Stuart D. Gathman wrote:
On Sat, 10 Sep 2005, Theo Schlossnagle wrote:
already doing this. However, SPF having awful (show stopping)
consequences for senders who's mail is forwarded by someone who doesn't
care about SPF means that you can have something mostly useless before
Forwarders, even those who don't care about SPF, do not break SPF.
The "show stopping consequences" you describe are simple a braindead
receiver, who installed/purchased a forwarding alias, but neglected
to account for it in their SPF setup.
You talk about a receiver that uses and alias and a receiver who
controls an SPF record as the same person. They hardly ever are. The
domain owner controls the SPF record, the 10k or 10M email accounts the
host have widely varying needs that often include being on the recieving
end of aliases the owner has absolutely no clue about.
You buy me a machine and internet connection capable of fully
processing 30000 messages/day, and I'll get my head examined. By rejecting
at SMTP time, I can keep the "idle" incoming SMTP bandwidth down to 56K.
You can process 30000 messages per day on a 386 -- that's less that half
a message per second. On a dual Opteron you should be able to achieve
30000 message per minute with good software (that includes SPF,
SenderID, feature extraction and content analsys, DNSBL, etc.). As for
the Internet connection, 30k message at 32k each (which is large) is a
90kbs connection. And as I said "any large provider that blocks
outright on SPF in the SMTP phase needs their head checked' I'll stick
to my guns and say that 90kbs is chump bandwidth for a large provider.
--
Theo
-------
Sender Policy Framework: http://spf.pobox.com/
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