Need Help With LOOKUP Transform

I have a very simple problem I am trying to solve.

I have a table with a "DateEntered" field, and I have an ssis pkg set up to load data from a file into the database table. I just want to make sure that no one loads the same file twice in one day.

For example, if today is 8/22/07, and "DateEntered" is "2007-08-22", then I want to add a Lookup transform to run a query that will check and see if there's any rows in the table with a "DateEntered" is "2007-08-22". If so, don't load the file again!

Here's my query:

SELECT Code
FROM myTable
WHERE DATEADD(dd, DATEDIFF(dd, 0, DateEntered), 0) = DATEADD(dd, DATEDIFF(dd, 0, GETDATE()), 0)

(all the dateadd stuff is doing is removing the time portion from the DateEntered field, so we are comparing apples to apples).

Now, if the query returns a bunch of "Codes" then we know that the data has already been entered for the day! So far, so good.

Now, how do I set up the Lookup to get it to work? I'm getting this error message:
Error 1 Validation error. Data Flow Task: Lookup [1299]: The lookup transform must contain at least one input column joined to a reference column, and none were specified. You must specify at least one join column. FXRateLoader.dtsx 0 0

But I thought I did this! On the columns tab, I have:
Lookup column: code
Lookup operation: Replace 'code'
Output alias: code

I have my error output set to:
Lookup output - redirect row

I don't know what I'm doing, obviously..........!

Need assistance :-)