Tony Finch <dot(_at_)dotat(_dot_)at> wrote:
On Fri, 14 May 2004, John Leslie wrote:
I find it worrysome that we're proposing things like
"v=spf1 mx -all"
meaning that the receiving MTA is expected to pull the entire list of
MX records, resolve them to IP addresses, and check whether the one
IP address of interest is in that set.
Most MTAs do this already (apart from the last step, which is a nugatory
amount of work). Why is it worrying?
Actually, MTAs when sending do a single DNS lookup on MX, which
returns the IP address(es) as Additional data; they pick the first
IP address of the lowest "preference" MX record returned. It is
quite rare for any more than that to be required (unless the MTA
persists in the absence of any MX records).
In sending, the MTA need only find any plausible (single) IP
address.
To accomplish the SPF function, a receiving MTA needs to find
the IP address for _every_ MX server before it can move on to the
next directive.
That is what constitutes an "arbitrary" computation load, and
it's fundamentally different from the computation load for sending.
--
John Leslie <john(_at_)jlc(_dot_)net>