[Dnsmasq-discuss] Listen to two interfaces?

Frederick C. Damen fred at damen.org
Tue Nov 30 18:52:08 GMT 2010


Query: How does dnsmasq associate the interface with the dhcp-range?

I do not detect an explicit connection, i.e. tags shared between config 
entries.  I assume the implicit association is that the IP address 
assigned to the interface outside of dnsmasq, e.g. ifconfig, needs to be 
within the address range within dhcp-range assigned within dnsmasq. (and 
dnsmasq marks the interfaces IP address as used).  What happens to the 
dhcp-range that does not include one of the interfaces ip addresses?

Fred

On 11/30/2010 11:50 AM, /dev/rob0 wrote:
> On Tue, Nov 30, 2010 at 05:12:40PM +0100, Markus Bergholz wrote:
>> i want
>> dhcp-range=192.168.1.2,192.168.1.1,255.255.255.0,12h
>> for both. but everytime i try to bridge, the hole network on server
>> side don't work anymore.
>> so i want to try without bridge.
>
> Fine, just use separate netblocks for each physical segment. No need
> for bridging.
>
>> could
>> interface=eth0
>> dhcp-range=192.168.1.2,192.168.1.1,255.255.0.0,12h
>> interface=wlan0
>> dhcp-range=192.168.2.2,192.168.2.1,255.255.0.0,12h
>> work?
>
> You have the end-addr argument numerically less than the start-addr.
> This is supposed to be a RANGE of addresses. Also, you have set each
> in a /16 (255.255.0.0 netmask), so those netblocks are overlapping.
>
> You might need to read up a bit on IP networking basics: addresses,
> netmasks/CIDR and subnetting. This might work for what you want:
>
> # eth0 being 192.168.4.1/24
> interface=eth0
> dhcp-range=192.168.4.128,192.168.4.191,255.255.255.0,12h
> # wlan0 being 192.168.5.1/24
> interface=wlan0
> dhcp-range=192.168.5.64,192.168.5.127,255.255.255.0,12h
>
> In that example I use addresses on subnetting boundaries rather than
> decimal numbering boundaries. The eth0 range could be described in
> CIDR terms as 192.168.4.128/26, and wlan0 range is 192.168.5.64/26.
> For NAT purposes, 192.168.4.0/23 covers them both.



More information about the Dnsmasq-discuss mailing list