1 <?xml version="1.0" encoding="utf-8"?>
2 <database title="Open vSwitch Configuration Database">
4 A database with this schema holds the configuration for one Open
5 vSwitch daemon. The top-level configuration for the daemon is the
6 <ref table="Open_vSwitch"/> table, which must have exactly one
7 record. Records in other tables are significant only when they
8 can be reached directly or indirectly from the <ref
9 table="Open_vSwitch"/> table. Records that are not reachable from
10 the <ref table="Open_vSwitch"/> table are automatically deleted
11 from the database, except for records in a few distinguished
15 <h2>Common Columns</h2>
18 Most tables contain two special columns, named <code>other_config</code>
19 and <code>external_ids</code>. These columns have the same form and
20 purpose each place that they appear, so we describe them here to save space
25 <dt><code>other_config</code>: map of string-string pairs</dt>
28 Key-value pairs for configuring rarely used features. Supported keys,
29 along with the forms taken by their values, are documented individually
33 A few tables do not have <code>other_config</code> columns because no
34 key-value pairs have yet been defined for them.
38 <dt><code>external_ids</code>: map of string-string pairs</dt>
40 Key-value pairs for use by external frameworks that integrate with Open
41 vSwitch, rather than by Open vSwitch itself. System integrators should
42 either use the Open vSwitch development mailing list to coordinate on
43 common key-value definitions, or choose key names that are likely to be
44 unique. In some cases, where key-value pairs have been defined that are
45 likely to be widely useful, they are documented individually for each
50 <table name="Open_vSwitch" title="Open vSwitch configuration.">
51 Configuration for an Open vSwitch daemon. There must be exactly
52 one record in the <ref table="Open_vSwitch"/> table.
54 <group title="Configuration">
55 <column name="bridges">
56 Set of bridges managed by the daemon.
60 SSL used globally by the daemon.
63 <column name="external_ids" key="system-id">
64 A unique identifier for the Open vSwitch's physical host.
65 The form of the identifier depends on the type of the host.
66 On a Citrix XenServer, this will likely be the same as
67 <ref column="external_ids" key="xs-system-uuid"/>.
70 <column name="external_ids" key="xs-system-uuid">
71 The Citrix XenServer universally unique identifier for the physical
72 host as displayed by <code>xe host-list</code>.
75 <column name="other_config" key="flow-restore-wait"
76 type='{"type": "boolean"}'>
78 When <code>ovs-vswitchd</code> starts up, it has an empty flow table
79 and therefore it handles all arriving packets in its default fashion
80 according to its configuration, by dropping them or sending them to
81 an OpenFlow controller or switching them as a standalone switch.
82 This behavior is ordinarily desirable. However, if
83 <code>ovs-vswitchd</code> is restarting as part of a ``hot-upgrade,''
84 then this leads to a relatively long period during which packets are
88 This option allows for improvement. When <code>ovs-vswitchd</code>
89 starts with this value set as <code>true</code>, it will neither
90 flush or expire previously set datapath flows nor will it send and
91 receive any packets to or from the datapath. When this value is
92 later set to <code>false</code>, <code>ovs-vswitchd</code> will
93 start receiving packets from the datapath and re-setup the flows.
96 Thus, with this option, the procedure for a hot-upgrade of
97 <code>ovs-vswitchd</code> becomes roughly the following:
101 Stop <code>ovs-vswitchd</code>.
104 Set <ref column="other_config" key="flow-restore-wait"/>
105 to <code>true</code>.
108 Start <code>ovs-vswitchd</code>.
111 Use <code>ovs-ofctl</code> (or some other program, such as an
112 OpenFlow controller) to restore the OpenFlow flow table
113 to the desired state.
116 Set <ref column="other_config" key="flow-restore-wait"/>
117 to <code>false</code> (or remove it entirely from the database).
121 The <code>ovs-ctl</code>'s ``restart'' and ``force-reload-kmod''
122 functions use the above config option during hot upgrades.
126 <column name="other_config" key="flow-eviction-threshold"
127 type='{"type": "integer", "minInteger": 0}'>
129 A number of flows as a nonnegative integer. This sets number of
130 flows at which eviction from the datapath flow table will be
131 triggered. If there are a large number of flows then increasing this
132 value to around the number of flows present can result in reduced CPU
133 usage and packet loss.
136 The default is 2500. Values below 100 will be rounded up to 100.
140 <column name="other_config" key="force-miss-model">
142 Specifies userspace behaviour for handling flow misses. This takes
143 precedence over flow-eviction-threshold.
147 <dt><code>auto</code></dt>
148 <dd>Handle automatically based on the flow-eviction-threshold and
149 the flow setup governer (default, recommended).</dd>
150 <dt><code>with-facets</code></dt>
151 <dd>Always create facets. Expensive kernel flow creation and
152 statistics tracking is always performed, even on flows with only
153 a small number of packets.</dd>
154 <dt><code>without-facets</code></dt>
155 <dd>Always handle without facets. Forces flow misses to be handled
156 in userspace. May cause an increase in CPU usage and packet loss
157 on high throughput.</dd>
162 <column name="other_config" key="n-handler-threads"
163 type='{"type": "integer", "minInteger": 1}'>
165 Specifies the number of threads for software datapaths to use for
166 handling new flows. The default is two less than the number of
167 online CPU cores (but at least 1).
170 This configuration is per datapath. If you have more than one
171 software datapath (e.g. some <code>system</code> bridges and some
172 <code>netdev</code> bridges), then the total number of threads is
173 <code>n-handler-threads</code> times the number of software
179 <group title="Status">
180 <column name="next_cfg">
181 Sequence number for client to increment. When a client modifies
182 any part of the database configuration and wishes to wait for
183 Open vSwitch to finish applying the changes, it may increment
184 this sequence number.
187 <column name="cur_cfg">
188 Sequence number that Open vSwitch sets to the current value of
189 <ref column="next_cfg"/> after it finishes applying a set of
190 configuration changes.
193 <group title="Statistics">
195 The <code>statistics</code> column contains key-value pairs that
196 report statistics about a system running an Open vSwitch. These are
197 updated periodically (currently, every 5 seconds). Key-value pairs
198 that cannot be determined or that do not apply to a platform are
202 <column name="other_config" key="enable-statistics"
203 type='{"type": "boolean"}'>
204 Statistics are disabled by default to avoid overhead in the common
205 case when statistics gathering is not useful. Set this value to
206 <code>true</code> to enable populating the <ref column="statistics"/>
207 column or to <code>false</code> to explicitly disable it.
210 <column name="statistics" key="cpu"
211 type='{"type": "integer", "minInteger": 1}'>
213 Number of CPU processors, threads, or cores currently online and
214 available to the operating system on which Open vSwitch is running,
215 as an integer. This may be less than the number installed, if some
216 are not online or if they are not available to the operating
220 Open vSwitch userspace processes are not multithreaded, but the
221 Linux kernel-based datapath is.
225 <column name="statistics" key="load_average">
226 A comma-separated list of three floating-point numbers,
227 representing the system load average over the last 1, 5, and 15
228 minutes, respectively.
231 <column name="statistics" key="memory">
233 A comma-separated list of integers, each of which represents a
234 quantity of memory in kilobytes that describes the operating
235 system on which Open vSwitch is running. In respective order,
240 <li>Total amount of RAM allocated to the OS.</li>
241 <li>RAM allocated to the OS that is in use.</li>
242 <li>RAM that can be flushed out to disk or otherwise discarded
243 if that space is needed for another purpose. This number is
244 necessarily less than or equal to the previous value.</li>
245 <li>Total disk space allocated for swap.</li>
246 <li>Swap space currently in use.</li>
250 On Linux, all five values can be determined and are included. On
251 other operating systems, only the first two values can be
252 determined, so the list will only have two values.
256 <column name="statistics" key="process_NAME">
258 One such key-value pair, with <code>NAME</code> replaced by
259 a process name, will exist for each running Open vSwitch
260 daemon process, with <var>name</var> replaced by the
261 daemon's name (e.g. <code>process_ovs-vswitchd</code>). The
262 value is a comma-separated list of integers. The integers
263 represent the following, with memory measured in kilobytes
264 and durations in milliseconds:
268 <li>The process's virtual memory size.</li>
269 <li>The process's resident set size.</li>
270 <li>The amount of user and system CPU time consumed by the
272 <li>The number of times that the process has crashed and been
273 automatically restarted by the monitor.</li>
274 <li>The duration since the process was started.</li>
275 <li>The duration for which the process has been running.</li>
279 The interpretation of some of these values depends on whether the
280 process was started with the <option>--monitor</option>. If it
281 was not, then the crash count will always be 0 and the two
282 durations will always be the same. If <option>--monitor</option>
283 was given, then the crash count may be positive; if it is, the
284 latter duration is the amount of time since the most recent crash
289 There will be one key-value pair for each file in Open vSwitch's
290 ``run directory'' (usually <code>/var/run/openvswitch</code>)
291 whose name ends in <code>.pid</code>, whose contents are a
292 process ID, and which is locked by a running process. The
293 <var>name</var> is taken from the pidfile's name.
