openstack-manuals/doc/admin-guide-network/ch_under_the_hood.xml
Jon Proulx c55a3345fd use '--disable-dhcp' option
The option "--enable_dhcp" for "neutron subnet-create" is not visible
in the command documentation. The visible option is "--disable_dhcp".

Change-Id: Ieeac6bafc7707b114842ba7bbdf5b1310430597d
Closes-Bug: 1221923
2013-09-09 15:45:33 -04:00

576 lines
38 KiB
XML

<chapter xmlns="http://docbook.org/ns/docbook" xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xlink="http://www.w3.org/1999/xlink" version="5.0" xml:id="ch_under_the_hood">
<title>Under the Hood</title>
<para>This chapter describes two networking scenarios and how the Open vSwitch plugin and
the Linux bridging plugin implement these scenarios.</para>
<section xml:id="under_the_hood_openvswitch">
<?dbhtml stop-chunking?>
<title>Open vSwitch</title>
<para>This section describes how the Open vSwitch plugin implements the OpenStack
Networking abstractions.</para>
<section xml:id="under_the_hood_openvswitch_configuration">
<title>Configuration</title>
<para>This example uses VLAN isolation on the switches to isolate tenant networks. This
configuration labels the physical network associated with the public network as
<literal>physnet1</literal>, and the physical network associated with the data
network as <literal>physnet2</literal>, which leads to the following configuration
options in
<filename>ovs_neutron_plugin.ini</filename>:<programlisting language="bash">[ovs]
tenant_network_type = vlan
network_vlan_ranges = physnet2:100:110
integration_bridge = br-int
bridge_mappings = physnet2:br-eth1</programlisting></para>
</section>
<section xml:id="under_the_hood_openvswitch_scenario1">
<title>Scenario 1: one tenant, two networks, one router</title>
<para>The first scenario has two private networks (<literal>net01</literal>, and
<literal>net02</literal>), each with one subnet
(<literal>net01_subnet01</literal>: 192.168.101.0/24,
<literal>net02_subnet01</literal>, 192.168.102.0/24). Both private networks are
attached to a router that contains them to the public network (10.64.201.0/24).</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>Under the <literal>service</literal> tenant, create the shared router, define the
public network, and set it as the default gateway of the
router<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list | awk '/service/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create router01</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant public01 \
--provider:network_type flat \
--provider:physical_network physnet1 \
--router:external=True</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name public01_subnet01 \
--gateway 10.64.201.254 public01 10.64.201.0/24 --disable-dhcp</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router01 public01</userinput></screen></para>
<para>Under the <literal>demo</literal> user tenant, create the private network
<literal>net01</literal> and corresponding subnet, and connect it to the
<literal>router01</literal> router. Configure it to use VLAN ID 101 on the
physical
switch.<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/demo/ {print $2}'</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net01 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 101</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net01_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net01_subnet01</userinput></screen></para>
<para>Similarly, for <literal>net02</literal>, using VLAN ID 102 on the physical
switch:<screen><prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net02 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 102</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net02_subnet01 net02 192.168.102.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net02_subnet01</userinput></screen></para>
<section xml:id="under_the_hood_openvswitch_scenario1_compute">
<title>Scenario 1: Compute host config</title>
<para>The following figure shows how to configure various Linux networking devices on the compute host:
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-ovs-compute.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<simplesect>
<title>Types of network devices</title>
<note><para>There are four distinct type of virtual networking devices: TAP devices,
veth pairs, Linux bridges, and Open vSwitch bridges. For an ethernet frame to travel
from <literal>eth0</literal> of virtual machine <literal>vm01</literal> to the
physical network, it must pass through nine devices inside of the host: TAP
<literal>vnet0</literal>, Linux bridge
<literal>qbr<replaceable>nnn</replaceable></literal>, veth pair
<literal>(qvb<replaceable>nnn</replaceable>,
qvo<replaceable>nnn</replaceable>)</literal>, Open vSwitch bridge
<literal>br-int</literal>, veth pair <literal>(int-br-eth1,
phy-br-eth1)</literal>, and, finally, the physical network interface card
<literal>eth1</literal>.</para></note>
<para>A <emphasis role="italic">TAP device</emphasis>, such as <literal>vnet0</literal>
is how hypervisors such as KVM and Xen implement a virtual network interface card
(typically called a VIF or vNIC). An ethernet frame sent to a TAP device is received
by the guest operating system.</para>
<para>A <emphasis role="italic">veth pair</emphasis> is a pair of virtual network
interfaces correctly directly together. An ethernet frame sent to one end of a veth
pair is received by the other end of a veth pair. OpenStack networking makes use of
veth pairs as virtual patch cables in order to make connections between virtual
bridges.</para>
<para>A <emphasis role="italic">Linux bridge</emphasis> behaves like a hub: you can
connect multiple (physical or virtual) network interfaces devices to a Linux bridge.
