Never had this problem before. When we generate categories, and bring up the diagram, all expected categories appear. But when we build the cube, for one of the dimensions, only one of the six categories appears.
Why would this happen?
The most likely cause is a lack of measure data for the particular categories that are missing. This can sometimes happen when building a model with multiple data sources.
If you have your transformer model setup to build your dimensions first, then your facts, the categories will be generated in the transformer *.mdl/*.pyi model for all dimensional values.
There is a property for each level that lets you change the inclusion of categories in the cube when the *.mdc file is built. If this property is set to 'Default (When Needed)' or 'Include When Needed', only those categories that have transactions from your fact table against them will be built into the cube.
If you change this setting to 'Always Include', your cube file will contain all category values, even when there is no data.
Hope this helps.
J