On a SQL 2016 Server I have a job that calls an SSIS package. That package is in a project in the SSISDB and has parameters. One of those parameters is a string type that is blank as a default.
I ran the job with the blank value for this parameter, and it ran successfully.
I then opened the job properties, went to the step that calls that package and went into the configuration and gave that parameter a value.
I ran the job again and it ran successfully, and the parameter value had the expected effect on the results.
Now I want to modify the job and set the parameter back to a blank string. Repeating what I did above, I open the configuration and completely delete the value of the parameter.
When I then try to save the Job Step properties, I get an error dialog with the message in my subject line.
If I put a non-empty value back in for that parameter, the job step can be saved.
When I first created the job, I was able to save it with an empty string for that value. It's just when I change the value to a non-empty string, run the job, and then try to change it back to empty, that I encounter this error.
I'm aware that I can run an alter job script or delete and re-create the job step to work around this. I'm not interested in knowing how to work around this problem.
My question is: Is this a known bug, or is there a trick to using the GUI to change the value of a parameter for execution from a string value to an empty string?
Can any of you do it, or is my issue reproducible? I've found nothing in my searches so far.
EDIT: following Aaron's comment, I tried entering a pair of single-quotes for the value of the parameter. The job accepted it and saved successfully. But then when I ran the job, the pair of single quotes were used literally as the value of the parameter, instead of an empty string.