I have an application that needs to work with SQL Servers where I am not in control of the security model installed on the back end.

Most installations will run with Windows Only, but others may have Mixed Mode.

Is it possible to write code which will detect a failure of one and then try the other ?

I tried doing a TRY...ENDTRY around the code but it didn't work very well.


On a slightly related topic, I saw a thread on here which talked about setting up logins in SQL which consist of IP addresses. I wonder if this will be necessary here, because otherwise if I need to access SQL using a login, I'll need some UID and PWD to put in my connection string...

Any suggestions please ?

Thanks
Simon