297 Currently Open vSwitch is only able to obtain all of the above
298 detail on Linux systems. On other systems, the same key-value
299 pairs will be present but the values will always be the empty
304 <column name="statistics" key="file_systems">
306 A space-separated list of information on local, writable file
307 systems. Each item in the list describes one file system and
308 consists in turn of a comma-separated list of the following:
312 <li>Mount point, e.g. <code>/</code> or <code>/var/log</code>.
313 Any spaces or commas in the mount point are replaced by
315 <li>Total size, in kilobytes, as an integer.</li>
316 <li>Amount of storage in use, in kilobytes, as an integer.</li>
320 This key-value pair is omitted if there are no local, writable
321 file systems or if Open vSwitch cannot obtain the needed
328 <group title="Version Reporting">
330 These columns report the types and versions of the hardware and
331 software running Open vSwitch. We recommend in general that software
332 should test whether specific features are supported instead of relying
333 on version number checks. These values are primarily intended for
334 reporting to human administrators.
337 <column name="ovs_version">
338 The Open vSwitch version number, e.g. <code>1.1.0</code>.
341 <column name="db_version">
343 The database schema version number in the form
344 <code><var>major</var>.<var>minor</var>.<var>tweak</var></code>,
345 e.g. <code>1.2.3</code>. Whenever the database schema is changed in
346 a non-backward compatible way (e.g. deleting a column or a table),
347 <var>major</var> is incremented. When the database schema is changed
348 in a backward compatible way (e.g. adding a new column),
349 <var>minor</var> is incremented. When the database schema is changed
350 cosmetically (e.g. reindenting its syntax), <var>tweak</var> is
355 The schema version is part of the database schema, so it can also be
356 retrieved by fetching the schema using the Open vSwitch database
361 <column name="system_type">
363 An identifier for the type of system on top of which Open vSwitch
364 runs, e.g. <code>XenServer</code> or <code>KVM</code>.
367 System integrators are responsible for choosing and setting an
368 appropriate value for this column.
372 <column name="system_version">
374 The version of the system identified by <ref column="system_type"/>,
375 e.g. <code>5.6.100-39265p</code> on XenServer 5.6.100 build 39265.
378 System integrators are responsible for choosing and setting an
379 appropriate value for this column.
385 <group title="Database Configuration">
387 These columns primarily configure the Open vSwitch database
388 (<code>ovsdb-server</code>), not the Open vSwitch switch
389 (<code>ovs-vswitchd</code>). The OVSDB database also uses the <ref
390 column="ssl"/> settings.
394 The Open vSwitch switch does read the database configuration to
395 determine remote IP addresses to which in-band control should apply.
398 <column name="manager_options">
399 Database clients to which the Open vSwitch database server should
400 connect or to which it should listen, along with options for how these
401 connection should be configured. See the <ref table="Manager"/> table
402 for more information.
406 <group title="Common Columns">
407 The overall purpose of these columns is described under <code>Common
408 Columns</code> at the beginning of this document.
410 <column name="other_config"/>
411 <column name="external_ids"/>
415 <table name="Bridge">
417 Configuration for a bridge within an
418 <ref table="Open_vSwitch"/>.
421 A <ref table="Bridge"/> record represents an Ethernet switch with one or
422 more ``ports,'' which are the <ref table="Port"/> records pointed to by
423 the <ref table="Bridge"/>'s <ref column="ports"/> column.
426 <group title="Core Features">
428 Bridge identifier. Should be alphanumeric and no more than about 8
429 bytes long. Must be unique among the names of ports, interfaces, and
433 <column name="ports">
434 Ports included in the bridge.
437 <column name="mirrors">
438 Port mirroring configuration.
441 <column name="netflow">
442 NetFlow configuration.
445 <column name="sflow">
446 sFlow(R) configuration.
449 <column name="ipfix">
453 <column name="flood_vlans">
455 VLAN IDs of VLANs on which MAC address learning should be disabled,
456 so that packets are flooded instead of being sent to specific ports
457 that are believed to contain packets' destination MACs. This should
458 ordinarily be used to disable MAC learning on VLANs used for
459 mirroring (RSPAN VLANs). It may also be useful for debugging.
462 SLB bonding (see the <ref table="Port" column="bond_mode"/> column in
463 the <ref table="Port"/> table) is incompatible with
464 <code>flood_vlans</code>. Consider using another bonding mode or
465 a different type of mirror instead.
470 <group title="OpenFlow Configuration">
471 <column name="controller">
473 OpenFlow controller set. If unset, then no OpenFlow controllers
478 If there are primary controllers, removing all of them clears the
479 flow table. If there are no primary controllers, adding one also
480 clears the flow table. Other changes to the set of controllers, such
481 as adding or removing a service controller, adding another primary
482 controller to supplement an existing primary controller, or removing
483 only one of two primary controllers, have no effect on the flow
488 <column name="flow_tables">
489 Configuration for OpenFlow tables. Each pair maps from an OpenFlow
490 table ID to configuration for that table.
493 <column name="fail_mode">
494 <p>When a controller is configured, it is, ordinarily, responsible
495 for setting up all flows on the switch. Thus, if the connection to
496 the controller fails, no new network connections can be set up.
497 If the connection to the controller stays down long enough,
498 no packets can pass through the switch at all. This setting
499 determines the switch's response to such a situation. It may be set
500 to one of the following:
502 <dt><code>standalone</code></dt>
503 <dd>If no message is received from the controller for three
504 times the inactivity probe interval
505 (see <ref column="inactivity_probe"/>), then Open vSwitch
506 will take over responsibility for setting up flows. In
507 this mode, Open vSwitch causes the bridge to act like an
508 ordinary MAC-learning switch. Open vSwitch will continue
509 to retry connecting to the controller in the background
510 and, when the connection succeeds, it will discontinue its
511 standalone behavior.</dd>
512 <dt><code>secure</code></dt>
513 <dd>Open vSwitch will not set up flows on its own when the
514 controller connection fails or when no controllers are
515 defined. The bridge will continue to retry connecting to
516 any defined controllers forever.</dd>
520 The default is <code>standalone</code> if the value is unset, but
521 future versions of Open vSwitch may change the default.
524 The <code>standalone</code> mode can create forwarding loops on a
525 bridge that has more than one uplink port unless STP is enabled. To
526 avoid loops on such a bridge, configure <code>secure</code> mode or
527 enable STP (see <ref column="stp_enable"/>).
529 <p>When more than one controller is configured,
530 <ref column="fail_mode"/> is considered only when none of the
531 configured controllers can be contacted.</p>
533 Changing <ref column="fail_mode"/> when no primary controllers are
534 configured clears the flow table.
538 <column name="datapath_id">
539 Reports the OpenFlow datapath ID in use. Exactly 16 hex digits.
540 (Setting this column has no useful effect. Set <ref
541 column="other-config" key="datapath-id"/> instead.)
544 <column name="other_config" key="datapath-id">
545 Exactly 16 hex digits to set the OpenFlow datapath ID to a specific
546 value. May not be all-zero.
549 <column name="other_config" key="dp-desc">
550 Human readable description of datapath. It it a maximum 256
551 byte-long free-form string to describe the datapath for
552 debugging purposes, e.g. <code>switch3 in room 3120</code>.
555 <column name="other_config" key="disable-in-band"
556 type='{"type": "boolean"}'>
557 If set to <code>true</code>, disable in-band control on the bridge
558 regardless of controller and manager settings.
561 <column name="other_config" key="in-band-queue"
562 type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'>
563 A queue ID as a nonnegative integer. This sets the OpenFlow queue ID
564 that will be used by flows set up by in-band control on this bridge.
565 If unset, or if the port used by an in-band control flow does not have
566 QoS configured, or if the port does not have a queue with the specified
567 ID, the default queue is used instead.
570 <column name="protocols">
571 List of OpenFlow protocols that may be used when negotiating a
572 connection with a controller. A default value of
573 <code>OpenFlow10</code> will be used if this column is empty.
577 <group title="Spanning Tree Configuration">
578 The IEEE 802.1D Spanning Tree Protocol (STP) is a network protocol
579 that ensures loop-free topologies. It allows redundant links to
580 be included in the network to provide automatic backup paths if
581 the active links fails.
583 <column name="stp_enable">
584 Enable spanning tree on the bridge. By default, STP is disabled
585 on bridges. Bond, internal, and mirror ports are not supported
586 and will not participate in the spanning tree.
589 <column name="other_config" key="stp-system-id">
590 The bridge's STP identifier (the lower 48 bits of the bridge-id)
592 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>.
593 By default, the identifier is the MAC address of the bridge.
596 <column name="other_config" key="stp-priority"
597 type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'>
598 The bridge's relative priority value for determining the root
599 bridge (the upper 16 bits of the bridge-id). A bridge with the
600 lowest bridge-id is elected the root. By default, the priority
604 <column name="other_config" key="stp-hello-time"
605 type='{"type": "integer", "minInteger": 1, "maxInteger": 10}'>
606 The interval between transmissions of hello messages by
607 designated ports, in seconds. By default the hello interval is
611 <column name="other_config" key="stp-max-age"
612 type='{"type": "integer", "minInteger": 6, "maxInteger": 40}'>
613 The maximum age of the information transmitted by the bridge
614 when it is the root bridge, in seconds. By default, the maximum
618 <column name="other_config" key="stp-forward-delay"
619 type='{"type": "integer", "minInteger": 4, "maxInteger": 30}'>
620 The delay to wait between transitioning root and designated
621 ports to <code>forwarding</code>, in seconds. By default, the
622 forwarding delay is 15 seconds.
