On Thu, 14 Mar 2013, Murphy, Sandra wrote:
what will an operator do differently for these two shades of grey?
Good question. *Operators*, would you ever treat these differently?
I thought Ruediger was saying that the need for this 4th state arose as
his company was studying using RPKI.
From what I understood he's missing a state for unverifiable resources. eg.
when the validator isn't reachable.
Reading the RFC6811 again, I think the state is covered by the NotFound
state; No VRP Covers the Route Prefix.
The problem arises when strict RPKI policing is used. Using a strict policy
you'd only accept Valid prefixes and drop _all_ the others.
For this case a 4th state could be introduced. This state would only be
assigned if there are no entries in the entire VRP. The strict policy should
then accept Valids and 4th state prefixes.
Jac
Ruediger, was there a difficulty? a need not met by just 3 states? can
you say what it was?
--Sandy, speaking as regular ol' wg member
_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr
--
Jac Kloots
Network Services
SURFnet bv
_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr
