Advanced features through API extensions

Several plug-ins implement API extensions that provide capabilities similar to what was available in nova-network. These plug-ins are likely to be of interest to the OpenStack community.

Provider networks

Networks can be categorized as either project networks or provider networks. Project networks are created by normal users and details about how they are physically realized are hidden from those users. Provider networks are created with administrative credentials, specifying the details of how the network is physically realized, usually to match some existing network in the data center.

Provider networks enable administrators to create networks that map directly to the physical networks in the data center. This is commonly used to give projects direct access to a public network that can be used to reach the Internet. It might also be used to integrate with VLANs in the network that already have a defined meaning (for example, enable a VM from the marketing department to be placed on the same VLAN as bare-metal marketing hosts in the same data center).

The provider extension allows administrators to explicitly manage the relationship between Networking virtual networks and underlying physical mechanisms such as VLANs and tunnels. When this extension is supported, Networking client users with administrative privileges see additional provider attributes on all virtual networks and are able to specify these attributes in order to create provider networks.

The provider extension is supported by the Open vSwitch and Linux Bridge plug-ins. Configuration of these plug-ins requires familiarity with this extension.


A number of terms are used in the provider extension and in the configuration of plug-ins supporting the provider extension:

Provider extension terminology



virtual network

A Networking L2 network (identified by a UUID and optional name) whose ports can be attached as vNICs to Compute instances and to various Networking agents. The Open vSwitch and Linux Bridge plug-ins each support several different mechanisms to realize virtual networks.

physical network

A network connecting virtualization hosts (such as compute nodes) with each other and with other network resources. Each physical network might support multiple virtual networks. The provider extension and the plug-in configurations identify physical networks using simple string names.

project network

A virtual network that a project or an administrator creates. The physical details of the network are not exposed to the project.

provider network

A virtual network administratively created to map to a specific network in the data center, typically to enable direct access to non-OpenStack resources on that network. Project can be given access to provider networks.

VLAN network

A virtual network implemented as packets on a specific physical network containing IEEE 802.1Q headers with a specific VID field value. VLAN networks sharing the same physical network are isolated from each other at L2 and can even have overlapping IP address spaces. Each distinct physical network supporting VLAN networks is treated as a separate VLAN trunk, with a distinct space of VID values. Valid VID values are 1 through 4094.

flat network

A virtual network implemented as packets on a specific physical network containing no IEEE 802.1Q header. Each physical network can realize at most one flat network.

local network

A virtual network that allows communication within each host, but not across a network. Local networks are intended mainly for single-node test scenarios, but can have other uses.

GRE network

A virtual network implemented as network packets encapsulated using GRE. GRE networks are also referred to as tunnels. GRE tunnel packets are routed by the IP routing table for the host, so GRE networks are not associated by Networking with specific physical networks.

Virtual Extensible LAN (VXLAN) network

VXLAN is a proposed encapsulation protocol for running an overlay network on existing Layer 3 infrastructure. An overlay network is a virtual network that is built on top of existing network Layer 2 and Layer 3 technologies to support elastic compute architectures.

The ML2, Open vSwitch, and Linux Bridge plug-ins support VLAN networks, flat networks, and local networks. Only the ML2 and Open vSwitch plug-ins currently support GRE and VXLAN networks, provided that the required features exist in the hosts Linux kernel, Open vSwitch, and iproute2 packages.

Provider attributes

The provider extension extends the Networking network resource with these attributes:

Provider network attributes

Attribute name


Default Value


provider: network_type



The physical mechanism by which the virtual network is implemented. Possible values are flat, vlan, local, gre, and vxlan, corresponding to flat networks, VLAN networks, local networks, GRE networks, and VXLAN networks as defined above. All types of provider networks can be created by administrators, while project networks can be implemented as vlan, gre, vxlan, or local network types depending on plug-in configuration.

provider: physical_network


If a physical network named “default” has been configured and if provider:network_type is flat or vlan, then “default” is used.

The name of the physical network over which the virtual network is implemented for flat and VLAN networks. Not applicable to the local, vxlan or gre network types.




For VLAN networks, the VLAN VID on the physical network that realizes the virtual network. Valid VLAN VIDs are 1 through 4094. For GRE networks, the tunnel ID. Valid tunnel IDs are any 32 bit unsigned integer. Not applicable to the flat or local network types.

To view or set provider extended attributes, a client must be authorized for the extension:provider_network:view and extension:provider_network:set actions in the Networking policy configuration. The default Networking configuration authorizes both actions for users with the admin role. An authorized client or an administrative user can view and set the provider extended attributes through Networking API calls. See the section called Authentication and authorization for details on policy configuration.

L3 routing and NAT

The Networking API provides abstract L2 network segments that are decoupled from the technology used to implement the L2 network. Networking includes an API extension that provides abstract L3 routers that API users can dynamically provision and configure. These Networking routers can connect multiple L2 Networking networks and can also provide a gateway that connects one or more private L2 networks to a shared external network. For example, a public network for access to the Internet. See the OpenStack Configuration Reference for details on common models of deploying Networking L3 routers.