Any ethernet frames that come in from one interface attached to the bridge is
transmitted to all of the other devices.</para>
<para>An <emphasis role="italic">Open vSwitch bridge</emphasis> behaves like a virtual
switch: network interface devices connect to Open vSwitch bridge's ports, and the
ports can be configured much like a physical switch's ports, including VLAN
configurations.</para>
</simplesect>
<simplesect>
<title>Integration bridge</title>
<para>The <literal>br-int</literal> OpenvSwitch bridge is the integration bridge: all of
the guests running on the compute host connect to this bridge. OpenStack Networking
implements isolation across these guests by configuring the
<literal>br-int</literal> ports.</para>
</simplesect>
<simplesect>
<title>Physical connectivity bridge</title>
<para>The <literal>br-eth1</literal> bridge provides connectivity to the physical
network interface card, <literal>eth1</literal>. It connects to the integration
bridge by a veth pair: <literal>(int-br-eth1, phy-br-eth1)</literal>.</para>
</simplesect>
<simplesect>
<title>VLAN translation</title>
<para>In this example, net01 and net02 have VLAN ids of 1 and 2, respectively. However,
the physical network in our example only supports VLAN IDs in the range 101 through 110. The
Open vSwitch agent is responsible for configuring flow rules on
<literal>br-int</literal> and <literal>br-eth1</literal> to do VLAN translation.
When <literal>br-eth1</literal> receives a frame marked with VLAN ID 1 on the port
associated with <literal>phy-br-eth1</literal>, it modifies the VLAN ID in the frame
to 101. Similarly, when <literal>br-int</literal> receives a frame marked with VLAN ID 101 on the port
associated with <literal>int-br-eth1</literal>, it modifies the VLAN ID in the frame
to 1.</para>
</simplesect>
<simplesect>
<title>Security groups: iptables and Linux bridges</title>
<para>Ideally, the TAP device <literal>vnet0</literal> would be connected directly to
the integration bridge, <literal>br-int</literal>. Unfortunately, this isn't
possible because of how OpenStack security groups are currently implemented.
OpenStack uses iptables rules on the TAP devices such as <literal>vnet0</literal> to
implement security groups, and Open vSwitch is not compatible with iptables rules
that are applied directly on TAP devices that are connected to an Open vSwitch
port.</para>
<para>OpenStack Networking uses an extra Linux bridge and a veth pair as a workaround for
this issue. Instead of connecting <literal>vnet0</literal> to an Open vSwitch
bridge, it is connected to a Linux bridge,
<literal>qbr<replaceable>XXX</replaceable></literal>. This bridge is
connected to the integration bridge, <literal>br-int</literal>, through the
<literal>(qvb<replaceable>XXX</replaceable>,
qvo<replaceable>XXX</replaceable>)</literal> veth pair.</para>
</simplesect>
</section>
<section xml:id="under_the_hood_openvswitch_scenario1_network">
<title>Scenario 1: Network host config</title>
<para>The network host runs the neutron-openvswitch-plugin-agent, the
neutron-dhcp-agent, neutron-l3-agent, and neutron-metadata-agent services.</para>
<para>On the network host, assume that eth0 is connected to the external network, and
eth1 is connected to the data network, which leads to the following configuration
in the
<filename>ovs_neutron_plugin.ini</filename> file:
<programlisting language="bash">[ovs]
tenant_network_type = vlan
network_vlan_ranges = physnet2:101:110
integration_bridge = br-int
bridge_mappings = physnet1:br-ex,physnet2:br-eth1</programlisting>
The following figure shows the network devices on the network host:</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-ovs-network.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>As on the compute host, there is an Open vSwitch integration bridge
(<literal>br-int</literal>) and an Open vSwitch bridge connected to the data
network (<literal>br-eth1</literal>), and the two are connected by a veth pair, and
the neutron-openvswitch-plugin-agent configures the ports on both switches to do
VLAN translation.</para>
<para>An additional Open vSwitch bridge, <literal>br-ex</literal>,
connects to the physical interface that is connected to the external network. In
this example, that physical interface is <literal>eth0</literal>.</para>
<note><para>While the integration bridge and the external bridge are connected by
a veth pair <literal>(int-br-ex, phy-br-ex)</literal>, this example uses layer 3
connectivity to route packets from the internal networks to the public network: no
packets traverse that veth pair in this example.</para></note>
