Keith Moore wrote:
On Oct 24, 2011, at 5:59 PM, Hector wrote:
There is no suggestion to change the framework. There is an
growing issue with increasing deployment of Greylisting.
exactly what is the issue with the increasing deployment of
greylisting? what problems does it cause that are addressed by a
BCP or SMTP extension?
Keith, this has already been stated many times in the past two weeks,
and in my view, adequate opinion exist that BCP/EXT is desirable to
address many different the issues and specifically, its now documented
in the SMTP extension draft:
http://www.ietf.org/internet-drafts/draft-santos-smtpgrey-00.txt
http://tools.ietf.org/html/draft-santos-smtpgrey-00
The background section is all you need to read.
Further, you have Steve Atkins's related I-D
http://www.ietf.org/id/draft-atkins-smtp-traffic-control-00.txt
can you actually quantify how bad the problem is?
well, first, I am not the one saying its a major problem, but
recognizing a growing issues that IMV needs to be address at some
point. Its not a major issue yet, but the "smoke" is there.
Yes, I can 100% measure it and I think anyone can as well that have a
Greylisting Server employed and/or is monitoring its how volumes of
mail is delivered.
For this, one has to believe that delays are not good. If they
believe a compromise is ok, then they may now see an problem.
I can also measure the benefits of supporting 4yz time hints with
actual real results supporting MTA recognition of existing Servers
with time hints in their 4yz responses. This is more than an
optimization, but an solution that lowers the impact of Greylisting
servers had on standard SMTP systems before the Greylisting concept
existed.
Maybe questions to ask are:
Why did (at least) seven different systems add time hints to their
responses?
Are their existing MTA that support it, or they did write their own
MTA to support it and/or prepared for the future?
Are they see and negative adaption to it?
Are they see any benefits?
Are any of those authors here?
Thanks
--