Folks, All my deployment is running on OVN and things are working great with octavia and its special lb-mgmt-net working. Recently I am deploying a new cloud using the OVS network plugin and found dhcp-agent namespace not pinging because I have a similar IP address configured on the host interface. My LBaaS nic /etc/netplan/00-network.yml ... ... # OS1_LBAAS bond0.15: id: 15 link: bond0 dhcp4: false dhcp6: false mtu: 1500 addresses: [ 192.168.32.201/24 ] Controller node interface # ifconfig bond0.15 bond0.15: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet 192.168.32.201 netmask 255.255.255.0 broadcast 192.168.32.255 inet6 fe80::b489:8ff:fe61:2298 prefixlen 64 scopeid 0x20<link> ether b6:89:08:61:22:98 txqueuelen 1000 (Ethernet) RX packets 1074073 bytes 54067608 (54.0 MB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 1594 bytes 78172 (78.1 KB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 octavia DHCP agent namespace (lb-mgmt-net) # ip netns exec qdhcp-1008bdd3-372a-44af-8fd3-00fb99b8e489 ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 13: tape6186f5a-40: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000 link/ether fa:16:3e:c7:bf:2d brd ff:ff:ff:ff:ff:ff inet 192.168.32.10/24 brd 192.168.32.255 scope global tape6186f5a-40 valid_lft forever preferred_lft forever inet 169.254.169.254/32 brd 169.254.169.254 scope global tape6186f5a-40 valid_lft forever preferred_lft forever My global.yml ################### # Octavia options ################### enable_octavia: "yes" octavia_network_interface: "bond0.15" octavia_amp_flavor: name: "amphora" is_public: no vcpus: 2 ram: 2048 disk: 5 octavia_amp_network: name: lb-mgmt-net provider_network_type: vlan provider_segmentation_id: 15 provider_physical_network: physnet1 external: false shared: false subnet: name: lb-mgmt-subnet cidr: "192.168.32.0/24" allocation_pool_start: "192.168.32.10" allocation_pool_end: "192.168.32.100" no_gateway_ip: yes enable_dhcp: yes When I ping from namespace to controller machine LBaaS interface I am not able to ping # ip netns exec qdhcp-1008bdd3-372a-44af-8fd3-00fb99b8e489 ping 192.168.32.201 PING 192.168.32.201 (192.168.32.201) 56(84) bytes of data. ^C --- 192.168.32.201 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 2039ms Seems like standard linux arp/kernel issue. How does it work? In OVN there is no namespace so everything works fine but in OVS because of namespace it doesn't work.