<simplesect><title>Open vSwitch internal ports</title>
<para>The network host uses Open vSwitch <emphasis role="italic">internal
ports</emphasis>. Internal ports enable you to assign one
or more IP addresses to an Open vSwitch bridge. In previous example, the
<literal>br-int</literal> bridge has four internal
ports: <literal>tap<replaceable>XXX</replaceable></literal>,
<literal>qr-<replaceable>YYY</replaceable></literal>,
<literal>qr-<replaceable>ZZZ</replaceable></literal>,
<literal>tap<replaceable>WWW</replaceable></literal>. Each internal port has
a separate IP address associated with it. An internal port,
<literal>qg-VVV</literal>, is on the <literal>br-ex</literal> bridge.</para>
</simplesect>
<simplesect><title>DHCP agent</title>
<para>By default, The OpenStack Networking DHCP agent uses a program called dnsmasq
to provide DHCP services to guests. OpenStack Networking must create an internal
port for each network that requires DHCP services and attach a dnsmasq process to
that port. In the previous example, the interface
<literal>tap<replaceable>XXX</replaceable></literal> is on subnet
<literal>net01_subnet01</literal>, and the interface
<literal>tap<replaceable>WWW</replaceable></literal> is on
<literal>net02_subnet01</literal>.</para>
</simplesect>
<simplesect>
<title>L3 agent (routing)</title>
<para>The OpenStack Networking L3 agent implements routing through the use of Open
vSwitch internal ports and relies on the network host to route the packets across
the interfaces. In this example: interface<literal>qr-YYY</literal>, which is on
subnet <literal>net01_subnet01</literal>, has an IP address of 192.168.101.1/24,
interface <literal>qr-<replaceable>ZZZ</replaceable></literal>, which is on subnet
<literal>net02_subnet01</literal>, has an IP address of
<literal>192.168.102.1/24</literal>, and interface
<literal>qg-<replaceable>VVV</replaceable></literal>, which has an IP
address of <literal>10.64.201.254/24</literal>. Because of each of these interfaces
is visible to the network host operating system, it will route the packets
appropriately across the interfaces, as long as an administrator has enabled IP
forwarding.</para>
<para>The L3 agent uses iptables to implement floating IPs to do the network address
translation (NAT).</para>
</simplesect>
<simplesect>
<title>Overlapping subnets and network namespaces</title>
<para>One problem with using the host to implement routing is that there is a chance
that one of the OpenStack Networking subnets might overlap with one of the physical
networks that the host uses. For example, if the management network is implemented
on <literal>eth2</literal> (not shown in the previous example), by coincidence happens
to also be on the <literal>192.168.101.0/24</literal> subnet, then this will cause
routing problems because it is impossible ot determine whether a packet on this
subnet should be sent to <literal>qr-YYY</literal> or <literal>eth2</literal>. In
general, if end-users are permitted to create their own logical networks and
subnets, then the system must be designed to avoid the possibility of such
collisions.</para>
<para>OpenStack Networking uses Linux <emphasis role="italic">network namespaces
</emphasis>to prevent collisions between the physical networks on the network host,
and the logical networks used by the virtual machines. It also prevents collisions
across different logical networks that are not routed to each other, as you will see
in the next scenario.</para>
<para>A network namespace can be thought of as an isolated environment that has its own
networking stack. A network namespace has its own network interfaces, routes, and
iptables rules. You can think of like a chroot jail, except for networking instead
of a file system. As an aside, LXC (Linux containers) use network namespaces to
implement networking virtualization.</para>
<para>OpenStack Networking creates network namespaces on the network host in order
to avoid subnet collisions.</para>
<para>Tn this example, there are three network namespaces, as depicted in the following figure.<itemizedlist>
<listitem>
<para><literal>qdhcp-<replaceable>aaa</replaceable></literal>: contains the
<literal>tap<replaceable>XXX</replaceable></literal> interface
and the dnsmasq process that listens on that interface, to provide DHCP
services for <literal>net01_subnet01</literal>. This allows overlapping
IPs between <literal>net01_subnet01</literal> and any other subnets on
the network host.</para>
</listitem>
<listitem>
<para><literal>qrouter-<replaceable>bbbb</replaceable></literal>: contains
