On Wed, Oct 16, 2002 at 11:07:55AM -0700, Alper E. YEGIN wrote: > > As far as I understand, people are not suggesting changing DAD, but instead > developing an optimized version of it. Both versions should be able > to co-exist, no interference.
That's exactly right, Alper. My reason for using the 'Override' bit rather than a dedicated 'Tentative' bit (like earlier Optimistic drafts) was because existing nodes (assuming they implement 2461/2462 correctly) will understand it already. There's a couple of odd cases involving an address collision and a new connection to that address within a very small time window, but they're quite improbable and easily recoverable by standard NUD without too much fuss. I'm in the process of coding the draft up as a patch to Linux, and crippling the random generator to induce collisions, so we'll be able to get some real test results soon too. If anyone has any questions about the draft details, ask away! -----Nick -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
