DNAME issues (was Re: ip6.int deprecation)

[Geoff Huston]

>IMHO decent support for DNAME would be great in general, but in this
>particular case I would vote for ripping ip6.int completely instead of
>replacing it with the appropriate DNAME. Products still using it have to
>be fixed.


This view is consistent with the majority of advice I received when 
preparing the ip6.int deprecation draft a while back. While some folk were 
in support of using a DNAME in ip6.int there was a larger body of opinion 
that suggested that DNAME was not sufficient operationally robust, and that 
"ripping ip6.int completely" was appropriate in this case.

regards,

Geoff