the <literal>qr-<replaceable>YYY</replaceable></literal>,
<literal>qr-<replaceable>ZZZ</replaceable></literal>, and
<literal>qg-<replaceable>VVV</replaceable></literal> interfaces,
and the corresponding routes. This namespace implements
<literal>router01</literal> in our example.</para>
</listitem>
<listitem>
<para><literal>qdhcp-<replaceable>ccc</replaceable></literal>: contains the
<literal>tap<replaceable>WWW</replaceable></literal> interface
and the dnsmasq process that listens on that interface, to provide DHCP
services for <literal>net02_subnet01</literal>. This allows overlapping
IPs between <literal>net02_subnet01</literal> and any other subnets on
the network host.</para>
</listitem>
</itemizedlist></para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-ovs-netns.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
</simplesect>
</section>
</section>
<section xml:id="under_the_hood_openvswitch_scenario2">
<title>Scenario 2: two tenants, two networks, two routers</title>
<para>In this scenario, tenant A and tenant B each have a
network with one subnet and one router that connects the
tenants to the public Internet.
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>Under the <literal>service</literal> tenant, define the public
network:<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list | awk '/service/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant public01 \
--provider:network_type flat \
--provider:physical_network physnet1 \
--router:external=True</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name public01_subnet01 \
--gateway 10.64.201.254 public01 10.64.201.0/24 --disable-dhcp</userinput></screen></para>
<para>Under the <literal>tenantA</literal> user tenant, create the tenant router and set
its gateway for the public
network.<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/tenantA/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create --tenant-id $tenant router01</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router01 public01</userinput></screen>
Then, define private network <literal>net01</literal> using VLAN ID 102 on the
physical switch, along with its subnet, and connect it to the router.
<screen><prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net01 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 101</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net01_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net01_subnet01</userinput></screen></para>
<para>Similarly, for <literal>tenantB</literal>, create a router and another network,
using VLAN ID 102 on the physical
switch:<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/tenantB/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create --tenant-id $tenant router02</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router02 public01</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net02 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 102</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net02_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router02 net02_subnet01</userinput></screen></para>
<section xml:id="under_the_hood_openvswitch_scenario2_compute">
<title>Scenario 2: Compute host config</title>
<para>The following figure shows how to configure Linux networking devices on the Compute host:
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-ovs-compute.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<note><para>The Compute host configuration resembles the
configuration in scenario 1. However, in scenario 1, a
guest connects to two subnets while in this scenario, the
subnets belong to different tenants.
</para></note>
</section>
<section xml:id="under_the_hood_openvswitch_scenario2_network">
<title>Scenario 2: Network host config</title>
<para>The following figure shows the network devices on the network host for the second
scenario.</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-ovs-network.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>In this configuration, the network namespaces are
organized to isolate the two subnets from each other as
shown in the following figure.
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-ovs-netns.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>In this scenario, there are four network namespaces
(<literal>qhdcp-<replaceable>aaa</replaceable></literal>,
<literal>qrouter-<replaceable>bbbb</replaceable></literal>,
<literal>qrouter-<replaceable>cccc</replaceable></literal>, and
<literal>qhdcp-<replaceable>dddd</replaceable></literal>), instead of three.
