On Tue, Oct 24, 2006 at 08:15:00PM -0700, Scott Kitterman wrote:
which in itself is correct: _test really is an invalid hostname.
What kind of entry did you try to create?
With DynDns it's an A, with noxa.de I tried directly a TXT.
I've tested the free service of DynDns, not the paid service.
Alex will no doubt correct me when I get this wrong, but here I go anyway...
<grin> you'd rather have a mistake be cause of problems? I think not.
summary: if CNAME points to non-existing domain, then the original
query is allowed to return NXDOMAIN. But this only is the case
when the original domain has a CNAME RR attached. In 1999, Vixie
made the mistake and in 2004 he corrected it. Now, in 2006, djb
still (ab)uses vixie's mistake to sell the misconception that
an empty resource set should result in NXDOMAIN.
http://www.ops.ietf.org/lists/namedroppers/namedroppers.2003/msg01586.html
This may be important for us. I'm thinking about the exists mechanism,
which could be made very light weight if no RR records are required.
However, if djb software, or anything based on that like pdns, is going
to mess things up then it won't work.
Anyway, I was still looking at this:
I think when you add an A record that has to be a valid hostname,
indeed, I think so as well. An A record points to a hostname (not
just to an interface) and thus cannot have underscores. I still
have a feeling that there may be a later RFC somewhat weakening
this principle.
Alex
-------
Sender Policy Framework: http://www.openspf.org/
Archives at http://archives.listbox.com/spf-discuss/current/
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to
http://v2.listbox.com/member/?listname=spf-discuss(_at_)v2(_dot_)listbox(_dot_)com