Re: [mif] [dhcwg] [DNSOP] [dnsext] 2nd Last Call for MIF DNS server selection document

[Danny Mayer <[e-mail]>]

On 10/23/2011 7:49 PM, Mark Andrews wrote:
> 
> In message <[e-mail]>, Matthew Pounse
> tt writes:
>>
>> On 2011/10/22, at 15:21, Keith Moore wrote:
>>
>>>
>>> On Oct 22, 2011, at 2:42 PM, Doug Barton wrote:
>>>
>>>> 1. I think we're all in agreement that dot-terminated names (e.g.,
>>>> example.) should not be subject to search lists. I personally don't have
>>>> any problems with any document mentioning that this is the expected
>>>> behavior.
>>>
>>> agree.  however there are standard protocols for which a trailing dot in a 
>> domain name is a syntax error.
>>
>> Any protocol that makes a standard FQDN a syntax error is itself in error.  N
>> ot to say that these don't exist, but if people are writing protocols that ca
>> n't deal with a properly formatted FQDN they need to stop.  Now.
> 
> Except it isn't a standard hostname.  Periods *seperate* labels in
> hostnames RFC 952.  They DO NOT appear at the end of hostnames.
> 
> Appending a period to the end of a name is user interface hack to
> prevent searching.  If is also a way to prevent the appending of
> the current origin to all names in a DNS master file as the current
> origin is always appended if it isn't done.
> 
> In addition single labels are not HEIRACHICAL / DOMAIN STYLE names
> as envisioned when we went from a flat namespace of simple hostnames
> to a heirarchical namespace.
> 
> 	foo.bar is a heirachical hostname.
> 	bar is a simple hostname.
> 
> Why are we trying to bring them back on a global context?

I'm not even sure why we are having this discussion. You are right, of
course.

DNS doesn't know anything about search lists, nor does it care. Any
valid string presented to DNS is by definition a FQDN and DNS will
proceed accordingly. If the application choosing to consult a domain
search list it's up to that application, but DNS is not involved in the
consultation.

Danny