I encountered the following error during a cube activation:
[ { "Data": [], "Type": "Error", "Message": "A unique index is automatically generated for the summary data table upon activation. Therefore, defining another unique index manually is not allowed." } ]
After investigating, I found that the associated data table for this cube summary table has a manually defined unique index. Are manually defined indexes (especially unique ones) on summary data tables supported?
In this case, the issue was caused by a manually defined unique index on the data table. To resolve this, I had to update my manually defined index to Unique = False.
A few important notes I wanted to post that others might find helpful:
Summary tables can have user-defined indexes, but these must not be unique.
The platform automatically defines a unique index based on the dimension fields in the cube summary definition. This is required for processing summary table events.
Multiple unique indexes on the table will conflict, which is what’s happening in this case.
If needed, you can reorder the dimension fields in the summary definition to influence how the platform generates its default unique index.
You can also add additional indexes, but again, they must be non-unique.