[ietf-dkim] The URL to my paper describing the DKIM policy options
dotis at mail-abuse.org
Tue Jul 25 07:56:02 PDT 2006
On Mon, 2006-07-24 at 22:27 -0700, Patrick Peterson wrote:
> > http://www.ietf.org/internet-drafts/draft-hallambaker-pcon-00.txt
> I think this is a great idea and am surprised it didn't generate more
> traffic on the list. It's not easy to cram needed new functionality into
> a backward-compatible solution.
This concept requires a wildcard PTR record at every label. This
assumes there are no other PTR record used beyond reverse lookups, which
may not be the case. Name errors would not be reported, which might
make other searches difficult and perhaps risk flooding caches with PTR
RRs. This will also require special treatments at delegation points.
The intent was to point to a general label where policy references are
found. Philip suggested this should point to an HTTP server. HTTP is
needed to support the size of an all encompassing policy response.
If the PTR record can not be overlaid, then a new RR type is needed.
Defining a new RR type allows this RR type to be wildcarded at every
label and contain the policy information specific for a particular
protocol. No multi-record lookup or HTTP would then be required. : )
Either way, it would be rather messy adding a RR at every node. If the
ultimate repository for this information is HTTP, it seems a protocol
using HTTP should be used instead. This HTTP service could locate all
the "registered" policy records. Perhaps a few organizations would be
willing to provide a policy redirection service.
More information about the ietf-dkim