Question

Starting from ActivePivot Sandbox 4.3.2 I changed my object feeding into the cube and redefined the fields, dimensions and measures of the cube. When I start the cube I see no error message in the logs.

However when I connect to the cube using ActivePivot Live 2.6.2 or Excel 2010 and run the following MDX query:

SELECT FROM [cubeName] WHERE ([Measures].[contributors.COUNT])

I see an empty pivot table, what can be the cause? How can it be diagnosed?

Answer 1

The most common reason for empty pivot tables on a non empty cube is the presence of slicing dimensions. If you have:

1. 2 slicing dimensions ‘A’ and ‘B’
2. a first fact contributing ‘b’ on ‘A’ and ‘a’ on ‘B’
3. a second fact contributing ‘a’ on ‘A’ and ‘b’ on ‘B’

then default members will be ‘a’ along ‘A’ and ‘a’ along ‘B’. The query you described would then return an empty pivot table as there is no fact with ‘a’ along ‘A’ and ‘a’ along ‘B’.

The second main reason is security filtering. You should retry with a user wihtout any access restrictions. This is easily feasible by requesting the query through the dedicated operation on the ActivePivotManager monitoring bean.

Of course, you should first check your cube is non empty (through the JConsole).