626 <group title="Other Features">
627 <column name="datapath_type">
628 Name of datapath provider. The kernel datapath has
629 type <code>system</code>. The userspace datapath has
630 type <code>netdev</code>.
633 <column name="external_ids" key="bridge-id">
634 A unique identifier of the bridge. On Citrix XenServer this will
635 commonly be the same as
636 <ref column="external_ids" key="xs-network-uuids"/>.
639 <column name="external_ids" key="xs-network-uuids">
640 Semicolon-delimited set of universally unique identifier(s) for the
641 network with which this bridge is associated on a Citrix XenServer
642 host. The network identifiers are RFC 4122 UUIDs as displayed by,
643 e.g., <code>xe network-list</code>.
646 <column name="other_config" key="hwaddr">
647 An Ethernet address in the form
648 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>
649 to set the hardware address of the local port and influence the
653 <column name="other_config" key="forward-bpdu"
654 type='{"type": "boolean"}'>
655 Option to allow forwarding of BPDU frames when NORMAL action is
656 invoked. Frames with reserved Ethernet addresses (e.g. STP
657 BPDU) will be forwarded when this option is enabled and the
658 switch is not providing that functionality. If STP is enabled
659 on the port, STP BPDUs will never be forwarded. If the Open
660 vSwitch bridge is used to connect different Ethernet networks,
661 and if Open vSwitch node does not run STP, then this option
662 should be enabled. Default is disabled, set to
663 <code>true</code> to enable.
665 The following destination MAC addresss will not be forwarded when this
668 <dt><code>01:80:c2:00:00:00</code></dt>
669 <dd>IEEE 802.1D Spanning Tree Protocol (STP).</dd>
671 <dt><code>01:80:c2:00:00:01</code></dt>
672 <dd>IEEE Pause frame.</dd>
674 <dt><code>01:80:c2:00:00:0<var>x</var></code></dt>
675 <dd>Other reserved protocols.</dd>
677 <dt><code>00:e0:2b:00:00:00</code></dt>
678 <dd>Extreme Discovery Protocol (EDP).</dd>
681 <code>00:e0:2b:00:00:04</code> and <code>00:e0:2b:00:00:06</code>
683 <dd>Ethernet Automatic Protection Switching (EAPS).</dd>
685 <dt><code>01:00:0c:cc:cc:cc</code></dt>
687 Cisco Discovery Protocol (CDP), VLAN Trunking Protocol (VTP),
688 Dynamic Trunking Protocol (DTP), Port Aggregation Protocol (PAgP),
692 <dt><code>01:00:0c:cc:cc:cd</code></dt>
693 <dd>Cisco Shared Spanning Tree Protocol PVSTP+.</dd>
695 <dt><code>01:00:0c:cd:cd:cd</code></dt>
696 <dd>Cisco STP Uplink Fast.</dd>
698 <dt><code>01:00:0c:00:00:00</code></dt>
699 <dd>Cisco Inter Switch Link.</dd>
701 <dt><code>01:00:0c:cc:cc:c<var>x</var></code></dt>
706 <column name="other_config" key="mac-aging-time"
707 type='{"type": "integer", "minInteger": 1}'>
709 The maximum number of seconds to retain a MAC learning entry for
710 which no packets have been seen. The default is currently 300
711 seconds (5 minutes). The value, if specified, is forced into a
712 reasonable range, currently 15 to 3600 seconds.
716 A short MAC aging time allows a network to more quickly detect that a
717 host is no longer connected to a switch port. However, it also makes
718 it more likely that packets will be flooded unnecessarily, when they
719 are addressed to a connected host that rarely transmits packets. To
720 reduce the incidence of unnecessary flooding, use a MAC aging time
721 longer than the maximum interval at which a host will ordinarily
726 <column name="other_config" key="mac-table-size"
727 type='{"type": "integer", "minInteger": 1}'>
729 The maximum number of MAC addresses to learn. The default is
730 currently 2048. The value, if specified, is forced into a reasonable
731 range, currently 10 to 1,000,000.
736 <group title="Bridge Status">
738 Status information about bridges.
740 <column name="status">
741 Key-value pairs that report bridge status.
743 <column name="status" key="stp_bridge_id">
745 The bridge-id (in hex) used in spanning tree advertisements.
746 Configuring the bridge-id is described in the
747 <code>stp-system-id</code> and <code>stp-priority</code> keys
748 of the <code>other_config</code> section earlier.
751 <column name="status" key="stp_designated_root">
753 The designated root (in hex) for this spanning tree.
756 <column name="status" key="stp_root_path_cost">
758 The path cost of reaching the designated bridge. A lower
764 <group title="Common Columns">
765 The overall purpose of these columns is described under <code>Common
766 Columns</code> at the beginning of this document.
768 <column name="other_config"/>
769 <column name="external_ids"/>
773 <table name="Port" table="Port or bond configuration.">
774 <p>A port within a <ref table="Bridge"/>.</p>
775 <p>Most commonly, a port has exactly one ``interface,'' pointed to by its
776 <ref column="interfaces"/> column. Such a port logically
777 corresponds to a port on a physical Ethernet switch. A port
778 with more than one interface is a ``bonded port'' (see
779 <ref group="Bonding Configuration"/>).</p>
780 <p>Some properties that one might think as belonging to a port are actually
781 part of the port's <ref table="Interface"/> members.</p>
784 Port name. Should be alphanumeric and no more than about 8
785 bytes long. May be the same as the interface name, for
786 non-bonded ports. Must otherwise be unique among the names of
787 ports, interfaces, and bridges on a host.
790 <column name="interfaces">
791 The port's interfaces. If there is more than one, this is a
795 <group title="VLAN Configuration">
796 <p>Bridge ports support the following types of VLAN configuration:</p>
801 A trunk port carries packets on one or more specified VLANs
802 specified in the <ref column="trunks"/> column (often, on every
803 VLAN). A packet that ingresses on a trunk port is in the VLAN
804 specified in its 802.1Q header, or VLAN 0 if the packet has no
805 802.1Q header. A packet that egresses through a trunk port will
806 have an 802.1Q header if it has a nonzero VLAN ID.
810 Any packet that ingresses on a trunk port tagged with a VLAN that
811 the port does not trunk is dropped.
818 An access port carries packets on exactly one VLAN specified in the
819 <ref column="tag"/> column. Packets egressing on an access port
820 have no 802.1Q header.
824 Any packet with an 802.1Q header with a nonzero VLAN ID that
825 ingresses on an access port is dropped, regardless of whether the
826 VLAN ID in the header is the access port's VLAN ID.
830 <dt>native-tagged</dt>
832 A native-tagged port resembles a trunk port, with the exception that
833 a packet without an 802.1Q header that ingresses on a native-tagged
834 port is in the ``native VLAN'' (specified in the <ref column="tag"/>
838 <dt>native-untagged</dt>
840 A native-untagged port resembles a native-tagged port, with the
841 exception that a packet that egresses on a native-untagged port in
842 the native VLAN will not have an 802.1Q header.
846 A packet will only egress through bridge ports that carry the VLAN of
847 the packet, as described by the rules above.
850 <column name="vlan_mode">
852 The VLAN mode of the port, as described above. When this column is
853 empty, a default mode is selected as follows:
857 If <ref column="tag"/> contains a value, the port is an access
858 port. The <ref column="trunks"/> column should be empty.
861 Otherwise, the port is a trunk port. The <ref column="trunks"/>
862 column value is honored if it is present.
869 For an access port, the port's implicitly tagged VLAN. For a
870 native-tagged or native-untagged port, the port's native VLAN. Must
871 be empty if this is a trunk port.
875 <column name="trunks">
877 For a trunk, native-tagged, or native-untagged port, the 802.1Q VLAN
878 or VLANs that this port trunks; if it is empty, then the port trunks
879 all VLANs. Must be empty if this is an access port.
882 A native-tagged or native-untagged port always trunks its native
883 VLAN, regardless of whether <ref column="trunks"/> includes that
888 <column name="other_config" key="priority-tags"
889 type='{"type": "boolean"}'>
891 An 802.1Q header contains two important pieces of information: a VLAN
892 ID and a priority. A frame with a zero VLAN ID, called a
893 ``priority-tagged'' frame, is supposed to be treated the same way as
894 a frame without an 802.1Q header at all (except for the priority).
898 However, some network elements ignore any frame that has 802.1Q
899 header at all, even when the VLAN ID is zero. Therefore, by default
900 Open vSwitch does not output priority-tagged frames, instead omitting
901 the 802.1Q header entirely if the VLAN ID is zero. Set this key to
902 <code>true</code> to enable priority-tagged frames on a port.
