docker: dnsmasq fails in hostmode to restart

[bebehei]

Attaching to pixieboot_pixieboot-itv_1
pixieboot-itv_1  | dnsmasq's configuration changed. Reloading.
pixieboot-itv_1  | ln: failed to create symbolic link '/etc/nginx/sites-enabled/pixieboot.conf': File exists
pixieboot-itv_1  | nginx's configuration changed. Reloading.
pixieboot-itv_1  |  * Starting DNS forwarder and DHCP server dnsmasq
pixieboot-itv_1  |
pixieboot-itv_1  | dnsmasq: failed to bind DHCP server socket: Address already in use
pixieboot-itv_1  |    ...fail!
pixieboot-itv_1  |  * Starting nginx nginx
pixieboot-itv_1  |    ...done.
pixieboot-itv_1  | tailf: stat of /var/log/dnsmasq.log failed: No such file or directory

In some cases, dnsmasq does not get killed from the previous invocation with the container. The next invocation of the pixieboot container is not able to start properly and fails silently.

It would be good to implement a new strategy in docker.sh to handle failures.