On Wed, Oct 5, 2016 at 3:58 AM, Nick Hilliard <[email protected]> wrote:
> [email protected] wrote: > > I'm all for a permanent solution, but routers just don't get updated > > with new features all that often. > > Extended communities have been around for years. Peoples' lack of > inclination to properly maintain their infrastructure is not a good > reason to introduce additional unnecessary complexity by defining > byzantine workarounds. > > There seems to be a good deal of vendor / operator support for large > communities. This fills in the rfc1997 gap for asn32s extremely well, > and should shortly start appearing in production code. > > noting I don't disagree with nick, but a bit more color / reason for my comment I think is helpful: o these communities are not expected to work across more than the immediate remote peer, except possibly in the case of IX's where the same next bit of logic should hold, I think. o the folk that want to exploit the use of these new communities have a reason to upgrade code to that which supports them o upgrade cycles are shorter if people have a reason to upgrade I think, therefore, that a complex solution as a stopgap isn't worth the work, when in 'short order' code and ability to use that on the network will be here.
_______________________________________________ GROW mailing list [email protected] https://www.ietf.org/mailman/listinfo/grow
