Hello!
I have a V2 dataflow that has been running successfully for a few weeks. No changes were made to the dataflow, but this morning my dataflow is failing with 0 upsert and 100% failure. I suspect Microsoft may be rolling out changes on the back end.
Here is the error in the error log (the same error for each record):
{"error":{"code":"0x0","message":"An error occurred while validating input parameters: Microsoft.OData.ODataException: Cannot convert the literal '2021-03-16T00:00:00+00:00' to the expected type 'Edm.Date'. ---> System.FormatException: String '2021-03-16T00:00:00+00:00' was not recognized as a valid Edm.Date.
Here is the context:
- The field in Dataverse has a Date Only data type with a Date Only behavior (Date Only/Date Only is used to prevent date calculation issues relating to date comparisons from systems in different time zones)
- The column type in Power Query is String, with strings in this format: 2021-03-24 (using the Date column type produces an error because it returns the date/time, not just the date)
As you can see, I was using strings precisely to avoid returning date/time to Dataverse. Unfortunately, this workaround stopped working today. I tried a few things to solve the problem:
1. Using a Date column type instead of a string type : Same error
2. Using a Date/Time column type instead of a string type : Same error
3. Replacing the dashes with slashes in the string (2021/03/24) : Same error
4. Changing the project locale : Same error
Although all new dataflows are created with the V2 engine, I fortunately had mothballed a few old V1 dataflows. Using the advanced editor, I copy and pasted the code from my V2 dataflow (which worked just fine yesterday) into an old V1 dataflow. Surprise: With exactly the same code as the V2 dataflow, the V1 import succeeds with 100% upserts.
Is anybody else having issues with V2 dataflow imports of Date Only fields?
Keep 'em flowin'
Charles
