0

I'm trying to diagnose a bug in a calculated measure in a SSAS cube, and while debugging I get this weird behaviour ...

select
    [Measures].[Net Lost Commitments] on columns 
  , {[Date Dimension].[Fiscal Year Hierarchy].[Fiscal Year].[2009\10],[Date Dimension].[Fiscal Year Hierarchy].[Fiscal Year].[2010\11]} on rows
from Compass3

...returns the following:

       Net Lost Commitments
2009\10  (null)
2010\11  9,937

I'm trying to figure out why the value for 2009\10 is null and so I've been adding various other calculations via a WITH clause to help track back the problem. But any WITH clause I add changes the results, even if the calculation in the WITH clause is not used. For instance ...

with
    member [Measures].[test] as null
select
    [Measures].[Net Lost Commitments] on columns 
  , {[Date Dimension].[Fiscal Year Hierarchy].[Fiscal Year].[2009\10],[Date Dimension].[Fiscal Year Hierarchy].[Fiscal Year].[2010\11]} on rows
from Compass3

...returns the following:

       Net Lost Commitments
2009\10  8,783
2010\11  9,937

Have you ever seen that before? What could possibly be the cause?

4

2 回答 2

0

我已经将度量定义重新设计成更简单的形式，这个问题似乎已经消失了。我仍然不知道这种看似随机的行为是如何发生的，但它不再是一个问题。

——马特。

于 2013-03-13T03:48:00.650 回答
0

也许计算成员的存在会影响 [Date Dimension] 聚合的求解顺序？[Net Lost Commitments] 是经过计算的衡量标准吗？如果是这样，您可以将其求解顺序更改为稍后在查询处理中吗？

于 2013-02-25T04:32:39.483 回答