906 Regardless of this setting, Open vSwitch omits the 802.1Q header on
907 output if both the VLAN ID and priority would be zero.
911 All frames output to native-tagged ports have a nonzero VLAN ID, so
912 this setting is not meaningful on native-tagged ports.
917 <group title="Bonding Configuration">
918 <p>A port that has more than one interface is a ``bonded port.'' Bonding
919 allows for load balancing and fail-over.</p>
922 The following types of bonding will work with any kind of upstream
923 switch. On the upstream switch, do not configure the interfaces as a
928 <dt><code>balance-slb</code></dt>
930 Balances flows among slaves based on source MAC address and output
931 VLAN, with periodic rebalancing as traffic patterns change.
934 <dt><code>active-backup</code></dt>
936 Assigns all flows to one slave, failing over to a backup slave when
937 the active slave is disabled. This is the only bonding mode in which
938 interfaces may be plugged into different upstream switches.
943 The following modes require the upstream switch to support 802.3ad with
944 successful LACP negotiation. If LACP negotiation fails and
945 other-config:lacp-fallback-ab is true, then <code>active-backup</code>
950 <dt><code>balance-tcp</code></dt>
952 Balances flows among slaves based on L2, L3, and L4 protocol
953 information such as destination MAC address, IP address, and TCP
958 <p>These columns apply only to bonded ports. Their values are
959 otherwise ignored.</p>
961 <column name="bond_mode">
962 <p>The type of bonding used for a bonded port. Defaults to
963 <code>active-backup</code> if unset.
967 <column name="other_config" key="bond-hash-basis"
968 type='{"type": "integer"}'>
969 An integer hashed along with flows when choosing output slaves in load
970 balanced bonds. When changed, all flows will be assigned different
971 hash values possibly causing slave selection decisions to change. Does
972 not affect bonding modes which do not employ load balancing such as
973 <code>active-backup</code>.
976 <group title="Link Failure Detection">
978 An important part of link bonding is detecting that links are down so
979 that they may be disabled. These settings determine how Open vSwitch
980 detects link failure.
983 <column name="other_config" key="bond-detect-mode"
984 type='{"type": "string", "enum": ["set", ["carrier", "miimon"]]}'>
985 The means used to detect link failures. Defaults to
986 <code>carrier</code> which uses each interface's carrier to detect
987 failures. When set to <code>miimon</code>, will check for failures
988 by polling each interface's MII.
991 <column name="other_config" key="bond-miimon-interval"
992 type='{"type": "integer"}'>
993 The interval, in milliseconds, between successive attempts to poll
994 each interface's MII. Relevant only when <ref column="other_config"
995 key="bond-detect-mode"/> is <code>miimon</code>.
998 <column name="bond_updelay">
1000 The number of milliseconds for which the link must stay up on an
1001 interface before the interface is considered to be up. Specify
1002 <code>0</code> to enable the interface immediately.
1006 This setting is honored only when at least one bonded interface is
1007 already enabled. When no interfaces are enabled, then the first
1008 bond interface to come up is enabled immediately.
1012 <column name="bond_downdelay">
1013 The number of milliseconds for which the link must stay down on an
1014 interface before the interface is considered to be down. Specify
1015 <code>0</code> to disable the interface immediately.
1019 <group title="LACP Configuration">
1021 LACP, the Link Aggregation Control Protocol, is an IEEE standard that
1022 allows switches to automatically detect that they are connected by
1023 multiple links and aggregate across those links. These settings
1024 control LACP behavior.
1027 <column name="lacp">
1028 Configures LACP on this port. LACP allows directly connected
1029 switches to negotiate which links may be bonded. LACP may be enabled
1030 on non-bonded ports for the benefit of any switches they may be
1031 connected to. <code>active</code> ports are allowed to initiate LACP
1032 negotiations. <code>passive</code> ports are allowed to participate
1033 in LACP negotiations initiated by a remote switch, but not allowed to
1034 initiate such negotiations themselves. If LACP is enabled on a port
1035 whose partner switch does not support LACP, the bond will be
1036 disabled, unless other-config:lacp-fallback-ab is set to true.
1037 Defaults to <code>off</code> if unset.
1040 <column name="other_config" key="lacp-system-id">
1041 The LACP system ID of this <ref table="Port"/>. The system ID of a
1042 LACP bond is used to identify itself to its partners. Must be a
1043 nonzero MAC address. Defaults to the bridge Ethernet address if
1047 <column name="other_config" key="lacp-system-priority"
1048 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
1049 The LACP system priority of this <ref table="Port"/>. In LACP
1050 negotiations, link status decisions are made by the system with the
1051 numerically lower priority.
1054 <column name="other_config" key="lacp-time"
1055 type='{"type": "string", "enum": ["set", ["fast", "slow"]]}'>
1057 The LACP timing which should be used on this <ref table="Port"/>.
1058 By default <code>slow</code> is used. When configured to be
1059 <code>fast</code> LACP heartbeats are requested at a rate of once
1060 per second causing connectivity problems to be detected more
1061 quickly. In <code>slow</code> mode, heartbeats are requested at a
1062 rate of once every 30 seconds.
1066 <column name="other_config" key="lacp-fallback-ab"
1067 type='{"type": "boolean"}'>
1069 Determines the behavior of openvswitch bond in LACP mode. If
1070 the partner switch does not support LACP, setting this option
1071 to <code>true</code> allows openvswitch to fallback to
1072 active-backup. If the option is set to <code>false</code>, the
1073 bond will be disabled. In both the cases, once the partner switch
1074 is configured to LACP mode, the bond will use LACP.
1079 <group title="Rebalancing Configuration">
1081 These settings control behavior when a bond is in
1082 <code>balance-slb</code> or <code>balance-tcp</code> mode.
1085 <column name="other_config" key="bond-rebalance-interval"
1086 type='{"type": "integer", "minInteger": 0, "maxInteger": 10000}'>
1087 For a load balanced bonded port, the number of milliseconds between
1088 successive attempts to rebalance the bond, that is, to move flows
1089 from one interface on the bond to another in an attempt to keep usage
1090 of each interface roughly equal. If zero, load balancing is disabled
1091 on the bond (link failure still cause flows to move). If
1092 less than 1000ms, the rebalance interval will be 1000ms.
1096 <column name="bond_fake_iface">
1097 For a bonded port, whether to create a fake internal interface with the
1098 name of the port. Use only for compatibility with legacy software that
1103 <group title="Spanning Tree Configuration">
1104 <column name="other_config" key="stp-enable"
1105 type='{"type": "boolean"}'>
1106 If spanning tree is enabled on the bridge, member ports are
1107 enabled by default (with the exception of bond, internal, and
1108 mirror ports which do not work with STP). If this column's
1109 value is <code>false</code> spanning tree is disabled on the
1113 <column name="other_config" key="stp-port-num"
1114 type='{"type": "integer", "minInteger": 1, "maxInteger": 255}'>
1115 The port number used for the lower 8 bits of the port-id. By
1116 default, the numbers will be assigned automatically. If any
1117 port's number is manually configured on a bridge, then they
1121 <column name="other_config" key="stp-port-priority"
1122 type='{"type": "integer", "minInteger": 0, "maxInteger": 255}'>
1123 The port's relative priority value for determining the root
1124 port (the upper 8 bits of the port-id). A port with a lower
1125 port-id will be chosen as the root port. By default, the
1129 <column name="other_config" key="stp-path-cost"
1130 type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'>
1131 Spanning tree path cost for the port. A lower number indicates
1132 a faster link. By default, the cost is based on the maximum
1137 <group title="Other Features">
1139 Quality of Service configuration for this port.
1143 The MAC address to use for this port for the purpose of choosing the
1144 bridge's MAC address. This column does not necessarily reflect the
1145 port's actual MAC address, nor will setting it change the port's actual
1149 <column name="fake_bridge">
1150 Does this port represent a sub-bridge for its tagged VLAN within the
1151 Bridge? See ovs-vsctl(8) for more information.
1154 <column name="external_ids" key="fake-bridge-id-*">
1155 External IDs for a fake bridge (see the <ref column="fake_bridge"/>
1156 column) are defined by prefixing a <ref table="Bridge"/> <ref
1157 table="Bridge" column="external_ids"/> key with
1158 <code>fake-bridge-</code>,
1159 e.g. <code>fake-bridge-xs-network-uuids</code>.
1163 <group title="Port Status">
1165 Status information about ports attached to bridges.
1167 <column name="status">
1168 Key-value pairs that report port status.
1170 <column name="status" key="stp_port_id">
1172 The port-id (in hex) used in spanning tree advertisements for
1173 this port. Configuring the port-id is described in the
1174 <code>stp-port-num</code> and <code>stp-port-priority</code>
1175 keys of the <code>other_config</code> section earlier.
