I was wondering what the difference would be to use this method, as opposed to a T-SQL method, ie not VB/dll one.

Like this one:
http://support.microsoft.com/default.aspx?scid=kb%3Ben-us%3B312839#3

As well, a more general and novice question, why would I use this method over something like plain old CDO where I don't have to put in my mail server name (Particularly when I do have a Virtual mail server or whatever else is necessary to send it)?

Thanks

Hi there

Actually, its the first ive seen the MSDN link, and there is no problem I see it using this method also. Its like all things tech, 101 different solutions to most problems; this is just one of them that works well from other code written, so im going down the service orientated path, a single buisness obj (aka the DLL) and many services that use it (t-sql, vb, vb-script, asp, etc..) from the business apps.

Cheers

Ck

Sounds good to me,

Thanks

I have used an extended stored procedure (freeware) for almost 2 years with success.  You may be interested in taking a look.  http://www.sqldev.net/xp/xpsmtp.htm

Great article, Chris!

I didn't know about SQL Server Agent tokens until now. I plan to use them for an ErrorLog in SQL Server, but I saw a problem that affects your code too: if the error message contains single quotes ('), the string built for the @message parameter ends there, and the remaining words of the error message will be interpreted as another command, usually leading to an error. The solution that I plan to use is to SET QUOTED_IDENTIFIER OFF and use double quotes (") instead of single quotes. There still is a chance of failure, but it's smaller, because there are much fewer error messages that contain double quotes compared to the error messages that contain single quotes.

Razvan

Thanks Razvan

I will check out single quotes, could be a nasty bug...