"Jon Kyme" <jrk(_at_)merseymail(_dot_)com> wrote:
You mean not accepting connections from some host drives them to
hammer on your firewall?
Yes. It increases the number of TCP SYN packets by a factor of 10
or more.
Accepting all connections increases the total traffic, but the
number of SYN packets drops dramatically.
But I don't want to increase the total traffic, surely?
I think that this is a truly exceptional situation.
I've seen it, as have a number of other people with spamtraps.
I believe you. That doesn't make it not exceptional (i.e. normal).
A spamtrap is a rather particular case. All messages are accepted - all
are successfully delivered ("success" and "deliver" have very particular
meanings in this context). No DSN are issued. No problem.
I think it's probably right to mention the spamtrap case in any document
dealing with filtering in SMTP (such as the one I'm *still* writing).
Thanks for bringing it up.
--
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg