Dataflow refresh - cannot convert null to type logical

[homol]

I have a Dataflow that looks good in the Power Query designer and appears to be in good order.  

However when I start the refresh, I get the following error:

Error Code: Mashup Exception Error, Error Details: Couldn't refresh the entity because of an issue with the mashup document MashupException.Error: Expression.Error: We cannot convert the value null to type Logical.

Does this mean that one of the values that its trying to insert into Dataverse is NULL or that something that appears to be functioning in my designer is actually erroring and I'm missing it?  Is that what a mashupexception represents?

[EricRegnier]

Hi @homol,

Usually a mashup exception is when the data type is incompatible upon insert. It doesn't mean that if the PowerQuery designer work that it will work on import. For ex, you can have service-side validation, alt keys, etc, that might prevent the transaction. What destination column type is it, lookup?

[homol]

well, it's hard to say, since I can't tell from the error dump, what is the affected column.  Or at least I don't think I can tell lol.  But yes, it's most likely a lookup that is a part of the key - that makes the most sense.  The "Logical" type is what was throwing me off - I started looking into my designer for True/False fields with Nulls lol.

[homol]

@EricRegnier 

Here's my particular setup:

I have an Investment table that has 2 different keys:

• A combo key of 3 different lookups, to validate uniqueness (Internal Partner, External Partner, and Type)
• An autonumber column that uniquely represents the Investment Code

I have a Tax Year table that has a Lookup to Investment - this is the table I'm trying to insert into - various years per investment.

My assumption was that I could simply get the autonumber value and associate it to my years and conduct my inserts.  I see this though in the Map tables screen.

Does this mean that I still need to have the other 3 values to make the insert work?  I'm assuming that this is what's causing my error.

[homol · Accepted solution]

So it turns out that, in my case, the issue was actually within the designer mashup. And the Logical even made sense, once I figured.  I had a new conditional column in my final table, used for insert.  This conditional column was checking for a value to be greater than 0.  If it was it would set the value to one column versus another.  However in the column I was comparing, there happened to be one row out of 1400 that had a null value incorrectly.  This was generating an [Error] just within that cell for that row inside the designer but I didn't catch it.  Once I fixed the conditional, everything inserted fine.

[EricRegnier]

Thanks for lettings us know!

[cbovikings]

I had the same issue. It was a calculated column in my dataflow. I had the formula depending on a date and one of them was null. I put a condition in the formula to check for null's. I was able to find this easily by going to my last applied step and clicking the top left corner and choosing 'Keep Errors'. This filtered all columns and showed rows with an error.