Since there is no connectivity between the two networks, and so each router is
implemented by a separate namespace.</para>
</section>
</section>
</section>
<section xml:id="under_the_hood_linuxbridge">
<title>Linux bridge</title>
<para>This section describes how the Linux bridge plugin
implements the OpenStack Networking abstractions. For
information about DHCP and L3 agents, see
<xref linkend="under_the_hood_openvswitch_scenario1" />.
</para>
<section xml:id="under_the_hood_linuxbridge_configuration">
<title>Configuration</title>
<para>This example uses VLAN isolation on the switches to isolate tenant networks. This configuration labels the physical
network associated with the public network as <literal>physnet1</literal>, and the
physical network associated with the data network as <literal>physnet2</literal>,
which leads to the following configuration options in
<filename>linuxbridge_conf.ini</filename>:<programlisting language="bash">[vlans]
tenant_network_type = vlan
network_vlan_ranges = physnet2:100:110
[linux_bridge]
physical_interface_mappings: physnet2:eth1</programlisting></para>
</section>
<section xml:id="under_the_hood_linuxbridge_scenario1">
<title>Scenario 1: one tenant, two networks, one router</title>
<para>The first scenario has two private networks (<literal>net01</literal>, and
<literal>net02</literal>), each with one subnet
(<literal>net01_subnet01</literal>: 192.168.101.0/24,
<literal>net02_subnet01</literal>, 192.168.102.0/24). Both private networks are
attached to a router that contains them to the public network (10.64.201.0/24).</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>Under the <literal>service</literal> tenant, create the shared router, define the
public network, and set it as the default gateway of the
router<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list | awk '/service/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create router01</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant public01 \
--provider:network_type flat \
--provider:physical_network physnet1 \
--router:external=True</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name public01_subnet01 \
--gateway 10.64.201.254 public01 10.64.201.0/24 --disable-dhcp</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router01 public01</userinput></screen></para>
<para>Under the <literal>demo</literal> user tenant, create the private network
<literal>net01</literal> and corresponding subnet, and connect it to the
<literal>router01</literal> router. Configure it to use VLAN ID 101 on the
physical
switch.<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/demo/ {print $2}'</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net01 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 101</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net01_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net01_subnet01</userinput></screen></para>
<para>Similarly, for <literal>net02</literal>, using VLAN ID 102 on the physical
switch:<screen><prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net02 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 102</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net02_subnet01 net02 192.168.102.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net02_subnet01</userinput></screen></para>
<section xml:id="under_the_hood_linuxbridge_scenario1_compute">
<title>Scenario 1: Compute host config</title>
<para>The following figure shows how to configure the various Linux networking devices on the
compute host.
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-linuxbridge-compute.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<simplesect>
<title>Types of network devices</title>
<note><para>There are three distinct type of virtual networking devices: TAP devices,
VLAN devices, and Linux bridges. For an ethernet frame to travel from
<literal>eth0</literal> of virtual machine <literal>vm01</literal>, to the
physical network, it must pass through four devices inside of the host: TAP
<literal>vnet0</literal>, Linux bridge
<literal>brq<replaceable>XXX</replaceable></literal>, VLAN
<literal>eth1.101)</literal>, and, finally, the physical network interface card
<literal>eth1</literal>.</para></note>
<para>A <emphasis role="italic">TAP device</emphasis>, such as <literal>vnet0</literal>
is how hypervisors such as KVM and Xen implement a virtual network interface card
(typically called a VIF or vNIC). An ethernet frame sent to a TAP device is received
by the guest operating system.</para>
<para>A <emphasis role="italic">VLAN device</emphasis> is associated with a VLAN tag
attaches to an existing interface device and adds or removes VLAN tags. In the
preceding example, VLAN device <literal>eth1.101</literal> is associated with VLAN ID
101 and is attached to interface <literal>eth1</literal>. Packets received from the
outside by <literal>eth1</literal> with VLAN tag 101 will be passed to device
<literal>eth1.101</literal>, which will then strip the tag. In the other
direction, any ethernet frame sent directly to eth1.101 will have VLAN tag 101 added
and will be forward to <literal>eth1</literal> for sending out to the
network.</para>
<para>A <emphasis role="italic">Linux bridge</emphasis> behaves like a hub: you can
connect multiple (physical or virtual) network interfaces devices to a Linux bridge.
