On Fri, Jul 18, 2003 at 02:46:17PM +0200, Dave Crocker wrote
In other words, we are working at too high a level for the choice of
networking-level protocol to be relevant.
However, if anyone can explain how IPv6 details are relevant to Internet
email-related spam control mechanisms, please do.
Maybe a better way to put it is that any proposal should not break
when ported to an IPV6 environment. DNSbls are going to be interesting.
IPV6 has only one address pointing to "self", i.e. ::1. IPV4 has 2^24
addresses in the 127.0.0.0/8 loopback block. This allows aggregators
like Osirusoft to return different answers for an IP address, depending
on which of several lists it may be in. zz.countries.nerd.dk returns
127.0.X.Y where X * 256 + Y = numeric ISO country code of the country of
origin of the IP address.
The current IPV4 decimal values of 127.X.Y.Z can be represented as
::FFFF:127.X.Y.Z under the legacy-support feature of IPV6. What worries
me is that outfits with a vested interest in adoption of IPV6 may go out
of their way to disrupt legacy support for IPV4 sooner rather than
later. And it is legacy support, make no mistake about it.
I suggest using link-local 64-bit address space for DNSbl return
values. This will minimize the chances of stepping on somebody else's
toes. For backwards compatability, may I suggest ::7F:0:0:0 through
::7F:FF:FF:FF. This will map exactly one-to-one with the space
::FFFF:127.0.0.0 through ::FFFF:127.255.255.255. A comparison that uses
the 3 or 4 least-significant-bytes won't care whether the answer came
from an IPV6 machine or an IPV4 machine.
--
Walter Dnes <waltdnes(_at_)waltdnes(_dot_)org>
Email users are divided into two classes;
1) Those who have effective spam-blocking
2) Those who wish they did
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg