Matt, we went backwards and forwards many times on these timeouts in earlier versions. I agree with what you say as a common sense implementation heuristic, but I don't see how to cover it algorithmically in the spec.
Since we are now at the stage of having resolved the IESG comments, do you really want another text update? Brian Matt Crawford wrote: > > > http://www.ietf.org/internet-drafts/draft-ietf-ipv6-flow-label-08.txt > > The host ought to wait distinctly more than 120 seconds before reusing > a flow label, lest the last packet of the previous flow with that label > be delayed by a second more than the first packet of the new flow and > some router takes action based on the assumption that the two packets > belong to the same flow. > > The amount by which the wait time should exceed 120 seconds should > naturally be determined by the maximum packet lifetime, but IPv6 > discarded that IPv4 notion. > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > [EMAIL PROTECTED] > Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Brian E Carpenter Distinguished Engineer, Internet Standards & Technology, IBM NEW ADDRESS <[EMAIL PROTECTED]> PLEASE UPDATE ADDRESS BOOK -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
