> Date: Fri, 19 Jul 2002 10:37:29 +1000
> From: [EMAIL PROTECTED]
> Message-ID: <[EMAIL PROTECTED]>
>
> | Robert, I mentioned that this was a issue several years ago
> | (on namedroppers / bind-workers) with respect to MX vs A
> | and said that the search should stop on a NODATA response.
> | You were quite happy for MX and A to hit different fqdn at
> | the time. It looks like AAAA vs A has changed your opinion.
>
> Sorry, I don't remember the context, but the two cases aren't the same
> at all. If I am looking for an MX record, I am looking for an MX
> record. And if I want the MX record for "foo" that's what I want to
> find. Only if there is no MX record for "foo" should a A lookup
> even be attempted.
>
> On the other hand, when I want "the address of foo" (which is what
> getaddrinfo() without a specific address type hint does), then I am
> expressly asking for either an A or an AAAA (or even an A6...)
Well given that if you are looking for a MX for "foo" you
want to send mail to foo and the fallback if there isn't a
MX for foo is to look for a address record I would say that
it is a indentical senario.
> | Agreed, however changing *default* resolver behaviour is
> | difficult as it ends up breaking things. "localhost"
> | lookups is what tends to break if you disable this.
>
> Only for badly configured zones - the name "localhost" should be
> configured in every zone that is in a search list (if it doesn't
> exist in a zone that is on the search list means that there is no
> "localhost" existing, and the lookup should fail, going on to,
> or locally configuring "localhost." isn't what should be being done.
>
> Because of the way the decision was made to ground FQDN's in a trailing dot,
> but almost never actually represent it, there's no way for the PTR record
> for 127.0.0.1 to actually return the FQDN "localhost." (or rather, that
> could be done, but it never is). What's always returned is the
> unqualified name, "localhost"
>
> | It's not only if the kernel is capable. It's also if there
> | are interfaces configured.
>
> Sure, there should be a simple API for "should I attempt this AF?"
>
> I'd also be somewhat cautious about advising people to run non-standard
> DNS setups. There are things that people who know what they're doing
> can safely do - but which aren't really practical to advise for the
> world to attempt to copy (and everyone snarfing the root zone probably
> doesn't scale, even if it isn't one of the more dangerous things).
And constantly querying the root servers does?
It's quite possible to make zone transfers scale. You get
your copy from your ISP. Your ISP gets a copy from the
roots / dedictated "stealth" root server.
Anyway this need to be analysed.
>
> kre
>
--
Mark Andrews, Internet Software Consortium
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: [EMAIL PROTECTED]
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
