drillthrough to obtain the most detailed records.

Hope this helps.

Van Dieu

|||

Do you think there will be a problem with the performance? To my understanding, the cube is going to have detailed granual level of data.