Is this the expected behavior?

Hi, I was having some problems with my PHP files (the first 3 HTML lines missing and Netscape trying to save the page to disk instead of showing it). So, I telnet'd my web server and manually issued a GET as follows: GET /anglers/index.php HTTP/1.1 And this was the response: HTTP/1.0 200 OK Date: Tue, 04 Dec 2001 21:09:57 GMT Server: WebSitePro/2.4.9 Accept-ranges: bytes <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//ES"> <!-- Copyright ® 1999-2001 Jogua Web Solutions v3.0 --> <!-- BOF: default.inc 2001-12-04 18:09:57 --> <html> <head> <title>Anglers Aventuras \ Inicio</title> etc. etc... Note the CRLFs missing after the headers :-/ So the browser -naturally- understand those HTML remarks as being part of the header, and the HTML starting after those two CRLF I have in my HTML file. I know I can "fix" this just typing two CRLFs at the very beggining of my PHP file, but I'm not supposed to do that. And even if I do that, the header is still incomplete (i.e. mime type missing). Is this a know bug, an unknown bug or my code? I don't think I screwed something up because I'm not playing with the headers here. This happens with 2 different servers, both running WebSite Pro 2.4 and PHP 4.0.6 (one Windows 2000 Advanced Server and the other Windows XP Pro). The only thing strange with my code is that I use many nested inclusions (not recursion, just including a file in an included file wich at it's time is included in another, etc.). Regards, Ernesto