SA Bugzilla – Bug 3493
RFE: new SPF-based whitelisting rules
Last modified: 2005-03-05 07:43:39 UTC
So some postings from Dan on the SPF list brings this to mind. I suggest we identify some "known-good" sending domains, like ebay.com, amazon.com, etc., where we can use rules like the following: header SPF_WHITELIST_DOMS From:addr =~ /\@(?:amazon\.com|ebay\.com|whatever)$/ meta SPF_WHITELIST_FROM (SPF_PASS && FROM_EQ_ENV_FROM && SPF_WHITELIST_DOMS) In other words, give whitelist points to mails where: 1. env-from is validated by SPF 2. header "From:" matches env-from 3. domain in question is *known* to be good The domains in SPF_WHITELIST_DOMS could be very extensive; in fact, we could probably do that better as an eval test/plugin with its own config items, a la spf-whitelist-domain ebay.com amazon.com aol.com yahoo.com ... sound useful? AFAICS this will be reliable whitelisting even despite SPF shortcomings.
more accuracy and performance bugs going to 3.1.0 milestone
Duplicate of Bug 3487. Although, currently, the implementation for bug 3487 doesn't require that the header 'From:' match the envelope header used for SPF validation. Is there really a benefit to doing so? If it comes from a known-good sending domain and passes SPF validation, do we really want to keep up with what they are sticking in the 'From:' header?
It might make sense to give more points when From: matches EnvelopeFrom. That being said, I'll leave it to Daryl to see how this fits into bug 3487, marking as a dup. *** This bug has been marked as a duplicate of 3487 ***