socket

[David Bolen]

"Billy Ng" <evebill8 at hotmail.com> writes:

>         Second, again, I want to understand more about how the SMTP protocol
> works for my TCP/IP class project.

And from a pure learning perspective, that's a perfectly fine idea,
one I strongly suggest to anyone trying to understand how the Internet
works under the covers.

In terms of the servers not especially behaving as you may be finding
documented in your book (or the relevant RFCs), I think what you'll
find is that most have simply removed handling of commands that turned
out to be abused by spammers.  Originally, both VRFY and EXPN did in
fact do what they were written to do in most servers (I used to use it
all the time via telnet to manually try to check an e-mail address).
But as the net grew, they quickly became abused to try to build up
and/or verify distribution lists for spam, so the implementations
tended to either refuse the commands or just always acknowledge them
regardless of content.

Also to be fair, in many cases originally VRFY would still not be 100%
accurate, since if a mail server was built to relay mail elsewhere,
then by definition it would VRFY any address that was properly formed
(but would be relayed off that machine) as being good - even if it
would eventually bounce once a message to that address was really
transmitted.  In that case the VRFY was just telling you the address
was good to the machine you were talking to, but unless you were
certain that machine was responsible for final delivery of the mail,
it might not guarantee delivery.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/