On Thu, 2009-04-02 at 15:51 +1030, Tim wrote:
On Thu, 2009-04-02 at 09:27 +1100, Simon Slater wrote:
> After a reboot I got the message:
>
> dnsmasq failed to bind DHCP server socket: address already in use.
Which may be that something *else* is using it, or that dnsmasq is
starting up in a way that causes problems with itself. I've seen that
sort of thing when there's a network with IPv4 and IPv6 addressing in
use, and a service starts up. It listens to both, and while starting up
complains that the port's already in use while trying to listen to the
second interface (which is really the first, with the IPv6 type of
addresses, as well as IPv4).
> In the startup sequence as it scrolls on the screen, dnsmasq is near
> the end, but is running:
> [root@dell ~]# service dnsmasq status
> dnsmasq (pid 2613) is running...
>
> So is the problem with the configuration or at startup? How
> do I find what is conflicting for the socket address?
man netstat
e.g. netstat -antuevp
Thanks Tim,
Now this shows that dnsmasq is using 192.168.122.1:53 with tcp and udp.
This is the link local address on a port for dns. Also udp on
0.0.0.0:67 which is one of the dhcp ports but for all networks?
Does dnsmasq need to use 192.168.122.1?
The first aim is to get dhcp going. Would 0.0.0.0:67 help or get in the
way?
--
Regards,
Simon Slater
Registered Linux User #463789. Be counted at:
http://counter.li.org/