I've deployed openstack yoga using kolla ansible, in three controller nodes, a storage cluster and multiple compute nodes. 
The deployment included using haproxy, keepalived and hacluster.
everything seems to work fine, I can launch instances and networking is doing great, however I found out a problem that I didn't get in all the deployment tests that i've run before.
I've combien openstack networks under the same network 10.10.10.0/24 and used the same ip address for both : kolla_internal_vip_address and kolla_external_vip_interface which is 10.10.10.254, and used an ip route to access to the openstack cluster from an external pc 
ip route add 10.10.10.0/24 via 192.168.129.29 dev eth0 
where 192.168.129.29 is the address of one of my controllers, however when I try to login to the horizon dashboard using 10.10.10.254 i don't get a response from the web page, and when I log to the dashboard using the ip address of one of my controller nodes, I cannot use the instance console since it is under the ip address of 10.10.10.254 which is assigned to both kolla_internal_vip_address and kolla_external_vip_interface.
although i can ping 10.10.10.254 from the external network using ip route.