I would prefer signed updated over unsigned updates as Jakob suggested. But strictly speaking, IMO we should only accept signed updates, because it's the number of AS that we add in the update that we are protecting. By accepting unsigned update we may accept unprotected path information.
- Shankar K A -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Jakob Heitz Sent: Wednesday, November 16, 2011 9:08 AM To: Russ White Cc: [email protected] Subject: Re: [sidr] Burstiness of BGP updates > -----Original Message----- > From: Russ White [mailto:[email protected]] > Sent: Tuesday, November 15, 2011 7:32 PM > To: Jakob Heitz > Cc: [email protected] > Subject: Re: [sidr] Burstiness of BGP updates > > > > The only utility I can see is in protecting reachability. > > The only problem I can imagine with installing an unsigned route > is > > that the destination becomes unreachable. If it was unreachable to > > begin with, no harm is done. > > When you're protecting reachability, what are you protecting? > Whether or not someone can reach something. I assume that the > "something" you're trying to protect reachability to would/must > include things where you enter your password. > > Hence, I look at this entire problem a little differently than simply > trying to enforce a small subset of policies, or as a theoretical > exercise... If we can't prevent real world consequences with this > work, then --why are we doing it? We are doing it to protect reachability. We are not protecting your password in clear text on the internet. -- Jakob Heitz. x25475. 510-566-2901 _______________________________________________ sidr mailing list [email protected] https://www.ietf.org/mailman/listinfo/sidr _______________________________________________ sidr mailing list [email protected] https://www.ietf.org/mailman/listinfo/sidr
