Fred, Good point regarding ICMP rate-control at logical (or pseudo) interface level, a typical example being tunnel interfaces. I hope Mukesh takes note. Thanks, Alex Fred Templin wrote: > Alex, > > --- Alex Conta <aconta@txc.com> wrote: > >>My concern is clarity. >> >>"The rate-limiting parameters SHOULD be configurable per node. >>and/or per interface if the node has dissimilar speed/bandwidth interfaces. >> >>is not clear enough. >> >>"The rate-limiting parameters SHOULD be configurable per node. >>If a node has dissimilar speed/bandwidth interfaces, the rate-limiting >>parameters SHOULD be configurable per interface. >> >>is clearer. > > > To be even clearer still, perhaps there should be some mention > of what is meant by "interface". In particular, it seems that > rate limiting could be either per-physical interface (like, > one rate limiting parameter per multi-access Ethernet link) > or per-logical interface (like, one rate limiting parameter > per pt-to-pt tunnel). > > In the former case, a single rate limiting parameter could > produce too few ICMPs for some neighbors but too many for > others. In the latter case, rate limiting parameters are > per-neighbor and provide the correct levels of ICMPs. But, > the physical/logical interface distinction should be > clearly spelled out. > > Thanks - Fred > osprey67@yahoo.com > > > >
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
-------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------