On 2019-02-09 19:39:34 -0500, valdis(_dot_)kletnieks(_at_)vt(_dot_)edu wrote:
On Sat, 09 Feb 2019 23:38:15 +0100, ?????????? ???????????????? said:
if on 250-GREYLIST on retries the IP address of the client is changed,
then
the Greylist-421-retry=00:00:10 puzzle is not solved and the time begins
again
from the beginning on each IP-client-address change. This is not in the
interest of the sender.
As I mentioned, I don't think it rises to the level of a SHALL - that's
usually
reserved for things where failure to do so causes a high risk of the
protocol
failing to work properly.
I also think a SHALL is inappropriate here on purely formal grounds: An
extension is by definition optional.
That fact that an extension is optional doesn't mean that no requirements
have to be met in order to implement the extension properly.
A client may or may not implement
it. Putting a mandatory requirement on a client which hasn't yet
indicated that it supports the extension is therefore impossible.
It's assumed that requirements given in an extension specification apply
to software that implements the extension in a compliant fashion.
Ned
_______________________________________________
ietf-smtp mailing list
ietf-smtp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf-smtp