1178 <column name="status" key="stp_state"
1179 type='{"type": "string", "enum": ["set",
1180 ["disabled", "listening", "learning",
1181 "forwarding", "blocking"]]}'>
1183 STP state of the port.
1186 <column name="status" key="stp_sec_in_state"
1187 type='{"type": "integer", "minInteger": 0}'>
1189 The amount of time (in seconds) port has been in the current
1193 <column name="status" key="stp_role"
1194 type='{"type": "string", "enum": ["set",
1195 ["root", "designated", "alternate"]]}'>
1197 STP role of the port.
1202 <group title="Port Statistics">
1204 Key-value pairs that report port statistics.
1206 <group title="Statistics: STP transmit and receive counters">
1207 <column name="statistics" key="stp_tx_count">
1208 Number of STP BPDUs sent on this port by the spanning
1211 <column name="statistics" key="stp_rx_count">
1212 Number of STP BPDUs received on this port and accepted by the
1213 spanning tree library.
1215 <column name="statistics" key="stp_error_count">
1216 Number of bad STP BPDUs received on this port. Bad BPDUs
1217 include runt packets and those with an unexpected protocol ID.
1222 <group title="Common Columns">
1223 The overall purpose of these columns is described under <code>Common
1224 Columns</code> at the beginning of this document.
1226 <column name="other_config"/>
1227 <column name="external_ids"/>
1231 <table name="Interface" title="One physical network device in a Port.">
1232 An interface within a <ref table="Port"/>.
1234 <group title="Core Features">
1235 <column name="name">
1236 Interface name. Should be alphanumeric and no more than about 8 bytes
1237 long. May be the same as the port name, for non-bonded ports. Must
1238 otherwise be unique among the names of ports, interfaces, and bridges
1242 <column name="ifindex">
1243 A positive interface index as defined for SNMP MIB-II in RFCs 1213 and
1244 2863, if the interface has one, otherwise 0. The ifindex is useful for
1245 seamless integration with protocols such as SNMP and sFlow.
1248 <column name="mac_in_use">
1249 The MAC address in use by this interface.
1253 <p>Ethernet address to set for this interface. If unset then the
1254 default MAC address is used:</p>
1256 <li>For the local interface, the default is the lowest-numbered MAC
1257 address among the other bridge ports, either the value of the
1258 <ref table="Port" column="mac"/> in its <ref table="Port"/> record,
1259 if set, or its actual MAC (for bonded ports, the MAC of its slave
1260 whose name is first in alphabetical order). Internal ports and
1261 bridge ports that are used as port mirroring destinations (see the
1262 <ref table="Mirror"/> table) are ignored.</li>
1263 <li>For other internal interfaces, the default MAC is randomly
1265 <li>External interfaces typically have a MAC address associated with
1266 their hardware.</li>
1268 <p>Some interfaces may not have a software-controllable MAC
1272 <group title="OpenFlow Port Number">
1274 When a client adds a new interface, Open vSwitch chooses an OpenFlow
1275 port number for the new port. If the client that adds the port fills
1276 in <ref column="ofport_request"/>, then Open vSwitch tries to use its
1277 value as the OpenFlow port number. Otherwise, or if the requested
1278 port number is already in use or cannot be used for another reason,
1279 Open vSwitch automatically assigns a free port number. Regardless of
1280 how the port number was obtained, Open vSwitch then reports in <ref
1281 column="ofport"/> the port number actually assigned.
1285 Open vSwitch limits the port numbers that it automatically assigns to
1286 the range 1 through 32,767, inclusive. Controllers therefore have
1287 free use of ports 32,768 and up.
1290 <column name="ofport">
1292 OpenFlow port number for this interface. Open vSwitch sets this
1293 column's value, so other clients should treat it as read-only.
1296 The OpenFlow ``local'' port (<code>OFPP_LOCAL</code>) is 65,534.
1297 The other valid port numbers are in the range 1 to 65,279,
1298 inclusive. Value -1 indicates an error adding the interface.
1302 <column name="ofport_request"
1303 type='{"type": "integer", "minInteger": 1, "maxInteger": 65279}'>
1305 Requested OpenFlow port number for this interface.
1309 Open vSwitch currently assigns the OpenFlow port number for an
1310 interface once, when the client first adds the interface. It does
1311 not change the port number later if the client sets or changes or
1312 clears <ref column="ofport_request"/>. Therefore, to ensure that
1313 <ref column="ofport_request"/> takes effect, the client should set
1314 it in the same database transaction that creates the interface.
1315 (Future versions of Open vSwitch might honor changes to <ref
1316 column="ofport_request"/>.)
1322 <group title="System-Specific Details">
1323 <column name="type">
1325 The interface type, one of:
1329 <dt><code>system</code></dt>
1330 <dd>An ordinary network device, e.g. <code>eth0</code> on Linux.
1331 Sometimes referred to as ``external interfaces'' since they are
1332 generally connected to hardware external to that on which the Open
1333 vSwitch is running. The empty string is a synonym for
1334 <code>system</code>.</dd>
1336 <dt><code>internal</code></dt>
1337 <dd>A simulated network device that sends and receives traffic. An
1338 internal interface whose <ref column="name"/> is the same as its
1339 bridge's <ref table="Open_vSwitch" column="name"/> is called the
1340 ``local interface.'' It does not make sense to bond an internal
1341 interface, so the terms ``port'' and ``interface'' are often used
1342 imprecisely for internal interfaces.</dd>
1344 <dt><code>tap</code></dt>
1345 <dd>A TUN/TAP device managed by Open vSwitch.</dd>
1347 <dt><code>gre</code></dt>
1349 An Ethernet over RFC 2890 Generic Routing Encapsulation over IPv4
1353 <dt><code>ipsec_gre</code></dt>
1355 An Ethernet over RFC 2890 Generic Routing Encapsulation over IPv4
1359 <dt><code>gre64</code></dt>
1361 It is same as GRE, but it allows 64 bit key. To store higher 32-bits
1362 of key, it uses GRE protocol sequence number field. This is non
1363 standard use of GRE protocol since OVS does not increment
1364 sequence number for every packet at time of encap as expected by
1365 standard GRE implementation. See <ref group="Tunnel Options"/>
1366 for information on configuring GRE tunnels.
1369 <dt><code>ipsec_gre64</code></dt>
1371 Same as IPSEC_GRE except 64 bit key.
1374 <dt><code>vxlan</code></dt>
1377 An Ethernet tunnel over the experimental, UDP-based VXLAN
1378 protocol described at
1379 <code>http://tools.ietf.org/html/draft-mahalingam-dutt-dcops-vxlan-03</code>.
1382 Open vSwitch uses UDP destination port 4789. The source port used for
1383 VXLAN traffic varies on a per-flow basis and is in the ephemeral port
1388 <dt><code>lisp</code></dt>
1390 A layer 3 tunnel over the experimental, UDP-based Locator/ID
1391 Separation Protocol (RFC 6830).
1394 <dt><code>patch</code></dt>
1396 A pair of virtual devices that act as a patch cable.
1399 <dt><code>null</code></dt>
1400 <dd>An ignored interface. Deprecated and slated for removal in
1406 <group title="Tunnel Options">
1408 These options apply to interfaces with <ref column="type"/> of
1409 <code>gre</code>, <code>ipsec_gre</code>, <code>gre64</code>,
1410 <code>ipsec_gre64</code>, <code>vxlan</code>, and <code>lisp</code>.
1414 Each tunnel must be uniquely identified by the combination of <ref
1415 column="type"/>, <ref column="options" key="remote_ip"/>, <ref
1416 column="options" key="local_ip"/>, and <ref column="options"
1417 key="in_key"/>. If two ports are defined that are the same except one
1418 has an optional identifier and the other does not, the more specific
1419 one is matched first. <ref column="options" key="in_key"/> is
1420 considered more specific than <ref column="options" key="local_ip"/> if
1421 a port defines one and another port defines the other.
1424 <column name="options" key="remote_ip">
1425 <p>Required. The remote tunnel endpoint, one of:</p>
1429 An IPv4 address (not a DNS name), e.g. <code>192.168.0.123</code>.
1430 Only unicast endpoints are supported.
1433 The word <code>flow</code>. The tunnel accepts packets from any
1434 remote tunnel endpoint. To process only packets from a specific
1435 remote tunnel endpoint, the flow entries may match on the
1436 <code>tun_src</code> field. When sending packets to a
1437 <code>remote_ip=flow</code> tunnel, the flow actions must
1438 explicitly set the <code>tun_dst</code> field to the IP address of
1439 the desired remote tunnel endpoint, e.g. with a
1440 <code>set_field</code> action.
1445 The remote tunnel endpoint for any packet received from a tunnel
1446 is available in the <code>tun_src</code> field for matching in the
1451 <column name="options" key="local_ip">
1453 Optional. The tunnel destination IP that received packets must
1454 match. Default is to match all addresses. If specified, may be one
1460 An IPv4 address (not a DNS name), e.g. <code>192.168.12.3</code>.
1463 The word <code>flow</code>. The tunnel accepts packets sent to any
1464 of the local IP addresses of the system running OVS. To process
1465 only packets sent to a specific IP address, the flow entries may
1466 match on the <code>tun_dst</code> field. When sending packets to a
1467 <code>local_ip=flow</code> tunnel, the flow actions may
1468 explicitly set the <code>tun_src</code> field to the desired IP
1469 address, e.g. with a <code>set_field</code> action. However, while
1470 routing the tunneled packet out, the local system may override the
1471 specified address with the local IP address configured for the
1472 outgoing system interface.
