IPv4 routes 'leaking' into RIPng (BIF)

bmanning@ISI.EDU bmanning@ISI.EDU
Wed, 5 Mar 1997 08:46:29 -0800 (PST)


> > What benefit was it giving you?  Surely, an IPv6
> > compatible IPv4 route is either going to be auto-tunnelled or routed via
> > IPv4.
> 
> Well, call me loony, but I thought that part of the idea behind 6bone was to 
> explore/examine the extent of the IPv6 feature set, particularly the -very-
> primitive routing capabilities.
> 
> I think the injection of this "spurious" route raised a couple of interesting 
> questions.  One respones was, "its confusing"  and there are your assertions
> above about autotunneling or routed via IPv4.  A careful examination of
> the prefix I injected will show that the IPv4 equivalent is never supposed
> to be routed in the IPv4 infrastructure.  A rough equal to the RFC 1918
> blocks. 
> 
> I think that before we, as operators, start making restrictive choices on
> IPv6 routing implementations, we should really have better criteria that
> the subjective assertions that have thus far been made.


	The prefix in question was  ::192.0.2.10

-- 
bill