Wildcards in reverse DNS
ccc2716 at vip.cybercity.dk
Sat Jan 6 22:08:04 UTC 2007
Marc Haber wrote:
> On Sat, Jan 06, 2007 at 09:55:25PM +0100, Sten Carlsen wrote:
>> As for 128bit adressing; as far as I can see (which may not be far
>> enough) the lower 64bits are meant for the MAC-address, the rest is
>> useful as addresses.
> It is common to use the MAC address, but one can use arbitrary numbers
> as well. Which comes in handy for an actual service which should not
> have its IP change when it moves to different hardware.
That was one of my points, I don't see the idea of changing all DNS
entries just because oone NIC burnt out and was replaced. Good that we
can avoid that.
>> What is the reason for this use of the lower 64 bits? Can you get rid of
>> ARP? Is this so much better than the self-assigned IPv4 addresses in use
>> today? On top of this, there are now attempts of hiding the MAC-address
>> to help privacy, which would otherwise be gone.
> It allows autoconfiguration. And why does one want to hide the MAC
> address? Knowing it does only help an attacker on the local network,
> and who is on the local network knows it anyway.
I for one don't need any manufacturer knowing what I use my purchase for
and where. For almost all traditional applications I have no big issue.
For some of those to come, I am not so sure; I can imagine a host of
"services" I will not want, all based on the manufacturer tracking his
For autoconfigurations, yes, it is nice to be able to do that with no
effort other than copying the MAC to the IP. Is that worth 64bits to be
transmitted everywhere? I don't see any other value in those bits; at
least if the world is autoconfigured. If that is not the case we still
have two classes of addresses: public and private; the low 64 bits still
do little for addressing as I see it. Maybe I still need more details?
No improvements come from shouting:
"MALE BOVINE MANURE!!!"
More information about the bind-users