Ok, right, didn't consider that (as I'm writing only a nameserver :-). Nevertheless, I would say that it's not eligible to consider root to be the default origin. I personally would reject zone which has no $ORIGIN directive at the beginning, nor a FQDN as the owner of the SOA record. But I suppose this is one of the points where the standards get blurry and any interpretation is right as long as one documents it properly.

LS

On 05-Jul-12 12:44, Miek Gieben wrote:
[ Quoting <lubos.slovak@nic.cz> in "Re: [knot-dns-users] origin doesn't..." ]
  
Hi Miek,

as I found out, according to RFC 1035: 'A free standing @ is used to denote the
current origin.' and from various other references to 'origin' it seems that
the default origin should be the zone name. This also corresponds with Bind
behaviour (see their manual: http://ftp.isc.org/isc/bind9/cur/9.9/doc/arm/
Bv9ARM.ch06.html#id2598969).
    
hello,

Yes, you are right. But even before you start parsing a stand-alone zonefile, you
need to to start somewhere.... If no $ORIGIN is defined and you have no clue on
which zone you are about to read, what $ORIGIN do you append to the ownername
of SOA record if that was unqualified?

(Again, if you are writing a nameserver, this case doesn't apply, because you
*always* now what zone you are reading, but for a stand alone library it is
important, see: https://github.com/miekg/dns/blob/master/zscan.go#L162)

Kind regards,
Miek Gieben
  
_______________________________________________ knot-dns-users mailing list knot-dns-users@lists.nic.cz https://lists.nic.cz/cgi-bin/mailman/listinfo/knot-dns-users


-- 
 Ľuboš Slovák                       Knot DNS
 CZ.NIC Labs          http://www.knot-dns.cz
 -------------------------------------------
 Americká 23, 120 00 Praha 2, Czech Republic

 Email: lubos.slovak@nic.cz
 WWW: http://labs.nic.cz   http://www.nic.cz
 ------------------------------------------- 

 Please consider the environment before printing this email.
 Join the campaign at http://thinkBeforePrinting.org