1475 This option is valid only for tunnels also configured with the
1476 <code>remote_ip=flow</code> option.
1482 The tunnel destination IP address for any packet received from a
1483 tunnel is available in the <code>tun_dst</code> field for matching in
1488 <column name="options" key="in_key">
1489 <p>Optional. The key that received packets must contain, one of:</p>
1493 <code>0</code>. The tunnel receives packets with no key or with a
1494 key of 0. This is equivalent to specifying no <ref column="options"
1495 key="in_key"/> at all.
1498 A positive 24-bit (for VXLAN and LISP), 32-bit (for GRE) or 64-bit
1499 (for GRE64) number. The tunnel receives only packets with the
1503 The word <code>flow</code>. The tunnel accepts packets with any
1504 key. The key will be placed in the <code>tun_id</code> field for
1505 matching in the flow table. The <code>ovs-ofctl</code> manual page
1506 contains additional information about matching fields in OpenFlow
1515 <column name="options" key="out_key">
1516 <p>Optional. The key to be set on outgoing packets, one of:</p>
1520 <code>0</code>. Packets sent through the tunnel will have no key.
1521 This is equivalent to specifying no <ref column="options"
1522 key="out_key"/> at all.
1525 A positive 24-bit (for VXLAN and LISP), 32-bit (for GRE) or 64-bit
1526 (for GRE64) number. Packets sent through the tunnel will have the
1530 The word <code>flow</code>. Packets sent through the tunnel will
1531 have the key set using the <code>set_tunnel</code> Nicira OpenFlow
1532 vendor extension (0 is used in the absence of an action). The
1533 <code>ovs-ofctl</code> manual page contains additional information
1534 about the Nicira OpenFlow vendor extensions.
1539 <column name="options" key="key">
1540 Optional. Shorthand to set <code>in_key</code> and
1541 <code>out_key</code> at the same time.
1544 <column name="options" key="tos">
1545 Optional. The value of the ToS bits to be set on the encapsulating
1546 packet. ToS is interpreted as DSCP and ECN bits, ECN part must be
1547 zero. It may also be the word <code>inherit</code>, in which case
1548 the ToS will be copied from the inner packet if it is IPv4 or IPv6
1549 (otherwise it will be 0). The ECN fields are always inherited.
1553 <column name="options" key="ttl">
1554 Optional. The TTL to be set on the encapsulating packet. It may also
1555 be the word <code>inherit</code>, in which case the TTL will be copied
1556 from the inner packet if it is IPv4 or IPv6 (otherwise it will be the
1557 system default, typically 64). Default is the system default TTL.
1560 <column name="options" key="df_default"
1561 type='{"type": "boolean"}'>
1562 Optional. If enabled, the Don't Fragment bit will be set on tunnel
1563 outer headers to allow path MTU discovery. Default is enabled; set
1564 to <code>false</code> to disable.
1567 <group title="Tunnel Options: gre and ipsec_gre only">
1569 Only <code>gre</code> and <code>ipsec_gre</code> interfaces support
1573 <column name="options" key="csum" type='{"type": "boolean"}'>
1575 Optional. Compute GRE checksums on outgoing packets. Default is
1576 disabled, set to <code>true</code> to enable. Checksums present on
1577 incoming packets will be validated regardless of this setting.
1581 GRE checksums impose a significant performance penalty because they
1582 cover the entire packet. The encapsulated L3, L4, and L7 packet
1583 contents typically have their own checksums, so this additional
1584 checksum only adds value for the GRE and encapsulated L2 headers.
1588 This option is supported for <code>ipsec_gre</code>, but not useful
1589 because GRE checksums are weaker than, and redundant with, IPsec
1590 payload authentication.
1595 <group title="Tunnel Options: ipsec_gre only">
1597 Only <code>ipsec_gre</code> interfaces support these options.
1600 <column name="options" key="peer_cert">
1601 Required for certificate authentication. A string containing the
1602 peer's certificate in PEM format. Additionally the host's
1603 certificate must be specified with the <code>certificate</code>
1607 <column name="options" key="certificate">
1608 Required for certificate authentication. The name of a PEM file
1609 containing a certificate that will be presented to the peer during
1613 <column name="options" key="private_key">
1614 Optional for certificate authentication. The name of a PEM file
1615 containing the private key associated with <code>certificate</code>.
1616 If <code>certificate</code> contains the private key, this option may
1620 <column name="options" key="psk">
1621 Required for pre-shared key authentication. Specifies a pre-shared
1622 key for authentication that must be identical on both sides of the
1628 <group title="Patch Options">
1630 Only <code>patch</code> interfaces support these options.
1633 <column name="options" key="peer">
1634 The <ref column="name"/> of the <ref table="Interface"/> for the other
1635 side of the patch. The named <ref table="Interface"/>'s own
1636 <code>peer</code> option must specify this <ref table="Interface"/>'s
1637 name. That is, the two patch interfaces must have reversed <ref
1638 column="name"/> and <code>peer</code> values.
1642 <group title="Interface Status">
1644 Status information about interfaces attached to bridges, updated every
1645 5 seconds. Not all interfaces have all of these properties; virtual
1646 interfaces don't have a link speed, for example. Non-applicable
1647 columns will have empty values.
1649 <column name="admin_state">
1651 The administrative state of the physical network link.
1655 <column name="link_state">
1657 The observed state of the physical network link. This is ordinarily
1658 the link's carrier status. If the interface's <ref table="Port"/> is
1659 a bond configured for miimon monitoring, it is instead the network
1660 link's miimon status.
1664 <column name="link_resets">
1666 The number of times Open vSwitch has observed the
1667 <ref column="link_state"/> of this <ref table="Interface"/> change.
1671 <column name="link_speed">
1673 The negotiated speed of the physical network link.
1674 Valid values are positive integers greater than 0.
1678 <column name="duplex">
1680 The duplex mode of the physical network link.
1686 The MTU (maximum transmission unit); i.e. the largest
1687 amount of data that can fit into a single Ethernet frame.
1688 The standard Ethernet MTU is 1500 bytes. Some physical media
1689 and many kinds of virtual interfaces can be configured with
1693 This column will be empty for an interface that does not
1694 have an MTU as, for example, some kinds of tunnels do not.
1698 <column name="lacp_current">
1699 Boolean value indicating LACP status for this interface. If true, this
1700 interface has current LACP information about its LACP partner. This
1701 information may be used to monitor the health of interfaces in a LACP
1702 enabled port. This column will be empty if LACP is not enabled.
1705 <column name="status">
1706 Key-value pairs that report port status. Supported status values are
1707 <ref column="type"/>-dependent; some interfaces may not have a valid
1708 <ref column="status" key="driver_name"/>, for example.
1711 <column name="status" key="driver_name">
1712 The name of the device driver controlling the network adapter.
1715 <column name="status" key="driver_version">
1716 The version string of the device driver controlling the network
1720 <column name="status" key="firmware_version">
1721 The version string of the network adapter's firmware, if available.
1724 <column name="status" key="source_ip">
1725 The source IP address used for an IPv4 tunnel end-point, such as
1729 <column name="status" key="tunnel_egress_iface">
1730 Egress interface for tunnels. Currently only relevant for GRE tunnels
1731 On Linux systems, this column will show the name of the interface
1732 which is responsible for routing traffic destined for the configured
1733 <ref column="options" key="remote_ip"/>. This could be an internal
1734 interface such as a bridge port.
1737 <column name="status" key="tunnel_egress_iface_carrier"
1738 type='{"type": "string", "enum": ["set", ["down", "up"]]}'>
1739 Whether carrier is detected on <ref column="status"
1740 key="tunnel_egress_iface"/>.
1744 <group title="Statistics">
1746 Key-value pairs that report interface statistics. The current
1747 implementation updates these counters periodically. Future
1748 implementations may update them when an interface is created, when they
1749 are queried (e.g. using an OVSDB <code>select</code> operation), and
1750 just before an interface is deleted due to virtual interface hot-unplug
1751 or VM shutdown, and perhaps at other times, but not on any regular
1755 These are the same statistics reported by OpenFlow in its <code>struct
1756 ofp_port_stats</code> structure. If an interface does not support a
1757 given statistic, then that pair is omitted.
1759 <group title="Statistics: Successful transmit and receive counters">
1760 <column name="statistics" key="rx_packets">
1761 Number of received packets.
1763 <column name="statistics" key="rx_bytes">
1764 Number of received bytes.
1766 <column name="statistics" key="tx_packets">
1767 Number of transmitted packets.
1769 <column name="statistics" key="tx_bytes">
1770 Number of transmitted bytes.
1773 <group title="Statistics: Receive errors">
1774 <column name="statistics" key="rx_dropped">
1775 Number of packets dropped by RX.
1777 <column name="statistics" key="rx_frame_err">
1778 Number of frame alignment errors.
1780 <column name="statistics" key="rx_over_err">
1781 Number of packets with RX overrun.
1783 <column name="statistics" key="rx_crc_err">
1784 Number of CRC errors.
1786 <column name="statistics" key="rx_errors">
1787 Total number of receive errors, greater than or equal to the sum of
1791 <group title="Statistics: Transmit errors">
1792 <column name="statistics" key="tx_dropped">
1793 Number of packets dropped by TX.
1795 <column name="statistics" key="collisions">
1796 Number of collisions.
1798 <column name="statistics" key="tx_errors">
1799 Total number of transmit errors, greater than or equal to the sum of
1805 <group title="Ingress Policing">
1807 These settings control ingress policing for packets received on this
1808 interface. On a physical interface, this limits the rate at which
1809 traffic is allowed into the system from the outside; on a virtual
1810 interface (one connected to a virtual machine), this limits the rate at
1811 which the VM is able to transmit.
