Bart Schaefer <schaefer(_at_)brasslantern(_dot_)com> wrote:
[ re: outgoing SMTP servers blocking all inbound traffic ]
Servers configured that way may find that they are not able to send
mail to aol.com addresses. AOL now periodically tests IPs from which it
recieves inbound mail to determine whether they are open relays. My
experience has been that if those IPs simply refuse inbound SMTP, they
get put on AOL's blacklist; to pass AOL's open relay test, the server
must both accept an SMTP connection, and then refuse to transmit.
That is, in order to discover if the messages were sent
consentually, AOL has implemented yet another broken "consent"
protocol. That "consent" protocol is unable to determine that two
methods for refusing to relay are semantically equivalent. Further,
if you don't consent to their broken "consent" protocol, they don't
consent to communicate with you at all.
So we need a BCP saying why their practices are unhelpful.
Or, we need a BCP from AOL, saying why their practices are helpful.
Or, we need to design a better consent protocol which doesn't have
these problems, so AOL can implement it.
While I understand that current network practices make it difficult
to implement a consent system, we should NOT give up trying to design
or implement such a system because of those difficulties. We are here
to design a consent system precisely BECAUSE of difficulties in
current practices.
Alan DeKok.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg