ietf-mxcomp
[Top] [All Lists]

Re: consensus call on pra/mailfrom deployment and versioning/scope

2004-09-08 14:25:31

Several items:

-= 1 =-
Someone asked for an example of a single record that works for both scopes:
        spf2.0/pra,mailfrom +mx -all

-= 2 =-
Yes, scopes would need an IANA registry. I would suggest that it needs to be limited in the greatest possible way (like amended only by a working group draft - anyone know language for this? an existing example of such a registry? -- e-mail me off-list.)

-= 3 =-
Please note that the Classic SPF's use of the HELO/EHLO domain when the reverse-path is null does not constitute a check of the HELO scope (such as done by SPF/HELO or CSV.) It is constructing what the return path should be even though the sending MTA has omitted it ("Mail From <>") to break possible mail loops. It is still checking the envelope concept of "Mail From". It is not making any assertions about the HELO/EHLO domain as used by a SMTP client.

-= 4 =-
I would not support a HELO scope for the same reasons mentioned by others. In particular, 1) CSV is this group's next agenda item, 2) The full syntax of SPF records is totally unwarranted for HELO checks (you only need "+a -all" and "-all".) 3) One generally never publishes records about HELO at the same domain name as MAIL FROM or PRA. Hence they'd never share a record.

-= 5 =-
As per my discussion in prior messages, the scope macro doesn't buy much. I am not in favor of it, though not strongly.

        - Mark


Mark Lentczner
http://www.ozonehouse.com/mark/
markl(_at_)glyphic(_dot_)com


<Prev in Thread] Current Thread [Next in Thread>