1814 Policing is a simple form of quality-of-service that simply drops
1815 packets received in excess of the configured rate. Due to its
1816 simplicity, policing is usually less accurate and less effective than
1817 egress QoS (which is configured using the <ref table="QoS"/> and <ref
1818 table="Queue"/> tables).
1821 Policing is currently implemented only on Linux. The Linux
1822 implementation uses a simple ``token bucket'' approach:
1826 The size of the bucket corresponds to <ref
1827 column="ingress_policing_burst"/>. Initially the bucket is full.
1830 Whenever a packet is received, its size (converted to tokens) is
1831 compared to the number of tokens currently in the bucket. If the
1832 required number of tokens are available, they are removed and the
1833 packet is forwarded. Otherwise, the packet is dropped.
1836 Whenever it is not full, the bucket is refilled with tokens at the
1837 rate specified by <ref column="ingress_policing_rate"/>.
1841 Policing interacts badly with some network protocols, and especially
1842 with fragmented IP packets. Suppose that there is enough network
1843 activity to keep the bucket nearly empty all the time. Then this token
1844 bucket algorithm will forward a single packet every so often, with the
1845 period depending on packet size and on the configured rate. All of the
1846 fragments of an IP packets are normally transmitted back-to-back, as a
1847 group. In such a situation, therefore, only one of these fragments
1848 will be forwarded and the rest will be dropped. IP does not provide
1849 any way for the intended recipient to ask for only the remaining
1850 fragments. In such a case there are two likely possibilities for what
1851 will happen next: either all of the fragments will eventually be
1852 retransmitted (as TCP will do), in which case the same problem will
1853 recur, or the sender will not realize that its packet has been dropped
1854 and data will simply be lost (as some UDP-based protocols will do).
1855 Either way, it is possible that no forward progress will ever occur.
1857 <column name="ingress_policing_rate">
1859 Maximum rate for data received on this interface, in kbps. Data
1860 received faster than this rate is dropped. Set to <code>0</code>
1861 (the default) to disable policing.
1865 <column name="ingress_policing_burst">
1866 <p>Maximum burst size for data received on this interface, in kb. The
1867 default burst size if set to <code>0</code> is 1000 kb. This value
1868 has no effect if <ref column="ingress_policing_rate"/>
1869 is <code>0</code>.</p>
1871 Specifying a larger burst size lets the algorithm be more forgiving,
1872 which is important for protocols like TCP that react severely to
1873 dropped packets. The burst size should be at least the size of the
1874 interface's MTU. Specifying a value that is numerically at least as
1875 large as 10% of <ref column="ingress_policing_rate"/> helps TCP come
1876 closer to achieving the full rate.
1881 <group title="Bidirectional Forwarding Detection (BFD)">
1883 BFD, defined in RFC 5880 and RFC 5881, allows point-to-point
1884 detection of connectivity failures by occasional transmission of
1885 BFD control messages. Open vSwitch implements BFD to serve
1886 as a more popular and standards compliant alternative to CFM.
1890 BFD operates by regularly transmitting BFD control messages at a rate
1891 negotiated independently in each direction. Each endpoint specifies
1892 the rate at which it expects to receive control messages, and the rate
1893 at which it is willing to transmit them. Open vSwitch uses a detection
1894 multiplier of three, meaning that an endpoint signals a connectivity
1895 fault if three consecutive BFD control messages fail to arrive. In the
1896 case of a unidirectional connectivity issue, the system not receiving
1897 BFD control messages signals the problem to its peer in the messages it
1902 The Open vSwitch implementation of BFD aims to comply faithfully
1903 with RFC 5880 requirements. Open vSwitch does not implement the
1904 optional Authentication or ``Echo Mode'' features.
1907 <group title="BFD Configuration">
1909 A controller sets up key-value pairs in the <ref column="bfd"/>
1910 column to enable and configure BFD.
1913 <column name="bfd" key="enable" type='{"type": "boolean"}'>
1914 True to enable BFD on this <ref table="Interface"/>.
1917 <column name="bfd" key="min_rx"
1918 type='{"type": "integer", "minInteger": 1}'>
1919 The shortest interval, in milliseconds, at which this BFD session
1920 offers to receive BFD control messages. The remote endpoint may
1921 choose to send messages at a slower rate. Defaults to
1925 <column name="bfd" key="min_tx"
1926 type='{"type": "integer", "minInteger": 1}'>
1927 The shortest interval, in milliseconds, at which this BFD session is
1928 willing to transmit BFD control messages. Messages will actually be
1929 transmitted at a slower rate if the remote endpoint is not willing to
1930 receive as quickly as specified. Defaults to <code>100</code>.
1933 <column name="bfd" key="decay_min_rx" type='{"type": "integer"}'>
1934 An alternate receive interval, in milliseconds, that must be greater
1935 than or equal to <ref column="bfd" key="min_rx"/>. The
1936 implementation switches from <ref column="bfd" key="min_rx"/> to <ref
1937 column="bfd" key="decay_min_rx"/> when there is no obvious incoming
1938 data traffic at the interface, to reduce the CPU and bandwidth cost
1939 of monitoring an idle interface. This feature may be disabled by
1940 setting a value of 0. This feature is reset whenever <ref
1941 column="bfd" key="decay_min_rx"/> or <ref column="bfd" key="min_rx"/>
1945 <column name="bfd" key="forwarding_if_rx" type='{"type": "boolean"}'>
1946 True to consider the interface capable of packet I/O as long as it
1947 continues to receive any packets (not just BFD packets). This
1948 prevents link congestion that causes consecutive BFD control packets
1949 to be lost from marking the interface down.
1952 <column name="bfd" key="cpath_down" type='{"type": "boolean"}'>
1953 Set to true to notify the remote endpoint that traffic should not be
1954 forwarded to this system for some reason other than a connectivty
1955 failure on the interface being monitored. The typical underlying
1956 reason is ``concatenated path down,'' that is, that connectivity
1957 beyond the local system is down. Defaults to false.
1960 <column name="bfd" key="check_tnl_key" type='{"type": "boolean"}'>
1961 Set to true to make BFD accept only control messages with a tunnel
1962 key of zero. By default, BFD accepts control messages with any
1966 <column name="bfd" key="bfd_dst_mac">
1967 Set to an Ethernet address in the form
1968 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>
1969 to set the MAC used as destination for transmitted BFD packets and
1970 expected as destination for received BFD packets. The default is
1971 <code>00:23:20:00:00:01</code>.
1975 <group title="BFD Status">
1977 The switch sets key-value pairs in the <ref column="bfd_status"/>
1978 column to report the status of BFD on this interface. When BFD is
1979 not enabled, with <ref column="bfd" key="enable"/>, the switch clears
1980 all key-value pairs from <ref column="bfd_status"/>.
1983 <column name="bfd_status" key="state"
1984 type='{"type": "string",
1985 "enum": ["set", ["admin_down", "down", "init", "up"]]}'>
1986 Reports the state of the BFD session. The BFD session is fully
1987 healthy and negotiated if <code>UP</code>.
1990 <column name="bfd_status" key="forwarding" type='{"type": "boolean"}'>
1991 Reports whether the BFD session believes this <ref
1992 table="Interface"/> may be used to forward traffic. Typically this
1993 means the local session is signaling <code>UP</code>, and the remote
1994 system isn't signaling a problem such as concatenated path down.
1997 <column name="bfd_status" key="diagnostic">
1998 In case of a problem, set to a short message that reports what the
1999 local BFD session thinks is wrong.
2002 <column name="bfd_status" key="remote_state"
2003 type='{"type": "string",
2004 "enum": ["set", ["admin_down", "down", "init", "up"]]}'>
2005 Reports the state of the remote endpoint's BFD session.
2008 <column name="bfd_status" key="remote_diagnostic">
2009 In case of a problem, set to a short message that reports what the
2010 remote endpoint's BFD session thinks is wrong.
2015 <group title="Connectivity Fault Management">
2017 802.1ag Connectivity Fault Management (CFM) allows a group of
2018 Maintenance Points (MPs) called a Maintenance Association (MA) to
2019 detect connectivity problems with each other. MPs within a MA should
2020 have complete and exclusive interconnectivity. This is verified by
2021 occasionally broadcasting Continuity Check Messages (CCMs) at a
2022 configurable transmission interval.
