On 10/08/2012, at 1:18 PM, Randy Bush wrote: >> an interesting outgrowth of the grandparenting could be the ability to >> 'avoid' LEA actions at middle tiers of the address allocation >> heirarchy... that's something to consider, i'd say. > > yep. the point is > o this is not pretty > o but it is well defined > o and can be useful as a hack around breakage > > so i wrote it up
And for that I am grateful. Truly. But I think what might be of value (huge earth shattering value) is to take this, and also an analysis of the RPKI for where actors can create valid objects which will result in routing outcomes that are not by the strict design of the resource holder. Such a document I think, in light of this observed functionality, is near mandatory to identify and perhaps ultimately address the imperfections. Not saying that RPKI and BGPSEC needs to be perfect first go, or even that we share a common definition of "imperfection" but I see this as a crack in the looking glass. T. _______________________________________________ sidr mailing list [email protected] https://www.ietf.org/mailman/listinfo/sidr
