[ppml] alternative realities (was PIv6 for legacy holders (/wRSA + efficient use))
Paul Vixie
paul at vix.com
Thu Aug 2 20:17:25 EDT 2007
[vixie]
> who is this "we" that you think is going to do the right thing when all
> of the wrong things are being done distantly from "us"?
[leibrand]
> I'm also optimistic that we've learned some lessons from the last round
> of deaggregation, and will have narrow-spectrum therapies (like
> as-pathlimit) to treat the next round.
i'm not getting through here. scott, there is no "we" in the "routing table
full" scenario. last time around "we" dodged the bullet because the network
was small, and CIDR was an easy way to accomodate future growth, and moore's
law was an easy way to cope with current growth. if pool depletion leads to
mass deaggregation and steps in the global table size, it'll be every network
for itself -- the ultimate assymetric cost:benefit equation -- a meltdown.
[leibrand]
> For example, there's no particular reason that a North American network
> needs to hear all the /24 deaggregates from Asia and Europe if they're
> covered by larger aggregates, and vice versa.
covered by larger aggregates having the same path? or would it then be
possible for competing wet transit operators to out-duel eachother using
announcement sizes? and are you SURE there would be any lost TE benefit,
even if anycast wasn't in use? these are distant people's networks, about
which comparatively few local decisions can be well informed.
[leibrand]
> It's not really a case that "they" are doing bad things, and "we" are doing
> everything right. Rather, deaggregation is something that is very useful
> locally, and not so useful further away, so if filtering becomes necessary,
> it would be appropriate for everyone to filter distantly-originated
> deaggregates, and continue listening to locally-originated ones.
that's an overly sweeping information losing generality.
(btw, i was serious when i said i hoped you'd tell the internap ipv6 story
in a nanog presentation... we've got to get more network operators thinking
about this subject.)
More information about the ARIN-PPML
mailing list