ietf-smtp
[Top] [All Lists]

Re: About TURN and ETRN command

2005-05-12 05:17:24

On Thu May 12 2005 07:24, Vijayan wrote:

i wanna clarify my doubt in using TURN and ETRN commands

You probably don't "wanna" use TURN; it has some security issues.

ETRN isn't generally useful; some of the problems are described along
with the specification, in RFC 1985.  Note that it can only be used
with ESMTP, and only if offered by the server as a supported extension.
 
This session will despatch the mails 1, 3 for mailbox 1 immediately.. and the 
remaining mails will be despatched after the smtp session quits..
is it like that..?????

No.

A single transaction consists of
 MAIL FROM: <path-for-notifications(_at_)example(_dot_)net>
 RCPT TO: <recipient(_at_)example(_dot_)edu>
 DATA
commands (after a session is started, of course). Between multiple
transactions, there is usually a RSET command.  And theree may be
multiple RCPT TO commands within a single transaction.

After a successful response to the DATA command, the message is queued
for delivery.  Even if the session times out or the TCP connection is
broken, that will happen (in conforming implementations).  A particular
implementation might immediately process (rather than defer queue
processing) a given transaction while other transactions are taking
place in a session.  So in general you cannot say when a particular
transaction will be processed -- once the server has issued a positive
response to the DATA command, it has taken responsibility for delivery.

ETRN is supposedly used to cause queued messages (on the server)
"destined" for some domain to be processed when the client sends an
ETRN command specifying that domain, as in
 ETRN foo.bar.grimble.pritz.example.com
One problem is that the server might not recognize that domain name;
it might queue messages for that domain under mx.blurfl.example.com,
for example (and in general, the client has no way of knowing that).
Note in particular that ETRN does not take a mailbox as the argument
(unlike MAIL FROM and RCPT TO).