[EMAIL PROTECTED] wrote:
>
> >>>>> "Mikael" == Mikael Olsson <[EMAIL PROTECTED]> writes:
>
> Mikael> Undoubtedly. (Which by the way is why the fix for all the current FTP
>problems,
> Mikael> IMHO, is to enforce passive mode FTP which exposes the servers but saves the
> Mikael> clients.) An unmanagable amount of workstations all in the hands of ...
>*shudder*
>
> Actually, the _correct_ fix is for clients and servers to implement
> the latest FTP i18n RFC, which makes NULL escaping CR characters
> mandatory. If clients and servers did that, and only allowed CRLF to
> terminate a command/response, we wouldn't have these problems.
Nope, that won't help. Go visit [EMAIL PROTECTED] and read the latest
fun we've been having.
> Of course, if you _really_ want to fix FTP, drop the silly data
> connection entirely and do the whole thing over a single
> connection. Of course, _that_ would basically be a new protocol with
> only a vague resemblence to FTP. Which is the whole point :)
I definately agree. I'd propose something like an inline command, "INLI",
(compare "PASV") to handle this. This way you can still be backwards
compatible for a while.
> Hmmm... HTTP/1.1 methods for NLST, PWD, etc... maybe ;)
Even better. I prefer the HTTP approach of not necessarily keeping
connections open over the FTP one of always keeping the command
connection open.
--
Mikael Olsson, EnterNet Sweden AB, Box 393, S-891 28 �RNSK�LDSVIK
Phone: +46 (0)660 105 50 Fax: +46 (0)660 122 50
Mobile: +46 (0)70 66 77 636
WWW: http://www.enternet.se/ E-mail: [EMAIL PROTECTED]
-
[To unsubscribe, send mail to [EMAIL PROTECTED] with
"unsubscribe firewalls" in the body of the message.]
