Commented SKIPCHECK causing cube load errors and stack overflow?

Post Reply
vladino
Posts: 110
Joined: Sat Nov 06, 2010 10:10 am
OLAP Product: Cognos TM1
Version: 10.2.2
Excel Version: Excel 2013

Commented SKIPCHECK causing cube load errors and stack overflow?

Post by vladino »

Hi guys,
we have commented out SKIPCHECK on one of our cubes. I know this is not the best approach but as we are unable to use feeders and we need to calculate ConsolidatedAvg values on hierarchy nodes we can't use feeders at all.

Is it possible that this is causing "Unable to load cube" in TM1 apps? The security is configured correctly and users are able to work with the cube. But sometimes they are getting "Unable to load cube" error and the view disappears from the application. When they log out, close the browser, log in again and open the application the view is working correctly.

Additionally, we are getting #N/A, circular references and stack overflow errors in the log file. But when I change the view (ie. pivot dimensions) the values are calculated. Measures affected are used for calculating BOP/EOP values.

I'm a bit confused by this behavior so any clarification is much appreciated!

BR
Vladino
tomok
MVP
Posts: 2831
Joined: Tue Feb 16, 2010 2:39 pm
OLAP Product: TM1, Palo
Version: Beginning of time thru 10.2
Excel Version: 2003-2007-2010-2013
Location: Atlanta, GA
Contact:

Re: Commented SKIPCHECK causing cube load errors and stack overflow?

Post by tomok »

I think what TM1 is trying to tell you is that you have some problems with your rules. Stack overflow is from too many dependencies in the rules. What I mean is Z is dependent on Y, which is dependent on X, which is dependent on W,.....and so forth. It could be that taking off SKIPCHECK exposed this. It looks like that's exactly what you have with the EOP to BOP linkage. When you are doing BOP to EOP linkage it can't be at the exact same dimension elements for all the other dimension in the cube. That would result in a circular reference. You have to break that linkage on at least one other dimension. Also, make sure you don't have any intersections where A is part of the calculation for B which is part of the calculation for C which is part of the calculation for A. Both of these errors are really easy to do when you have time period linkage like you do. NA means you have another math error, usually division by zero using the "/", instead of "\".
Tom O'Kelley - Manager Finance Systems
American Tower
http://www.onlinecourtreservations.com/
Post Reply