did you restarted neutron-server services on network and controller node. Did you try restarting httpd and memcached service?

the physical ethernet connection to controller from neutron may be disconnected or your neutron-server service may be down.

first check wheather you are able to ping between neutron & controller.

then check wheather messaging service is running or not.

I would suggest you look into your services to determine wether or not they are all running as intended. On a controller node you can try the following:

nova service-list
neutron agent-list
rabbitmqctl status

Alternatively, you can also run the following commands to get a response from the OS:

On Ubuntu:

cd /etc/init.d; for i in $(ls nova-*); do service $i status; done
cd /etc/init.d; for i in $(ls neutron-*); do service $i status; done

On CentOS:

cd /etc/init.d; for i in $(ls openstack-nova-*); do service $i status; done
cd /etc/init.d; for i in $(ls openstack-neutron-*); do service $i status; done

If you can post the responses from these commands perhaps we can determine what the problem is.