Install the packages: # apt-get install neutron-plugin-openvswitch-agent \ neutron-dhcp-agent neutron-l3-agent # yum install openstack-neutron openstack-neutron-openvswitch\ openvswitch-switch # zypper install openstack-neutron openstack-neutron-openvswitch-agent \ openvswitch-switch In Fedora, the Open vSwitch kernel module is compiled without tunnel support. If you need gre tunnels and network namespaces, you must recompile this package from source. For information, see the Open vSwitch site. If you recompile the kernel module, you must remove and delete the existing one. Start Open vSwitch: # service openvswitch-switch start # chkconfig openvswitch-switch on # service openvswitch-switch start # systemctl start openvswitch-switch.service # systemctl enable openvswitch-switch.service Create an internal and external network bridge. The introduction in this guide describes the purposes of these bridges. # ovs-vsctl add-br br-ex # ovs-vsctl add-port br-ex eth1 # ovs-vsctl add-br br-int To configure the bridges, complete the following steps: Change the eth1 entry in /etc/network/interfaces, as follows: auto eth1 iface eth1 inet manual up ip address add 0/0 dev $IFACE up ip link set $IFACE up down ip link set $IFACE down Add br-ex to /etc/network/interfaces, as follows: auto br-ex iface br-ex inet static address 10.0.0.9 netmask 255.255.255.0 gateway 10.0.0.1 Edit /etc/sysconf/network-scripts/ifcfg-eth1, as follows: # External DEVICE=eth1 TYPE=Ethernet BOOTPROTO=none NM_CONTROLLED=no BRIDGE=br-ex ONBOOT=yes Edit /etc/sysconf/network-scripts/ifcfg-br-ex, as follows: #Public Bridge DEVICE=br-ex TYPE=Bridge BOOTPROTO=static IPADDR=10.10.10.9 NETMASK=255.255.255.0 NM_CONTROLLED=no ONBOOT=yes Remove the IP address from eth1 add it to br-ex, as follows: # ip addr del 10.0.0.9/24 dev eth1 # ip addr add 10.0.0.9/24 dev br-ex Restart networking, as follows: # service networking restart Enable a simple NAT service so that Compute nodes can access the Internet through the Cloud Controller: # iptables -A FORWARD -i eth0 -o br-ex -s 10.10.10.0/24 -m conntrack \ --ctstate NEW -j ACCEPT # iptables -A FORWARD -m conntrack --ctstate ESTABLISHED,RELATED -j ACCEPT # iptables -A POSTROUTING -s 10.10.10.0/24 -t nat -j MASQUERADE These rules are lost on reboot. To save or restore the rules, use an appropriate tool from your distribution to save and restore them.

Edit /etc/neutron/neutron.conf, as follows: [DEFAULT] verbose = True rabbit_password = password rabbit_host = 10.10.10.10 [keystone_authtoken] auth_host = 10.10.10.10 admin_tenant_name = service admin_user = neutron admin_password = password[DEFAULT] verbose = True rpc_backend = cinder.openstack.common.rpc.impl_qpid sql_connection = mysql://cinder:password@localhost/cinder qpid_user = guest qpid_password = quest [keystone_authtoken] auth_host = 10.10.10.10 admin_tenant_name = service admin_user = neutron admin_password = password[DEFAULT] verbose = True rabbit_password = password rabbit_host = 10.10.10.10 [keystone_authtoken] auth_host = 10.10.10.10 admin_tenant_name = service admin_user = neutron admin_password = password Edit /etc/neutron/plugins/openvswitch/ovs_neutron_plugin.ini, as follows: [database] connection = mysql://neutron:password@10.10.10.10/neutron [ovs] tenant_network_type = gre tunnel_id_ranges = 1:1000 enable_tunneling = True local_ip = 10.10.10.9 [securitygroup] firewall_driver = \ neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver Choose tunnel mode because you do not have to configure your physical switches for VLANs. Edit /etc/neutron/dhcp_agent.ini, as follows: [DEFAULT] enable_isolated_metadata = True enable_metadata_network = True Edit /etc/neutron/metadata_agent.ini, as follows: [DEFAULT] auth_url = http://10.10.10.10:35357/v2.0 auth_region = RegionOne admin_tenant_name = service admin_user = neutron admin_password = password nova_metadata_ip = 10.10.10.10 metadata_proxy_shared_secret = password Start the services, as follows: # service neutron-plugin-openvswitch-agent start # service neutron-dhcp-agent restart # service neutron-metadata-agent restart # service neutron-l3-agent restart # service neutron-server restart # service neutron-openvswitch-agent restart # service neutron-dhcp-agent restart # service neutron-l3-agent restart # chkconfig neutron-server on # chkconfig neutron-openvswitch-agent on # chkconfig neutron-dhcp-agent on # chkconfig neutron-l3-agent on # systemctl restart neutron-server # systemctl restart neutron-openvswitch-agent # systemctl restart neutron-dhcp-agent # systemctl restart neutron-l3-agent # systemctl enable neutron-server # systemctl enable neutron-openvswitch-agent # systemctl enable neutron-dhcp-agent # systemctl enable neutron-l3-agent To debug errors that prevent the Networking Service from starting, review the /var/log/neutron/*.log files.