At 20:43 -0800 1/3/06, Dan Fandrich wrote:
>What you gave above isn't a URL. It may very well be that your conversion
>of the above string to a URL may be the issue.

What the OP quoted looks like an HTTP GET string starting with a solidus and terminating with (space) HTTP/1.1.

It's probably necessary to discover the host domain and the http or https which preceded it in the log of a test with a browser. Port 433 implies a URL starting with https. That needs to be prepended to the "URL" as in the OP message. The trailing space and the HTTP stuff needs to be removed.

As Dan said, the destination domain may not be the same as the previous page.

And. . . You may need to deliver at least a sessionID cookie.

My favorite for getting the details is the log file generated by iCab on a Mac but Livehttpheaders for use with FireFox is another option. Both are better than a sniffer which never sees the domain name in text form.

It might also be worth while to look at the GET string like this:

/ESD20/JSCDL/jdk/1.5.0_06/jdk-1_5_0-doc.zip
?AuthParam=1136326505_767ee1c5f15cfadefb8c8fc789ec792b
&TUrl=EMr8DPgljlCngjJlNnNDcFaPvU3tT+IO859ZF2EwLi4b1e0IkW/TF/LVaA==
&TicketId=dlV8PANPM+8z/Q==
&GroupName=SDLC
&BHost=sdlcweb8a.sun.com
&FilePath=/ESD20/JSCDL/jdk/1.5.0_06/jdk-1_5_0-doc.zip
&File=jdk-1_5_0-doc.zip
HTTP/1.1

You might well find that most of the needed items are right in the original HTML document as hidden input fields or in its cookies.

-- 
--> From the U S of A, the only socialist country that refuses to admit it. <--