Re: [dnsext] we need help to make names the same, was draft-yao-dnsext-identical-resolution-02 comment

[Mark Andrews <[e-mail]>]

In message <[e-mail]>, "John R. Levine" wr
ites:
> > The number of these domain names varies from two (or four if a final sigma
> > is present) to some tens of domains, not millions and certainly not
> > billions.
> 
> Well, yes, but that's in one component.  If someone has a name A.B.C.gr, 
> where each of A B C has ten variants, now there's a thousand equivalent 
> names.  I would think that a design that handled the variants in C but not 
> in A and B is not worth implementing.
> 
> If you only care about C, you can do that with bundling, have the registry 
> assign all variants of a requested name to the registrant, and let the 
> registrant deal with making them the same, by copies of the zone.  It's 
> not particularly elegant, but it requires no changes to existing protocol 
> and the software is not very complicated.
> 
> > We asked for a DNS rr that will allow two, three, four, or tens of 
> > chosen domain names to act as if they were interchangeable in all the 
> > branches of the domain name tree, starting from the top. They could be 
> > perfectly normal Latin domain names that point to the same services and 
> > the administrator decides to administer them as one.
> 
> Except that, as often noted, if the servers they point to don't have some 
> way to discover that the names are intended to be equivalent, the services 
> won't work.  The assumptions so far seem to be either that the services 
> are something like telnet that doesn't care what its name is, or that 
> they'll be manually provisioned, which seems unlikely to work except in 
> the tiniest cases.

This is where if CNAME's were being properly handled you wouldn't have a
issue.  Give the service the indirection support and aliasing support.
HTTP uses the alias record for indirection which is just wrong.
 
> Regards,
> John Levine, [e-mail], Primary Perpetrator of "The Internet for Dummies
> ",
> Please consider the environment before reading this e-mail. http://jl.ly
> _______________________________________________
> dnsext mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/dnsext
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: [phone]                 INTERNET: [e-mail]