reverse lookup to CNAME

Matus UHLAR - fantomas uhlar at fantomas.sk
Tue Jan 27 12:29:02 UTC 2009


> > On 23.01.09 23:06, Barry Margolin wrote:
> > > Why don't you just use normal reverse DNS:
> > >
> > > zone for 1.1.1.in-addr.arpa
> > >
> > > 1 IN PTR metis.local.
> > >   IN PTR bob-www-sol-l01.local.
> > 
> > accorging to the above, metis.local is a CNAME, so the 
> > reverse should point to bob-www-sol-l01.local. - pointing it 
> > to metis.local. would be incorrect.
> > And although two or PTRs usually make no problems, it's 
> > recommended not to do that, because some SWs jsut can't 
> > handle that (even if they would all be correct). So, keep 
> > only PTR to bob-www-sol-l01.local.

On 26.01.09 08:20, Ben Bridges wrote:
> Section 3.3.12 of RFC 1035 (PTR RDATA format) appears to say that no
> constraints are placed upon a domain-name that is the RDATA of a PTR
> record, that it is simple data and no special meaning or further
> processing is associated with it (hence implying that it is ok if that
> domain-name is defined as a CNAME somewhere else in the domain space).
> Is that not the case?  Is there some other part of the DNS specification
> that forbids it?

RFC 2181 sect. 10.2. mentions that 

"...the relevant section of RFC1034 (section 3.6.2) indicates that the value of
a PTR record should be a canonical name."

"Note that while the value of a PTR record must not be an alias"...

I know about SWs that reject the PTR if it points to CNAME.
-- 
Matus UHLAR - fantomas, uhlar at fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
WinError #99999: Out of error messages.



More information about the bind-users mailing list