I get the error "An error occurred while communicating with the data source all inputs to union must have matching column types" when running my flow. I was using it without problems all 2024, but now I modified it because I had a date filter (show only 2024) that I needed to update. I even rolled back my filter but keep getting the error. Talked to the DB administrator and he says they haven´t made any changes to sources (variable names, columns, DB types). Is it a known error? Please help.
PS: I get the error in desktop and in web.
Hi @Kate Sales
You are correct,
This error usually comes up when one or more columns going into the Union have different data types, even though there hasn’t been any change to the database schema.
I’d first check the inputs right before the Union and compare the column names and data types. I’d mainly look at date and numeric fields, as well as any fields that might contain NULLs.
If you find a mismatch, try explicitly converting the fields to the same type before the Union, for example:
DATE → DATE
INTEGER → INTEGER
STRING → STRING
Since the flow was working fine before, I’d also try refreshing the input metadata/schema and then checking the Union again. Sometimes the metadata gets stale, or an upstream field starts being interpreted differently, which can cause this error.
Thankyou!
