Re: [sidr] Injecting idea of "freshness of repository data" into BGP

[Jeffrey Haas <[e-mail]>]

On Wed, Mar 28, 2012 at 09:02:24PM -0400, Danny McPherson wrote:
> On Mar 28, 2012, at 4:19 AM, Jeffrey Haas wrote:
> > Per my mic comment at IETF 83:
> > During the San Diego interim session we had discussed potentially signaling
> > in BGP the idea that a given AS may have fresher data available in its
> > repository. 
> 
> Shouldn't this problem be solved in the resource certification infrastructure (i.e., RPKI) - signaling RPKI freshness in BGP and distributing to literally millions of routers seems like a REALLY bad idea to me.

One "route" per AS in the system which is a 20byte value max (per the cert)
hardly seems that scary. :-)

But that said, I don't object to some sort of mechanism used as part of the
RPKI infra would do such a "you may want to refresh" request.  My thought is
that since the certs in question are required for validation, the routing
system already has a strong interest in making sure downstreams can validate
(or invalidate!) routes.


-- Jeff