RPKI for dummies

John Kristoff jtk at depaul.edu
Thu Aug 20 14:15:29 UTC 2020


On Thu, 20 Aug 2020 13:20:53 +0000
Dovid Bender <dovid at telecurve.com> wrote:

> How do ISP's that receive my advertisement (either directly from me,
> meaning my upstreams or my upstreams upstream) verify against the
> cert that the advertisement is coming from me?

Nothing about your BGP announcements needs to change.  Through ARIN you
create one or more route origin authorizations (ROAs) with your public
key.  ARIN can even do all the work of creating the key pair for you if
you like.  You might try creating test ROAs in their operational test
and evaluation environment (OTE) environment to see how this process of
creating a ROA works.

ISPs obtain these ROAs apart and separately from the BGP  system.  ISPs
that fetch your ROA(s) and other RPKI objects through the RPKI
ecosystem, perform validation, and communicate AS origin and prefix
information contained in these ROAs to BGP routers.  At that point
this information is used to inform the route decision process,
comparing received routes with processed ROAs as part of a route
import policy.

> If say we have Medium ISP (AS1000) -> Large ISP (AS200) in the above
> case AS200 know it's peering with AS1000 so it will take all
> advertisements. What's stopping AS1000 from adding a router to their
> network to impersonate me,  make it look like I am peering with them
> and then they re-advertise the path to Large ISP?

In a nutshell, today, ISPs will only be able to validate the prefix and
origin AS you publish in the ROA, this is known as route origin
validation (ROV).  Today someone could advertise your prefix and
post-pend your AS to appear as the origin.

People are working madly on solutions to protecting other parts of the
BGP route attributes the origin AS, but nothing is currently, widely
deployed to provide that protection with the RPKI today.

John



More information about the NANOG mailing list