i have a question regarding mondrian
My facts table
customer | regionid | customertype | productsold
------------------------------------------------
A | 1 | T1 | P1
A | 1 | T1 | P2
A | 1 | T1 | P3
B | 2 | T1 | P1
B | 2 | T1 | P2
C | 1 | T2 | P1
C | 1 | T2 | P2
C | 1 | T2 | P3
C | 1 | T2 | P4
... so on
Region dimension
regionid | state | district | subdistrict
-----------------------------------------
CustomerType dimension
customertype | name | more column
----------------------------------
ProductSold dimension
productsold | name | more column
--------------------------------
Each customer have 1 to 1 relation with region and customertype, and 1 to many with productsold
my MDX query like
select
{[Measures].[CustomerCount]} on columns,
{([Region].[AllRegion], [CustomerTypes].[AllCustomerTypes],
[ProductSold].[AllProductSold])} on rows
from [Customers]
I configured distinct-count for CustomerCount, but still the result is duplicate.
<Measure name="CustomerCount" column="Customer" aggregator="distinct-count" visible="true">
Any advice appreciate. Thanks.
I would have expected your MDX statement to return just a single cell (containing the distinct count of 3) since your rows axis includes the All member from each dimension.
I would suggest you enable SQL logging in Mondrian and check the Segment.load query since it should be a pretty simple SQL statement to return the distinct count of Customers.
Mondrian 9.3 and onwards uses log4j2 so see this example for configuration: https://github.com/pentaho/mondrian/blob/master/mondrian/src/it/resources/log4j2.xml
Prior versions of Mondrian still use log4j1.2 so see this example for configuration: https://github.com/pentaho/mondrian/blob/9.1.0.0/mondrian/src/it/resources/log4j.xml