The biggest problem I have with this document is among those pointed out by
Wouter:
* The rule that .local names MUST be sent to mdns(port 5353). I feel
this is a little too strong, there are sites out there that have set ups
with .local in their unicast DNS. Propose: SHOULD.
As stated above, it's already a somewhat common practice to use .local
in *private* DNS namespaces (e.g., corporate networks), whether we like
it or not, and the current text in the mdns draft section 3 is incompatible
with this (i.e., it proposes to break them).
The current practice is cited in many places including:
http://tools.ietf.org/html/draft-kato-dnsop-local-zones-00
While it has yet been described in a RFC, .local is used to provide a
local subspace of the DNS tree. Formal delegation process has not been
completed for this TLD. In spite of this informal status, .local has
been used in many installations regardless of the awareness of the
users.
http://en.wikipedia.org/wiki/Top-level_domain
The top-level pseudo domain local is required by the Zeroconf protocol.
It is also used by many organizations internally, which may become a
problem for those users as Zeroconf becomes more popular.
And there's lots of places people have complained about this conflict
with mdns, such as:
http://lists.apple.com/archives/Macnetworkprog/2004/Oct/msg00089.html
http://www.markwilson.co.uk/blog/2007/11/managing-simultaneous-access-to-resources-from-both-internal-and-external-dns-namespaces.htm
http://www.macosxhints.com/article.php?story=20040806232315819
etc
-Dave
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf