By my count that went:
keep: 4
modify: 20
remove: 0
A clear consensus to modify. (See the attached for who I think went
for what.)
6 of the 20 for modify also expressed a preference for the text from
draft-levine-dkim-adsp-00.
So I take it that the authors should include a modified domain existence
check along the lines discussed in the subsequent thread. Probably the
best thing is if the authors agree some modified text amongst
themselves and post that either separately or in an updated draft (which
I believe is nearly baked).
Stephen.
PS: sorry for being slow with this again;-)
Stephen Farrell wrote:
There has been considerable debate in the past few weeks regarding the
need for a check for domain existence in ADSP.
I think we've had sufficient time for debating this, let's decide.
Please respond to this by Friday June 6th.
The text in question (from section 4.2.2 of draft-ietf-dkim-ssp-03)
is as follows:
2. _Verify Domain Exists._ The host MUST perform a DNS query for a
record corresponding to the Author Domain (with no prefix). The
type of the query can be of any type, since this step is only to
determine if the domain itself exists in DNS. This query MAY be
done in parallel with the query made in step 2. If the result of
this query is an "NXDOMAIN" error, the algorithm MUST terminate
with an appropriate error.
NON-NORMATIVE DISCUSSION: Any resource record type could be
used for this query since the existence of a resource record
of any type will prevent an "NXDOMAIN" error. MX is a
reasonable choice for this purpose is because this record type
is thought to be the most common for likely domains, and will
therefore result in a result which can be more readily cached
than a negative result.
There are three options that have been actively discussed:
a. Keep. Retain the current text as-is.
b. Modify, i.e. keep, but with a different set of records. It was
suggested that the current NXDOMAIN is incorrect, and that MX, A, and
AAAA records for the domain should be queried, with the existence of
any of these records indicating a domain that is potentially used for
email. If we have consensus for this option, then we may well need a
subsequent poll to decide the details.
c. Remove. Remove the text as being out of scope for the ADSP
specification. Some text may need to be added pointing out the need for
a domain existence check elsewhere. If the consensus is for removal,
then we should consider what, if anything, the specification should
refer to for performing the domain existence check.
Please just answer "keep", "modify", or "remove" in this thread, and use
a different subject line for any discussion.
Thanks,
Stephen.
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html
keep:
fenton(_at_)cisco(_dot_)com
eric+dkim(_at_)sendmail(_dot_)org
aiversonlists(_at_)spamresource(_dot_)com
arvel(_dot_)hathcock(_at_)altn(_dot_)com
modify:
lear(_at_)cisco(_dot_)com
steve(_at_)blighty(_dot_)com
jmacdonald(_at_)e-dialog(_dot_)com
robert(_at_)barclayfamily(_dot_)com
nobody(_at_)xyzzy(_dot_)claranet(_dot_)de
jbacksch-ietf-dkim(_at_)tca-os(_dot_)de
dotis(_at_)mail-abuse(_dot_)org
tony(_at_)att(_dot_)com
chl(_at_)clerew(_dot_)man(_dot_)ac(_dot_)uk
Bill(_dot_)Oxley(_at_)cox(_dot_)com (? not entirely sure)
MHammer(_at_)ag(_dot_)com
deepvoice(_at_)gmail(_dot_)com
wietse(_at_)porcupine(_dot_)org
ietf-dkim(_at_)kitterman(_dot_)com
modify according to draft-levine-dkim-adsp:
dot(_at_)dotat(_dot_)at
johnl(_at_)iecc(_dot_)com
esiegel(_at_)constantcontact(_dot_)com
dhc(_at_)dcrocker(_dot_)net
jdfalk(_at_)returnpath(_dot_)net
jon(_at_)callas(_dot_)org
delete/remove: none
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html