I fully agree with James. Even without supporting RO, the CN can communicate with any MNs using a bidir-tunnel (which is your concern, though).
>From an implementor's point of view, manating RO will pose noticeable extra work to all IPv6 implementors. It is not so easy to implement RO (which means implementing the RR procedure and BCE management). Can't we deploy RO gradually? At the initial stage, the IPv6 nodes which support RO may be restricted to some domains (such as an IP telephone carrier). But even in such an environment, we are able to communicate with those nodes using our nodes which are not support RO. If people want to communicate with moving nodes and want to comunicate with nodes even when they are moving, more need for supporting RO is raised. Some time ago, the MIP6 draft said that all IPv6 nodes had to support HAO. Because of this, some implementations currently shipped process HAO. This is bad. Bad because processing unverified HAO has threats already discussed in mobile-ip WG. We now have a option not to mandate HAO processing. In this case, we communicate with those not supporint RO using bidir-tunnel. I think more needs for mobility applications will make 'SHOLD' to 'MUST' in fact, like the TCP flowcontrol. Best Regards, --- Keiichi SHIMA IIJ Research Laboratory <[EMAIL PROTECTED]> KAME Project <[EMAIL PROTECTED]> -------------------------------------------------------------------- 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] --------------------------------------------------------------------
