Wildcards in reverse DNS

[Sten Carlsen]

Karl Auer wrote:
> On Sun, 2007-01-07 at 00:15 +0100, Sten Carlsen wrote:
>   
>>> Hm. So having several million trillion addresses PER SUBNET (2^64) still
>>> isn't enough for you?
>>>
>>>       
>> YES, if they were available. Since they are used for the MAC address
>> reall life will provide only ONE out of those 2^64.
>>     
>
> ?!? You can put as many addresses as you want on one interface - the MAC
> address stuff is only for stateless autoconfiguration. Much more
> importantly, you can have 2^64 different hosts (or 2^64 different
> interfaces) in a single subnet - more if you extend your host part beyon
> 64 bits. That is more addresses than there are MAC addresses - more IP
> addresses than there are *possible* MAC addresses...
>
> Perhaps I misunderstand your point. What exactly is the problem as you
> see it?
>   
I agree, I was too quick.
>   
>> MAC addresses are supposed to be unique so only one out of the possible
>> addresses will be used; except if privacy addresses are being taken into
>> account.
>>     
>
> There is no problem at all with you and me sharing a MAC address as long
> as we have a router between us...
>   
As long as MAC-addresses are unique we will not share them. Even if I
have 1000 devices I "waste" 50+ bits; I guess that this is just my
natural tendency to not overdo field lengths and not transmit too many
bits, but also not too few. Network bandwidth is becoming cheaper so it
matters less.

My main concern will still be privacy. Let that rest for now, it will
not be any different whatever I say, I will have to learn to work with
it. :-)
> Regards, K.
>  
>   

-- 
Best regards

Sten Carlsen

No improvements come from shouting:

       "MALE BOVINE MANURE!!!"