Terry,
I see slight differences in the RPKI, in that we have overlap and
uniqueness issues. I agree that no one can force any replying party
to use any particular single or set of trust anchors. But lets say
two entities near this sphere of influence have the same prefix in
their TA which filters down to two contradictory ROAs. How does a
relying party make that "TA use" statement? How do they reasonably
or programmatically decide? Does your software help them choose
which is true? Or is this just a position of the RPKI is TA value
agnostic? And if that is so, should the drafts be made to reflect
that? Last time I looked there were numerous references to the
IANA->RIR->etc hierarchy and technically that is fine - but I think
misses David's scenario.
In the local TA management scheme I described, an RP gets to specify
the set of resources it believes is associated with a given key
(cert). The software enables an RP to assert what it "knows" to be
true, overriding what the repositories and the putative TAs may
assert. So, if there is a conflict relative to the info that the RP
says it knows to be true, what the RP says it knows to be true
prevails (locally). If the conflict arises at a point that is below a
cert that the RP ad adopted as a TA, and if the RP has no other info
about the resources in question, then no, the software can't help
with such a problem problem.
I think references to the IANA->RIR->... hierarchy are appropriate
for describing the extant allocation process. Moreover, it is
appropriate to note that the RPKI is designed to mimic that
hierarchy. It is also important to note that the choice of TAs is
always a local matter. If folks want, I can prepare an
(informational) I-D that describes one way of managing TA locally, as
per my presentation at the SIDR meeting.
So it certainly has its place - but doesn't address David's
position completely where an organisation, such as the UK
government, might wish to validly assert their own independent
trust anchor for their 51/8 SIDR hierarchy and have the rest of
the relying parties on the internet collect it and validate routes
against it.
The problem is that nobody, including the UK gov, can ensure that
every ISP in the world will recognize it as a TA, period.
Sure. But if we are looking for a qualitative trust statement of
TAs, where does one look? Or don't we? Is this just not a problem to
care about?
Statements about perceived trust in TAs are useful in PKIs that
anoint 3rd parties as TAs, independent of real world authorization.
The RPLI is not such a PKI. Instead it seeks to have the real world
entities that manage allocation of resources act as CAs. I would urge
us to NOT try to make the RPKI into a trusted 3rd party PKI.
Steve
_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr
