Perl is supposed to be able to interpret various unicode encodings if proper Byte Order Marks appear at the beginning of the script file, but this may conflict with having your server rely on the #! line. Can you tell the server to use perl without having to check the shebang line?
Don't know if/how mod_perl will react to BOMs, either.
Update:Is it really utf8 or is it utf16?
Comment on Re: calling the interpreter from a utf8 script
Hi,
The Bom seems to be the problem. Notepad seems to add 3 extra bytes at the beginning of the file to identify it as a utf8 file before it encodes the shebang line. I am almost sure this is the problem. Will get back with an update