Putting iPads in their own IP pool?

[Miroslav Lauš]

On Oct 8, 2010, at 10:18 PM, Bob Proulx wrote:

> Miroslav Lauš wrote:
>> Unfortunetly, iPads does not identify itself using dhcp option 60:
>> betelgeuse:~ mirek$ sudo tcpdump -vvvv -s0 -ni en1 port 67 or port 68
>> [...good dump information...]
> 
> Hmm...  That is unfortunate.  But it does look like they send the
> hostname by defailt.
> 
>> 	    Hostname Option 12, length 9: "beetlepad"
> 
> And here I see in the logs lines such as:
> 
>  Oct  8 14:02:23 fc dhcpd: DHCPREQUEST for 172.17.31.59 from c8:bc:c8:2e:49:46 (Elijahs-iPad) via eth0
> 
> And the dhcpd.leases file shows lines:
>  client-hostname "Laura-iPad";
> 
> By convention here the hostnames contain the string "iPad" in them.
> That is good enough as a 99% solution.  Again, a false positive is
> better than a false negative.  I agree that chasing ethernet addresses
> is not wonderful.  Using the client supplied hostname seems like a
> good way to go.
> 
> Looking through the docs to try to determine how to match on the
> hostname...
> 
> ?? class "ipad" {
> ??   match option client-hostname;
> ?? }
> ?? subclass "ipad" "iPad";
> 

you can try with that:

class "ipad" {
  match suffix(option host-name,4)
}
subclass "ipad" "iPad";

or without subclass in this case:

class "ipad" {
  match if suffix(option host-name,4) = "iPad"
}

-m

> Bob
> _______________________________________________
> dhcp-users mailing list
> dhcp-users at lists.isc.org
> https://lists.isc.org/mailman/listinfo/dhcp-users