Data Flow Incremental Refresh Issue

[Dave82]

I have a Data Flow with several tables and we use Deployment Pipelines to push from Dev, Test, and Production.  This has been working fine for the past few months.  I noticed a few days ago one the incremental refreshes for one of the data flows stopped working.

I am getting errors like this:

I looked at the Test environment I see this: (working)

let RangeStart = #datetime(0001,01,01,00,00,00) meta [IsParameterQuery=true, List={}, Type="DateTime", IsParameterQueryRequired=true, Description="autogenerated_for_incremental_refresh"] in RangeStart

And Prod is now looks like this: (for some reason tables are turning into calculated tables and RangeStart/End are convering into tables)

let

  RangeStart = #datetime(0001,01,01,00,00,00) meta [IsParameterQuery=true, List={}, Type="DateTime", IsParameterQueryRequired=true, Description="autogenerated_for_incremental_refresh"],

  #"Convert to table" = Table.FromValue(RangeStart)

in

  #"Convert to table"

I tried copying the query text from test into the prod RangeStart/RangeEnd to have it not be a table type and I resave the data flow, but everytime I save it and reopen, I notice it autoconverts the RangeStart/End back into a table type as seen above. 

When I paste the test RangeStart/End query from Test into Prod I see warnings too:

If I redeploy this data flow from test to prod and do the first refresh, it again tries to convert the ranges from dateTime to table types again and fails.  Any suggestions?

[Anonymous]

Hi @Dave82 - have you considered the following:  Premium features of dataflows - Power BI | Microsoft Docs.  Is it possible that your Prod environment is hosted in Premium Capacity, so the newly deployed dataflow are being converted to linked entity.  It suggests the In the Enhanced compute engine settings, is set to "On" enable-the-enhanced-compute-engine .  This should be changed to "Disable".