CNAME problem

Nate Campi nate at wired.com
Mon Dec 31 05:49:13 UTC 2001


On Mon, Dec 31, 2001 at 04:11:59PM +1100, Mark_Andrews at isc.org wrote:
> 
> > Before I categorically tell my client that 10001.org could not be aliased, 
> > please someone confirm or contradict and if you would, state your sources. 
> > I studied ORA's "DNS and BIND" in detail but could not find reference that 
> > would help with this. If it can be done, please tell me how.
> 
> 	It's always been illegal.
> 
> RFC 1034:
> "The domain system provides such a feature using the canonical name
> (CNAME) RR.  A CNAME RR identifies its owner name as an alias, and
> specifies the corresponding canonical name in the RDATA section of the
> RR.  If a CNAME RR is present at a node, no other data should be
> present; this ensures that the data for a canonical name and its aliases
> cannot be different.  This rule also insures that a cached CNAME can be
> used without checking with an authoritative server for other RR types."
> 
> 	The SOA, NS and MX records constitute "other data".

If you think about it, it simply doesn't make sense to alias a domain.
When you query for the SOA record, what do you expect to get, the SOA
for the alias or for the canonical name? You would get the canonical 
name, which is clearly not what you want. BIND won't even load a zone
with an error like that.

OBTW, "DNS & BIND", 4th ed, page 498. It's the first thing in those 
three or four pages on CNAMEs that is covered, quite clearly. The book 
does mention DNAME records with BIND 9, you might want to look into that.
I don't use BIND 9 so I can't comment on it.
-- 
Nate Campi | Terra Lycos DNS | WiReD UNIX Operations

The porcupine with the sharpest quills gets stuck on a tree more often.



More information about the bind-users mailing list