Chris Siebenmann wrote:
> You write:
> | >> Don't use SRS where the
> | >> SPF record contains "+all".
> | >
> | > Instead, just reject the message. It's almost certainly spam.
> |
> | Probably won't even see it here.... rDNS fail IS a hard-fail.
> |
> | But I have been known to tweak SA to add penalty points to users
> | of DK, hashcash, and S**. And enough points to hard-fail those
> | organizations that 'bless' direct advertising registrants with a 'free
> | pass' of negative points.
>
> I suspect that you'll see more and more domains with SPF records,
> because it is becoming a more and more reliable way of getting GMail
> to accept your email relatively promptly. We ran into this issue here,
> in part because a lot of our users forward their email to GMail, and
> so I held my nose and added a SPF record.
>
> - cks, who is not a fan of SPF in general
Interesting possibility.
But with a year-plus of logs showing consistent QT and DT to the Gmail's
receiving MTA of only 2 or 3 seconds off a server clear out in Hong Kong, China
.... I'm not sure what could be improved if I *were* to adopt any of those
'goodies' not in the decades-old smtp specs.
What sort of delays were you seeing, and could there have been some other cause?
As to 'more and more domains'. Published? perhaps. Required? Questionable.
I think that is more likely with DKIM than either SPF or SRS, and not all that
hard and fast with any of them.
Not only old-line 'purists' that insist the original smtp spec - if only FULLY
IMPLEMENTED - is good enough a barrier to abuse, and simpler - but a far greater
number of lazy or 'iggerant'.
;-)
Bill
