At 3:46 +0200 2003/09/04, Oliver Eikemeier wrote:
>Btw, if you do not receive my mail:
>
>SMTP error from remote mailer after RCPT TO:<andy@???>:
>host lyra.dyndns.org [217.237.102.138]:
>550-Verification failed for <eikemeier@???>
>550-Called: 62.138.193.82
>550-Sent: RCPT TO:<eikemeier@???>
>550-Response: 550 Sorry, your host 217.237.102.138 is listed at
>dnsbl.njabl.org
>550 Sender verify failed
>
>
>You *shouldn't* do callouts from a dialup
>(pD9ED668A.dip0.t-ipconnect.de), otherwise
>you wont ever receive mail from an AOL user (or from us...)
in my opinion, as I wrote a long time ago, callbacks (and
consequently bounces too) should not be subjected to any checks until
the data phase (unless the number of recipients is larger than 1..).
At the data phase they can then be scored and rejected (and since
there is only one recipient this can still be recipient dependent).
Incidentally, at the current state of the protocol, I think that
multiple recipients should be rejected outright for null senders.
Giuliano
--
H U M P H
|| |||
software
Java & C++ Server/Client/Human Interface applications on MacOS - MacOS X
http://www.humph.com/
