On Sunday, January 29, 2012 08:18:33 PM Murray S. Kucherawy wrote:
> I think it's prudent for us to add two more report types for both ADSP and
> DKIM: "p", meaning the DKIM or ADSP test failed for local policy reasons,
> and "o", meaning there was a problem with DKIM or ADSP that was something
> other than the report types already listed.
> 
> DKIM in particular can be presented a perfectly valid DKIM signature that
> verifies, but the verifier could decide it doesn't want to accept the
> signature anyway for policy reasons, including (real use cases here)
> Subject: wasn't signed, or "l=" didn't cover enough of the message
> percentage-wise.  The existing report requests don't cover those.
> 
> I don't know if there's an SPF equivalent for "p", but I suspect adding the
> "o" case would be a good idea.

"p" is embedded in SPF already.  If the SPF result is fail and you are 
rejecting due to SPF, it's by definition a policy issue.

I would suggest that "o" isn't needed for SPF since we've already covered the 
spectrum of reasons for an SPF related rejection and so if one is rejecting a 
mail due to some other reason, it's not an SPF auth failure that's causing the 
rejection, it's another type of policy of some kind (e.g. passes SPF, but it's 
a known spammer domain, so reject the mail isn't an SPF issue).

Scott K
_______________________________________________
marf mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/marf

Reply via email to