On Thu, Oct 17, 2002 at 01:11:18PM +0300, Pekka Savola wrote:
>
> Yes, I believe the above draft is a good, simple starting point.
>
> There are still some unnecessary assumptions and extra text, I'll send
> specific comments later.
It seems that if you will trust RA's to build your address and select a
default router, then having DNS info (not just server IP) in the RA is
at first approximation a reasonable method (with obvious caveats).
There seem to be many methods applied to discovery in many IPv6 tools, be
it DNS discovery, ISATAP router discovery, etc... e.g.
by anycast address
by router advertisement piggyback
by well-known name
by well-known site-local unicast address
by well-known site-local multicast address
by DHCPv6
by Service Location Protocol
by ...
none of which may necessarily be secure.
It seems that DNS would be the only additional service required for
discovery via RA(?), but is there any merit to encouraging consistency in
what other discovery tools use? (e.g. it would maybe seem odd to have
each method used by different tools and to have to support them all, or
is that seen as preferable for flexibility?)
A review against existing proposals might be interesting?
Tim
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
