On Wed, Feb 15, 2006 at 11:08:33PM -0500, Stuart D. Gathman wrote:
On Tue, 14 Feb 2006, Craig Whitmore wrote:
Not really SPF. but...
I am using Exim at the moment and have SRS coded messages on all
forwarded messages and BATV coded messages on all sent out messages
directly .
Would there be any problem in people's opinions to use SRS for all
SMTP sent out messages as well instead of BATV. Or is using SRS
->forward. BATV->smtp send.. the "best" way of doing things? and
SRS on ALL messages ugly.
BATV for sent, SRS for forwarded is better than SRS only. I had used
old SES (before they included headers and body) for the purpose, but
ran into difficult with greylisting (sigs change too often) - so had to switch
back to SRS only. Now that you've prompted me to examine BATV, I think I'll
use that instead of old SES. With daily timestamps, it shouldn't be a problem
with greylisting.
This is another example of the whitelisting problem you spoke about: a
greylister _could_ extract the underlying address, and a sender could work
around known greylisters that don't, by varying the timestamp algo ?
It seems that although the localpart hack buys a lot with servers who aren't
aware of it, it buys even more when they are.
SRS only has the advantage of being simpler (don't have to test whether
mail is forwarded). But it sure is ugly.
To think I used to laugh at X400 addresses :)
I can't help thinking that all this putting structure into the local part
is an ugly hack and that thinking in terms of a clean design first and
only then how to hack it into the existing infrastructure might be the
answer (but I expect that's not a new idea).
Regards,
Paddy
--
Perl 6 will give you the big knob. -- Larry Wall
-------
Sender Policy Framework: http://www.openspf.org/
Archives at http://archives.listbox.com/spf-discuss/current/
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to
http://v2.listbox.com/member/?listname=spf-discuss(_at_)v2(_dot_)listbox(_dot_)com