Top Level Domains

Kevin Darcy kcd at daimlerchrysler.com
Mon Jun 7 21:29:05 UTC 2004


Gnos Theos wrote:

>phil-news-nospam at ipal.net wrote in message news:<c9le8p$7e1$1 at sf1.isc.org>...
>  
>
>>On Tue, 01 Jun 2004 23:37:39 -0400 Kevin Darcy <kcd at daimlerchrysler.com> wrote:
>>
>>| Gnos Theos wrote:
>>| 
>>|>Hello,
>>|>
>>|>Does anyone know of a list of TLDs that are 
>>|>two levels, e.g. .co.uk, or .co.jp ?
>>|>
>>| Um, such a thing cannot exist by definition, since a TLD is a 
>>| *top*-level domain. .co.uk and .co.jp are SLDs (second-level domains) 
>>| under the .uk and .jp TLDs, respectively.
>>| 
>>| Perhaps you mean "a list of SLDs under which domain registration is 
>>| possible" (???) I'm not aware of any such comprehensive list, but at 
>>| least with the right terminology you might have more luck finding one....
>>| 
>>|                                                                         
>>|                                                   - Kevin
>>| 
>>| P.S. The .us domain accepts registrations under <state>.us, but only 
>>| from state agencies or departments, so I'm not sure whether you'd want 
>>| to add those 50 SLDs to your list or not...
>>
>>Possibly the reason for this request is to implement a means identify a
>>common authority level.  That is, to strip down a host name to just the
>>registered name level, and use that for comparison.  The idea I have for
>>it is to check for proper sender host where SPF data is absent.  Since
>>you cannot assume an outbound mail server (SMTP client) is listed in the
>>MX records for the sender's RHS domain, that test isn't reliable.  However,
>>striping the MX names down to registration authority level, and comparing
>>that to the similarly stripped down SMTP client reverse name, could lead
>>to a sufficiently reliable level as to be practical to deploy (it would
>>not be used in cases where SPF data is present).
>>
>>What I describe is my own interest in that kind of data.  I really don't
>>know what the OP wants it for.
>>    
>>
>
>
>What I am trying to do is to determine the based domain from a host name. 
>suppose you have the name of a web server. Some people put the domain
>name into a DNS record so that it works as a host name. For example
>
>http://google.com will work as well as http://www.google.com
>
>[user at squid user]$ nslookup google.com
>Server:  SQUID.SUNPERF.COM
>Address:  10.78.109.4
>
>Non-authoritative answer:
>Name:    google.com
>Addresses:  216.239.39.99, 216.239.57.99, 216.239.37.99
>
>[user at squid user]$ nslookup www.google.com
>Server:  SQUID.SUNPERF.COM
>Address:  10.78.109.4
>
>Non-authoritative answer:
>Name:    www.google.akadns.net
>Addresses:  216.239.57.147, 216.239.57.99, 216.239.57.104
>Aliases:  www.google.com
>
>Here is a non-comprehensive list of google domains: 
>
>..google.co.cr
>..google.co.hu
>..google.co.il
>..google.co.in
>..google.co.je
>..google.co.jp
>..google.co.kr
>..google.co.ls
>..google.co.nz
>..google.co.th
>..google.co.uk
>..google.co.ve
>
>For all of these you need 3 sections, a tld, plus two "sub-domains", 
>whereas you only need a tld plus one "sub-domain" for google.com. The
>semantics of the problem are difficult. I am trying to determine what
>part of a hostname is a domain assigned to a specific "person". For example
>..co.jp is not assigned per se to any one individual, but google.co.jp
>is assigned to a "person" (Note: a corporation is treated as a person
>in some legal instances). Nonetheless they could create sd2.sd1.google.co..jp,
>and I would still only be interested in .google.co.jp for my purposes.
>
>I am trying to deal with the problem that you can have
>
>domain.jp
>host.domain.jp
>host.domain.co.jp
>host.sub-domain.domain.co.jp
>host.domain.ne.jp
>host.domain.or.jp
>
>and only return domain.jp, domain.co.jp, domain.ne.jp, or domain.or.jp
>respectively. 
>
>I can ad hoc deal with problems as they arrise, but that is always
>painful and fraught with error. 
>
I'm not sure why you consider this a "problem". host.domain.jp and 
host.domain.co.jp, for example, are logically different from each other, 
and from all of the other names on that list. The fact that they may 
point to the same server, server farm, load-balancing device, web 
content, or whatever, is basically just a happy coincidence. How is this 
a "problem"?

                                                                         
                                          - Kevin




More information about the bind-users mailing list