recursion for reverse/in-addr.arpa zones

[Barry Margolin]

In article <ghub48$92r$1 at sf1.isc.org>, "Todd Snyder" <tsnyder at rim.com> 
wrote:

> On our slave, there are no specific declarations for the 10.131.10 zone,
> or even 10.131, just 10.
>  
> On the server we're slaving off of, there would probably be more, but I
> don't know as I'm not in control of that server/servers.

Since your server is a slave, the delegtion records in the 
10.in-addr.arpa zone will be received in the zone transfer.

> Will reverse lookups by default continue to look for more specific
> domains, recursing as necessary?  If so, how far will it go?  I'm

There's nothing special about reverse domains.  All lookups follow 
delegations, going as far as necessary to get the answer.

> slaving an "A" class, and it went and found a "C".  If we'd had the "B"
> declared, would it have stopped there, or kept going?

If the B contains a delegation to a C, it will go there.

>  
> This behaviour seems odd to me, and I've not been able to find
> information about this behaviour in the book(s).

It's just the basic DNS protocol.  If a name is in a delegated 
subdomain, you follow the NS records to get the answer.  Read the 
resolver algorithm description in RFC 1034.

-- 
Barry Margolin, barmar at alum.mit.edu
Arlington, MA
*** PLEASE don't copy me on replies, I'll read them in the group ***