When a star schema is used as an input type when building a cube, the NWAY is not always compacted. When the NWAY is not compacted, the wrong measures are returned if the NWAY is used to satisfy a query.

When the issue does occur, there is no indication that an error has occurred. The only way to identify the issue is by manually validating query results. Data discrepancies should be easily identifiable.

If the issue does occur, a workaround is available. When you build the cube for the first time, withhold at least one record. Then perform an incremental update that loads the withheld record or records. When you perform these tasks, the measures are summarized correctly, and the cube displays the correct values when queried.

Click the Hot Fix tab in this note to access the hot fix for this issue.

Operating System and Release Information

A fix for this issue for Base SAS 9.3_M2 is available at:

A fix for this issue for Base SAS 9.21_M3 is available at: