Categories
How to create a logical join with two facts and a confirmed dimension?

Hi All,
I am trying to create a logical join with two facts and a confirmed dimension.
However I see this error message when I create an analysis by pulling one column form each table (Fact1, Fact2, Dim1).
State: HY000. Code: 14025. [nQSError: 14025] No fact table exists at the requested level of detail:
I am expecting the OBIEE server to generate the query some thing like this, please suggest how to achieve this.
Select
Fact1.column1, Fact2.column2, Dim1.column3
from
Fact1, Fact2, Dim1
where
((Dim1.code = Fact2.code) and (Dim1.code=Fact1.code));
Thanks!
Answers
-
Can you also show us the physical joins?
Can you also show what has been modelled on the dimension hierarchies in question and the settings on the table source for the fact / dimension relationships in question?
0 -
When you use conformed dimensions and multiple fact tables, your business models require hierarchies and you must set content level based on what fact exists at what level of detail.
0 -
If your relationship is as simple as you describe I would expect (for it to work); -
Fact 1 - relationship to dimension 1 set to detail
Dimension 1 - has a dimension hierarchy with a minimum of total level and detail level, relationship to fact1 and fact2 set at detail level
Fact 2- relationship to dimension 1 set to detail
Physical joins model Fact1 -> Dim1 <- Fact2
Logical Relationships modelled Fact1 -> Dim1 <- Fact2
0 -
Yes, I did.
0 -
This is what I did.
Physical Joins
Logical joins
Dimension Hierarchy
Realtionship to FACT1
Realtionship to FACT2
When I ran the analysis using the code which is the key from the three tables, I see this error.
Please suggest.
Thanks!
0