On 06/27/2013 06:14 AM, Fred Baker (fred) wrote: > > On Jun 26, 2013, at 4:52 PM, Fernando Gont <fg...@si6networks.com> > wrote: > >> On 06/26/2013 12:56 PM, Mark ZZZ Smith wrote: >>> >>> One of the issues with conventional ICMP PTB based PMTUD is the >>> default rate limits on ICMP messages on broadband routers with PPPoE >>> subscribers i.e. dumbell [1500][1492][1500] MTUs, slowing down PMTUD. >>> Upping the limit on ICMP PTBs is an obscure parameter to have to >>> remember to change, and it isn't clear exactly what to change it to - >>> IIRC, the value I've used in the past is 1000/s up from 10/s. RFC4821 >>> would be nice because it would avoid control plane processing that >>> ICMP based PMTUD causes. >> >> Many implementers don't want to rely on a 100% ICMP-less PMTUD because >> of the potentially-lng convergence time (i.e. time to discover the PMTU). > > And when ICMP is blocked, and the convergence interval is infinite...?
Not sure what you mean. There ae two ways to implement RFC4821: 1) As a complete replacement of ICMP-based PMTUD, or, 2) For PMTUD back-whole detection Implementers I've talked to are fine with 2, but not with 1. If you process ICMPs, then when ICMP errors are received you don't need to rely on timeouts. Thanks, -- Fernando Gont SI6 Networks e-mail: fg...@si6networks.com PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492 -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
