Dataflow Refresh Problem - We cannot convert the value #datetime() to type Date- URGENT!

[Anonymous]

Since Monday our dataflows stopped updating

The error that is given is the following: “Error: Expression.Error: We cannot convert the value #datetime(2019 9 26 0 0 0) to type Date. .. RootActivityId = da928912-31f6-40a6-994c-761c55104956.Param1 = Expression.Error: We cannot convert the value #datetime(2019 9 26 0 0 0) to type Date. Request ID: a033c9df-0c02-7ccf-df39-c48a1c29335c.

 And it happens in all 3 workspaces (dev-test=prod and with different tables and different dates!)

We also notice that the datetime function in the error message is missing commas. We haven't made any changes in the code and are not explicitly doing any datetime to date conversion... 

This is a critical problem for us because the whole organization is now dealing with 2 days old data. Please advice!

[Anonymous]

For us the workaround was casting all our dates as datetimes. Things worked afterwards.

[v-chuncz-msft]

@Anonymous 

A similar issue has been reported internally, so stay tuned.

CRI 245776965

[v-chuncz-msft]

@Anonymous 

The fix is deployed. You may check it again.

[ChemEnger]

So, I wonder if this might be useful - we have just switched to a Premium workspace and a Dataflow that has been happily chugging along for months just filpped out on us.

No explicit conversion to date type, in fact deliberatly kept as datetime to allow incremental refresh.  However, there is a Table.Group line in the query and it looks like the conversion has added a type nullable datetime to the group command (we need the latest date as part of the grouping).  Changing this got the query up & running straightaway 😊