The L3 router provides basic NAT capabilities on gateway ports that uplink the router to external networks. This router SNATs all traffic by default and supports floating IPs, which creates a static one-to-one mapping from a public IP on the external network to a private IP on one of the other subnets attached to the router. This allows a project to selectively expose VMs on private networks to other hosts on the external network (and often to all hosts on the Internet). You can allocate and map floating IPs from one port to another, as needed.

Basic L3 operations

External networks are visible to all users. However, the default policy settings enable only administrative users to create, update, and delete external networks.

This table shows example openstack commands that enable you to complete basic L3 operations:

Basic L3 Operations



Creates external networks.

$ openstack network create public --external
$ openstack subnet create --network public --subnet-range subnetname

Lists external networks.

$ openstack network list --external

Creates an internal-only router that connects to multiple L2 networks privately.

$ openstack network create net1
$ openstack subnet create --network net1 --subnet-range subnetname1
$ openstack network create net2
$ openstack subnet create --network net2 --subnet-range subnetname2
$ openstack router create router1
$ openstack router add subnet router1 subnetname1
$ openstack router add subnet router1 subnetname2

An internal router port can have only one IPv4 subnet and multiple IPv6 subnets that belong to the same network ID. When you call router-interface-add with an IPv6 subnet, this operation adds the interface to an existing internal port with the same network ID. If a port with the same network ID does not exist, a new port is created.

Connects a router to an external network, which enables that router to act as a NAT gateway for external connectivity.

$ openstack router set --external-gateway EXT_NET_ID router1
$ openstack router set --route destination=,gateway= router1

The router obtains an interface with the gateway_ip address of the subnet and this interface is attached to a port on the L2 Networking network associated with the subnet. The router also gets a gateway interface to the specified external network. This provides SNAT connectivity to the external network as well as support for floating IPs allocated on that external networks. Commonly an external network maps to a network in the provider.

Lists routers.

$ openstack router list

Shows information for a specified router.

$ openstack router show ROUTER_ID

Shows all internal interfaces for a router.

$ openstack port list --router  ROUTER_ID
$ openstack port list --router  ROUTER_NAME

Identifies the PORT_ID that represents the VM NIC to which the floating IP should map.

$ openstack port list -c ID -c "Fixed IP Addresses" --server INSTANCE_ID

This port must be on a Networking subnet that is attached to a router uplinked to the external network used to create the floating IP. Conceptually, this is because the router must be able to perform the Destination NAT (DNAT) rewriting of packets from the floating IP address (chosen from a subnet on the external network) to the internal fixed IP (chosen from a private subnet that is behind the router).

Creates a floating IP address and associates it with a port.

$ openstack floating ip create EXT_NET_ID
$ openstack floating ip add port FLOATING_IP_ID --port-id INTERNAL_VM_PORT_ID

Creates a floating IP on a specific subnet in the external network.

$ openstack floating ip create EXT_NET_ID --subnet SUBNET_ID

If there are multiple subnets in the external network, you can choose a specific subnet based on quality and costs.

Creates a floating IP address and associates it with a port, in a single step.

$ openstack floating ip create --port INTERNAL_VM_PORT_ID EXT_NET_ID

Lists floating IPs

$ openstack floating ip list

Finds floating IP for a specified VM port.

$ openstack floating ip list --port INTERNAL_VM_PORT_ID

Disassociates a floating IP address.

$ openstack floating ip remove port FLOATING_IP_ID

Deletes the floating IP address.

$ openstack floating ip delete FLOATING_IP_ID

Clears the gateway.

$ openstack router unset --external-gateway router1

Removes the interfaces from the router.

$ openstack router remove subnet router1 SUBNET_ID

If this subnet ID is the last subnet on the port, this operation deletes the port itself.

Deletes the router.

$ openstack router delete router1

Security groups

Security groups and security group rules allow administrators and projects to specify the type of traffic and direction (ingress/egress) that is allowed to pass through a port. A security group is a container for security group rules.

When a port is created in Networking it is associated with a security group. If a security group is not specified the port is associated with a ‘default’ security group. By default, this group drops all ingress traffic and allows all egress. Rules can be added to this group in order to change the behavior.

To use the Compute security group APIs or use Compute to orchestrate the creation of ports for instances on specific security groups, you must complete additional configuration. You must configure the /etc/nova/nova.conf file and set the use_neutron=True option on every node that runs nova-compute, nova-conductor and nova-api. After you make this change, restart those nova services to pick up this change. Then, you can use both the Compute and OpenStack Network security group APIs at the same time.


  • To use the Compute security group API with Networking, the Networking plug-in must implement the security group API. The following plug-ins currently implement this: ML2, Open vSwitch, Linux Bridge, NEC, and VMware NSX.

  • You must configure the correct firewall driver in the securitygroup section of the plug-in/agent configuration file. Some plug-ins and agents, such as Linux Bridge Agent and Open vSwitch Agent, use the no-operation driver as the default, which results in non-working security groups.