Any ethernet frames that come in from one interface attached to the bridge is
transmitted to all of the other devices.</para>
</simplesect>
</section>
<section xml:id="under_the_hood_linuxbridge_scenario1_network">
<title>Scenario 1: Network host config</title>
<para>The following figure shows the network devices on the network host.</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-linuxbridge-network.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>The following figure shows how the Linux bridge plugin uses network namespaces to
provide isolation.</para><note><para>veth pairs form connections between the
Linux bridges and the network namespaces.</para></note><mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-1-linuxbridge-netns.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
</section>
</section>
<section xml:id="under_the_hood_linuxbridge_scenario2">
<title>Scenario 2: two tenants, two networks, two routers</title>
<para>The second scenario has two tenants (A, B). Each tenant has a network with
one subnet, and each one has a router that connects them to the public
Internet.</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>Under the <literal>service</literal> tenant, define the public
network:<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list | awk '/service/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant public01 \
--provider:network_type flat \
--provider:physical_network physnet1 \
--router:external=True</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name public01_subnet01 \
--gateway 10.64.201.254 public01 10.64.201.0/24 --disable-dhcp</userinput></screen></para>
<para>Under the <literal>tenantA</literal> user tenant, create the tenant router and set
its gateway for the public
network.<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/tenantA/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create --tenant-id $tenant router01</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router01 public01</userinput></screen>
Then, define private network <literal>net01</literal> using VLAN ID 102 on the
physical switch, along with its subnet, and connect it to the router.
<screen><prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net01 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 101</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net01_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router01 net01_subnet01</userinput></screen></para>
<para>Similarly, for <literal>tenantB</literal>, create a router and another network,
using VLAN ID 102 on the physical
switch:<screen><prompt>$</prompt> <userinput>tenant=$(keystone tenant-list|awk '/tenantB/ {print $2}')</userinput>
<prompt>$</prompt> <userinput>neutron router-create --tenant-id $tenant router02</userinput>
<prompt>$</prompt> <userinput>neutron router-gateway-set router02 public01</userinput>
<prompt>$</prompt> <userinput>neutron net-create --tenant-id $tenant net02 \
--provider:network_type vlan \
--provider:physical_network physnet2 \
--provider:segmentation_id 102</userinput>
<prompt>$</prompt> <userinput>neutron subnet-create --tenant-id $tenant --name net02_subnet01 net01 192.168.101.0/24</userinput>
<prompt>$</prompt> <userinput>neutron router-interface-add router02 net02_subnet01</userinput></screen></para>
<section xml:id="under_the_hood_linuxbridge_scenario2_compute">
<title>Scenario 2: Compute host config</title>
<para>The following figure shows how the various Linux networking devices would be configured on the
compute host under this scenario.
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-linuxbridge-compute.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<note><para>The configuration on the compute host is very similar to the configuration in scenario 1. The
only real difference is that scenario 1 had a guest that was connected to two
subnets, and in this scenario, the subnets belong to different tenants.</para></note>
</section>
<section xml:id="under_the_hood_linuxbridge_scenario2_network">
<title>Scenario 2: Network host config</title>
<para>The following figure shows the network devices on the network host for the second
scenario.</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-linuxbridge-network.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>The main difference between the configuration in this scenario and the previous one
is the organization of the network namespaces, in order to provide isolation
across the two subnets, as shown in the following figure.</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/under-the-hood-scenario-2-linuxbridge-netns.png" contentwidth="6in"/>
</imageobject>
</mediaobject>
<para>In this scenario, there are four network namespaces
(<literal>qhdcp-<replaceable>aaa</replaceable></literal>,
<literal>qrouter-<replaceable>bbbb</replaceable></literal>,
<literal>qrouter-<replaceable>cccc</replaceable></literal>, and
<literal>qhdcp-<replaceable>dddd</replaceable></literal>), instead of three.
Since there is no connectivity between the two networks, and so each router is
implemented by a separate namespace.</para>
</section>
</section>
</section>
</chapter>