@Jonathan Smith helped me out most recently in this thread...

Rocket U2 | UniVerse & UniData

Rocketsoftware		remove preview
Rocket U2 | UniVerse & UniData I've overcome most of the issues discussed earlier in the following two threads. The background there may help in addressing this question.Rocket U2 | UniVerse View this on Rocketsoftware >

Is there a similar  parameter I can set to get around this in the UCI Configuration Editor? 

I've tried "casting" the AVG_COST_PD column differently so that it would return, but to no avail.
As usual, I can query this table fine in the U2ODBC Tester, but not in SQL Server Management Studio, PowerBI and other Microsoft Products. 

Although Microsoft Query legacy connection can also return the full table to Excel, and I don't see any outlandish values in the AVG_COST_PD column there.

The full error now is:
The OLE DB provider "MSDASQL" for linked server "mylinkedservername" supplied invalid metadata for column "AVG_COST_PD". The precision exceeded the allowable maximum.

Karl,

I suspect AVG_COST_PD is a virtual dictionary , the easiest method might be to edit the AVG_POST_PD dictionary itself and force the calculated value to be returned with a set number of decimal points and if further needed also change the dictionary to force the sql type conversion.

Thanks,

Thanks again Jonathan.  I don't have direct control or access to the U2 environment, but will pass this along to those who do.

What about this HS.SCRUB utility I'm reading about?  If the admins of the U2 environment ran that, would it help clean some of this up?

Karl,

I really feel that you need to sit down with the developers and admins of the UniVerse environment to discuss what you are trying to achieve and how they can assist you. The major problem here is that UniVerse is a typeless database so there are no rules as to what can be stored in each field in the file. When you accessing this data in the way you are, then type is being implied via the ODBC driver. Hence just trying to resolve the problem on your end may not be enough to resolve the further issues you are running in to, they may have to be resolved at the U2 level.

HS.SCRUB is a utility used in general to perpare the dictionary and data of a UniVerse files for ODBC access. Only after a discussion with the developers and admins of the UniVerse environment will they then be able to decide if this will assist you in what you are looking to do. My initial thought would be that it will not  but I am missing a lot of information about the data being held in the UniVerse files and the dictionaries.

You will not solve this just by looking at this from the Client side, you really need to sit down and have a discussion with the UniVerse developers and admins as I suspect at this point changes at that level need to happen.

Regards,