  • When using the security group API through Compute, security groups are applied to all ports on an instance. The reason for this is that Compute security group APIs are instances based and not port based as Networking.

  • When creating or updating a port with a specified security group, the admin tenant can use the security groups of other tenants.

Basic security group operations

This table shows example neutron commands that enable you to complete basic security group operations:

Basic security group operations



Creates a security group for our web servers.

$ openstack security group create webservers \
 --description "security group for webservers"

Lists security groups.

$ openstack security group list

Creates a security group rule to allow port 80 ingress.

$ openstack security group rule create --ingress \
  --protocol tcp SECURITY_GROUP_UUID

Lists security group rules.

$ openstack security group rule list

Deletes a security group rule.

$ openstack security group rule delete SECURITY_GROUP_RULE_UUID

Deletes a security group.

$ openstack security group delete SECURITY_GROUP_UUID

Creates a port and associates two security groups.

$ openstack port create port1 --security-group SECURITY_GROUP_ID1 \
  --security-group SECURITY_GROUP_ID2 --network NETWORK_ID

Removes security groups from a port.

$ openstack port set --no-security-group PORT_ID

Plug-in specific extensions

Each vendor can choose to implement additional API extensions to the core API. This section describes the extensions for each plug-in.

Big Switch plug-in extensions

This section explains the Big Switch neutron plug-in-specific extension.

Big Switch router rules

Big Switch allows router rules to be added to each project router. These rules can be used to enforce routing policies such as denying traffic between subnets or traffic to external networks. By enforcing these at the router level, network segmentation policies can be enforced across many VMs that have differing security groups.

Router rule attributes

Each project router has a set of router rules associated with it. Each router rule has the attributes in this table. Router rules and their attributes can be set using the neutron router-update command, through the horizon interface or the Networking API.

Big Switch Router rule attributes

Attribute name


Input type




A valid CIDR or one of the keywords ‘any’ or ‘external’

The network that a packet’s source IP must match for the rule to be applied.



A valid CIDR or one of the keywords ‘any’ or ‘external’

The network that a packet’s destination IP must match for the rule to be applied.



‘permit’ or ‘deny’

Determines whether or not the matched packets will allowed to cross the router.



A plus-separated (+) list of next-hop IP addresses. For example,

Overrides the default virtual router used to handle traffic for packets that match the rule.

Order of rule processing

The order of router rules has no effect. Overlapping rules are evaluated using longest prefix matching on the source and destination fields. The source field is matched first so it always takes higher precedence over the destination field. In other words, longest prefix matching is used on the destination field only if there are multiple matching rules with the same source.

Big Switch router rules operations

Router rules are configured with a router update operation in OpenStack Networking. The update overrides any previous rules so all rules must be provided at the same time.

Update a router with rules to permit traffic by default but block traffic from external networks to the subnet:

$ neutron router-update ROUTER_UUID --router_rules type=dict list=true \
  source=any,destination=any,action=permit \

Specify alternate next-hop addresses for a specific subnet:

$ neutron router-update ROUTER_UUID --router_rules type=dict list=true  \
  source=any,destination=any,action=permit \

Block traffic between two subnets while allowing everything else:

$ neutron router-update ROUTER_UUID --router_rules type=dict list=true \
  source=any,destination=any,action=permit \

L3 metering

The L3 metering API extension enables administrators to configure IP ranges and assign a specified label to them to be able to measure traffic that goes through a virtual router.

The L3 metering extension is decoupled from the technology that implements the measurement. Two abstractions have been added: One is the metering label that can contain metering rules. Because a metering label is associated with a project, all virtual routers in this project are associated with this label.

Basic L3 metering operations

Only administrators can manage the L3 metering labels and rules.

This table shows example neutron commands that enable you to complete basic L3 metering operations:

Basic L3 operations



Creates a metering label.

$ openstack network meter label create LABEL1 \
  --description "DESCRIPTION_LABEL1"

Lists metering labels.

$ openstack network meter label list

Shows information for a specified label.

$ openstack network meter label show LABEL_UUID
$ openstack network meter label show LABEL1

Deletes a metering label.

$ openstack network meter label delete LABEL_UUID
$ openstack network meter label delete LABEL1

Creates a metering rule.

$ openstack network meter label rule create LABEL_UUID \
  --remote-ip-prefix CIDR \
  --direction DIRECTION --exclude

For example:

$ openstack network meter label rule create label1 \
  --remote-ip-prefix --direction ingress
$ openstack network meter label rule create label1 \
  --remote-ip-prefix --exclude

Lists metering all label rules.

$ openstack network meter label rule list

Shows information for a specified label rule.

$ openstack network meter label rule show RULE_UUID

Deletes a metering label rule.

$ openstack network meter label rule delete RULE_UUID

Lists the value of created metering label rules.

$ ceilometer sample-list -m SNMP_MEASUREMENT

For example:

$ ceilometer sample-list -m
$ ceilometer sample-list -m
$ ceilometer sample-list -m
$ ceilometer sample-list -m