My first thought is that you could use
wget to get the signon page from the web server and save it to a file. Then you could grep the file to make sure the file is a real signon page. The syntax in your script would look something like this:
#get the Signon page from the server
wget demo1.erpassociates.com:9080/HR91DMO/signon.html
#Make sure it's real:
grep response=grep "login" signon.html
#Do something if $response is empty
if ${response} ...
I believe there's a wget version for Windows as well.
I'm sure there are other ways to do this, that's just the first one that comes to mind.
-Brent