User Security in Dimension - Cube Designer

[pgorge12]

Hi, I am trying to publish a cube and have the data in the cube seen based on the user who logs in. I've made this cube using a framework package and importing regular dimensions from it. My user security dimension, which shows the users and their respective business associations, has a filter. You can see this in the picture.           
                                                                       
All my filter have the property 'Exclude Facts Without Corresponding Dim Keys' set to false as is recommended.                               
                                                                       
When I apply this filter, the data shows my username as expected. However, when I go into the members of the Hierarchy, it doesn't return me (or whoever is the user). This looks this way on the cube in Analytics too after I published it.     
                                                                       
I removed the portion of the filter that asks for the user session parameter. When I check the members of the Hierarchy, it shows all the possible users. Again, when I publish the cube after this, I see all the users in the cube and I can see that the cube understands the difference between users and the data they are seeing. (example: some see more revenue than others).                   
                                                                       
I am also wondering if cube designer only understands that the cube will be presented based on the signon who published the cube. When I publish the cube, whatever I see, everyone else sees. Obviously we want users to only see what they should see. Does this have to do with the pre-publishing process when I click 'publish'?
                                                                       
I know this is a lot but I want to know how tor resolve this or how to make Cube Designer understand the governors and filtering so that it knows which signon is which. We are trying to make security official when publishing cubes. It's testing purposes but we need to know how make cube designer understand that depth of signon logic asap.         
                                                                       
Thanks!

[bus_pass_man]

I'm not quite sure I understand what you've done.  Just to clarify, you've set up security views in your dimensions but they're not working.  Is that a correct understanding?  I've got the sinking feeling that you haven't done that.

What's this dimension with the filter meant to do? 

Macros in dimensions are resolved at cube startup time and with the values of whomever published the cube.  The only exception are the macros in the lookup table security filters.   That seems to match what you've reported as your observation.




Here's some stuff which might help you.
http://www.redbooks.ibm.com/redbooks/pdfs/sg248064.pdf

https://www.ibm.com/support/knowledgecenter/en/SSEP7J_11.0.0/com.ibm.swg.ba.cognos.ug_cog_rlp.doc/c_cog_rlp_modelingusingrolap.html


hope that helps.