On Tue, Jul 18, 2006 at 11:12:16AM +0200, Olaf M. Kolkman wrote:
> So I would rather see that turned around: Within the context of
> specific resource records underscore labels can define a semantic
> scope. In other words a domain name with an underscored label only
> has specific semantic scope when used in combination with a RR type
> for which the semantics are described/defined.
this I'd like to support, maybe s/semantic scope/resolution context/.
Such context can either be provided by the prefix or by the "suffix",
i.e. some ancestor domain, not necessarily the TLD. A precedent is that
A RRs have a very specific meaning in the "normal" tree and a different
one in the reverse tree. This overloading/context building has its
drawbacks, though.
> *._tcp.example.com 86400 IN TXT "I only care for sip"
>
> Here the underscored labels do not have any relevance to the TXT
> record. Besides the same TXT RR would show up were a dns client to
> query for a tcp service other than sip.
(With qtype==srv?) Other than that, yes, those RRs can't be forbidden,
they just have no defined meaning -- as don't TXT RRs in today's
forward tree.
RFC 2782 seems to suggest that the SRV RR type is only defined for
"underscored" domains. That I'd consider broken, the RR type definition
needs to be owner-agnostic. A particular application ("using SRV RRs to locate
the foo service") however may well make use of or depend on the contextual
information.
> The use of underscore labels with a particular RR is just convention
> and the chances of folk not sticking to the convention (and creating
> interop problems) becomes more problematic when the RR type has a
> more general use. In other words there are still good reasons to
> specify new RRs for specific use.
$ack++;
-Peter
.
dnsop resources:_____________________________________________________
web user interface: http://darkwing.uoregon.edu/~llynch/dnsop.html
mhonarc archive: http://darkwing.uoregon.edu/~llynch/dnsop/index.html
