]> The Neutron project provides virtual networking services among devices that are managed by the OpenStack compute service. The &APIv2; combines the Quantum API v1.1 with some essential Internet Protocol Address Management (IPAM) capabilities from the Melange API. These IPAM capabilities enable you to: Associate IP address blocks and other network configuration settings required by a network device, such as a default gateway and dns-servers settings, with an OpenStack Networking network. Allocate an IP address from a block and associate it with a device that is attached to the network through an OpenStack Networking port. To do this, the &APIv2; introduces the subnet entity. A subnet can represent either an IP version 4 or version 6 address block. Each OpenStack Networking network commonly has one or more subnets. When you create a port on the network, an available fixed IP address is allocated to it from one the designated subnets for each IP version. When you delete the port, the allocated addresses return to the pool of available IPs on the subnet. &APIv2; users can choose a specific IP address from the block or let OpenStack Networking choose the first available IP address. The Quantum API v1.x was removed from the source code tree. To use the Quantum API v1.x, install the Quantum Essex release.

Term Description CRUD Create, read, update, and delete (CRUD) are the basic functions of persistent storage in computer programming. entity Any piece of hardware or software that can connect to the network services provided by OpenStack Networking. An entity can use OpenStack Networking services by implementing a VIF. IPAM Internet Protocol Address Management (IPAM) is a means of planning, tracking, and managing the Internet Protocol (IP) address space that is used in a network. layer-2 network A virtual Ethernet network that is managed by the OpenStack Networking service. Currently, OpenStack Networking manages only Ethernet networks. network An isolated virtual layer-2 broadcast domain that is typically reserved for the tenant who created it unless the network is configured to be shared. Tenants can create multiple networks until they reach the thresholds specified by per-tenant quotas. plug-in A software component that implements &APIv2;. port A virtual switch port on a logical network switch. Virtual instances attach their interfaces into ports. The logical port also defines the MAC address and the IP addresses to be assigned to the interfaces plugged into them. When creating a port, any unallocated IP in the subnet can be stated specifically, even if the IP address is not in the allocation pool. Enabling users to specify the IP explicitly allows them to retain a particular subset of the subnet IPs for static allocation. subnet An IP address block that can be used to assign IP addresses to virtual instances. Each subnet must have a CIDR and must be associated with a network. IPs can be either selected from the whole subnet CIDR or from allocation pools that can be specified by the user. VM A virtual machine (VM) is a completely isolated guest operating system installation within a normal host operating system.

The high-level task flow for OpenStack Networking involves creating a network, associating a subnet with that network, and booting a VM that is attached to the network. Clean-up includes deleting the VM, deleting any ports associated with the network, and deleting the networks. OpenStack Networking deletes any subnets associated with the deleted network. The tenant creates a network. For example, the tenant creates the net1 network. Its ID is net1_id. The tenant associates a subnet with that network. For example, the tenant associates the 10.0.0.0/24 subnet with the net1 network. The tenant boots a virtual machine (VM) and specifies a single NIC that connects to the network. The following examples use the nova client to boot a VM. In the first example, Nova contacts OpenStack Networking to create the NIC and attach it to the net1 network, with the ID net1_id: $ nova boot <server_name> --image <image> --flavor <flavor> --nic net-id=<net1_id> In a second example, you first create the port1, port and then you boot the VM with a specified port. OpenStack Networking creates a NIC and attaches it to the port1 port, with the ID port1_id: $ nova boot <server_name> --image <image> --flavor <flavor> --nic port-id=<port1_id> OpenStack Networking chooses and assigns an IP address to the port1 port. For more information about the nova boot command, enter: $ nova help boot The tenant deletes the VM. Nova contacts OpenStack Networking and deletes the port1 port. The allocated IP address is returned to the pool of available IP addresses. If the tenant created any ports and associated them with the network, the tenant deletes the ports. The tenant deletes the network. This operation deletes an OpenStack Networking network and its associated subnets provided that no port is currently configured on the network.

Virtual networking services are implemented through a plug-in. A plug-in can use different techniques and technologies to provide isolated virtual networks to tenants. A plug-in also provides other services, such as IP address management. Because each plug-in implements all the operations included in &APIv2;, do not be concerned about which plug-in is used. However, some plug-ins might expose additional capabilities through API extensions, which this document discusses. For more information about the extensions exposed by a particular plug-in, see the plug-in documentation.