2026 According to the 802.1ag specification, each Maintenance Point should
2027 be configured out-of-band with a list of Remote Maintenance Points it
2028 should have connectivity to. Open vSwitch differs from the
2029 specification in this area. It simply assumes the link is faulted if
2030 no Remote Maintenance Points are reachable, and considers it not
2035 When operating over tunnels which have no <code>in_key</code>, or an
2036 <code>in_key</code> of <code>flow</code>. CFM will only accept CCMs
2037 with a tunnel key of zero.
2040 <column name="cfm_mpid">
2041 A Maintenance Point ID (MPID) uniquely identifies each endpoint within
2042 a Maintenance Association. The MPID is used to identify this endpoint
2043 to other Maintenance Points in the MA. Each end of a link being
2044 monitored should have a different MPID. Must be configured to enable
2045 CFM on this <ref table="Interface"/>.
2048 <column name="cfm_flap_count">
2049 Counts the number of cfm fault flapps since boot. A flap is
2050 considered to be a change of the <ref column="cfm_fault"/> value.
2053 <column name="cfm_fault">
2055 Indicates a connectivity fault triggered by an inability to receive
2056 heartbeats from any remote endpoint. When a fault is triggered on
2057 <ref table="Interface"/>s participating in bonds, they will be
2061 Faults can be triggered for several reasons. Most importantly they
2062 are triggered when no CCMs are received for a period of 3.5 times the
2063 transmission interval. Faults are also triggered when any CCMs
2064 indicate that a Remote Maintenance Point is not receiving CCMs but
2065 able to send them. Finally, a fault is triggered if a CCM is
2066 received which indicates unexpected configuration. Notably, this
2067 case arises when a CCM is received which advertises the local MPID.
2071 <column name="cfm_fault_status" key="recv">
2072 Indicates a CFM fault was triggered due to a lack of CCMs received on
2073 the <ref table="Interface"/>.
2076 <column name="cfm_fault_status" key="rdi">
2077 Indicates a CFM fault was triggered due to the reception of a CCM with
2078 the RDI bit flagged. Endpoints set the RDI bit in their CCMs when they
2079 are not receiving CCMs themselves. This typically indicates a
2080 unidirectional connectivity failure.
2083 <column name="cfm_fault_status" key="maid">
2084 Indicates a CFM fault was triggered due to the reception of a CCM with
2085 a MAID other than the one Open vSwitch uses. CFM broadcasts are tagged
2086 with an identification number in addition to the MPID called the MAID.
2087 Open vSwitch only supports receiving CCM broadcasts tagged with the
2088 MAID it uses internally.
2091 <column name="cfm_fault_status" key="loopback">
2092 Indicates a CFM fault was triggered due to the reception of a CCM
2093 advertising the same MPID configured in the <ref column="cfm_mpid"/>
2094 column of this <ref table="Interface"/>. This may indicate a loop in
2098 <column name="cfm_fault_status" key="overflow">
2099 Indicates a CFM fault was triggered because the CFM module received
2100 CCMs from more remote endpoints than it can keep track of.
2103 <column name="cfm_fault_status" key="override">
2104 Indicates a CFM fault was manually triggered by an administrator using
2105 an <code>ovs-appctl</code> command.
2108 <column name="cfm_fault_status" key="interval">
2109 Indicates a CFM fault was triggered due to the reception of a CCM
2110 frame having an invalid interval.
2113 <column name="cfm_remote_opstate">
2114 <p>When in extended mode, indicates the operational state of the
2115 remote endpoint as either <code>up</code> or <code>down</code>. See
2116 <ref column="other_config" key="cfm_opstate"/>.
2120 <column name="cfm_health">
2122 Indicates the health of the interface as a percentage of CCM frames
2123 received over 21 <ref column="other_config" key="cfm_interval"/>s.
2124 The health of an interface is undefined if it is communicating with
2125 more than one <ref column="cfm_remote_mpids"/>. It reduces if
2126 healthy heartbeats are not received at the expected rate, and
2127 gradually improves as healthy heartbeats are received at the desired
2128 rate. Every 21 <ref column="other_config" key="cfm_interval"/>s, the
2129 health of the interface is refreshed.
2132 As mentioned above, the faults can be triggered for several reasons.
2133 The link health will deteriorate even if heartbeats are received but
2134 they are reported to be unhealthy. An unhealthy heartbeat in this
2135 context is a heartbeat for which either some fault is set or is out
2136 of sequence. The interface health can be 100 only on receiving
2137 healthy heartbeats at the desired rate.
2141 <column name="cfm_remote_mpids">
2142 When CFM is properly configured, Open vSwitch will occasionally
2143 receive CCM broadcasts. These broadcasts contain the MPID of the
2144 sending Maintenance Point. The list of MPIDs from which this
2145 <ref table="Interface"/> is receiving broadcasts from is regularly
2146 collected and written to this column.
2149 <column name="other_config" key="cfm_interval"
2150 type='{"type": "integer"}'>
2152 The interval, in milliseconds, between transmissions of CFM
2153 heartbeats. Three missed heartbeat receptions indicate a
2158 In standard operation only intervals of 3, 10, 100, 1,000, 10,000,
2159 60,000, or 600,000 ms are supported. Other values will be rounded
2160 down to the nearest value on the list. Extended mode (see <ref
2161 column="other_config" key="cfm_extended"/>) supports any interval up
2162 to 65,535 ms. In either mode, the default is 1000 ms.
2165 <p>We do not recommend using intervals less than 100 ms.</p>
2168 <column name="other_config" key="cfm_extended"
2169 type='{"type": "boolean"}'>
2170 When <code>true</code>, the CFM module operates in extended mode. This
2171 causes it to use a nonstandard destination address to avoid conflicting
2172 with compliant implementations which may be running concurrently on the
2173 network. Furthermore, extended mode increases the accuracy of the
2174 <code>cfm_interval</code> configuration parameter by breaking wire
2175 compatibility with 802.1ag compliant implementations. Defaults to
2179 <column name="other_config" key="cfm_demand" type='{"type": "boolean"}'>
2181 When <code>true</code>, and
2182 <ref column="other_config" key="cfm_extended"/> is true, the CFM
2183 module operates in demand mode. When in demand mode, traffic
2184 received on the <ref table="Interface"/> is used to indicate
2185 liveness. CCMs are still transmitted and received, but if the
2186 <ref table="Interface"/> is receiving traffic, their absence does not
2187 cause a connectivity fault.
2191 Demand mode has a couple of caveats:
2194 To ensure that ovs-vswitchd has enough time to pull statistics
2195 from the datapath, the fault detection interval is set to
2196 3.5 * MAX(<ref column="other_config" key="cfm_interval"/>, 500)
2201 To avoid ambiguity, demand mode disables itself when there are
2202 multiple remote maintenance points.
2206 If the <ref table="Interface"/> is heavily congested, CCMs
2207 containing the <ref column="other_config" key="cfm_opstate"/>
2208 status may be dropped causing changes in the operational state to
2209 be delayed. Similarly, if CCMs containing the RDI bit are not
2210 received, unidirectional link failures may not be detected.
2216 <column name="other_config" key="cfm_opstate"
2217 type='{"type": "string", "enum": ["set", ["down", "up"]]}'>
2218 When <code>down</code>, the CFM module marks all CCMs it generates as
2219 operationally down without triggering a fault. This allows remote
2220 maintenance points to choose not to forward traffic to the
2221 <ref table="Interface"/> on which this CFM module is running.
2222 Currently, in Open vSwitch, the opdown bit of CCMs affects
2223 <ref table="Interface"/>s participating in bonds, and the bundle
2224 OpenFlow action. This setting is ignored when CFM is not in extended
2225 mode. Defaults to <code>up</code>.
2228 <column name="other_config" key="cfm_ccm_vlan"
2229 type='{"type": "integer", "minInteger": 1, "maxInteger": 4095}'>
2230 When set, the CFM module will apply a VLAN tag to all CCMs it generates
2231 with the given value. May be the string <code>random</code> in which
2232 case each CCM will be tagged with a different randomly generated VLAN.
2235 <column name="other_config" key="cfm_ccm_pcp"
2236 type='{"type": "integer", "minInteger": 1, "maxInteger": 7}'>
2237 When set, the CFM module will apply a VLAN tag to all CCMs it generates
2238 with the given PCP value, the VLAN ID of the tag is governed by the
2239 value of <ref column="other_config" key="cfm_ccm_vlan"/>. If
2240 <ref column="other_config" key="cfm_ccm_vlan"/> is unset, a VLAN ID of
2246 <group title="Bonding Configuration">
2247 <column name="other_config" key="lacp-port-id"
2248 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
2249 The LACP port ID of this <ref table="Interface"/>. Port IDs are
2250 used in LACP negotiations to identify individual ports
2251 participating in a bond.
2254 <column name="other_config" key="lacp-port-priority"
2255 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
2256 The LACP port priority of this <ref table="Interface"/>. In LACP
2257 negotiations <ref table="Interface"/>s with numerically lower
2258 priorities are preferred for aggregation.
2261 <column name="other_config" key="lacp-aggregation-key"
2262 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
2263 The LACP aggregation key of this <ref table="Interface"/>. <ref
2264 table="Interface"/>s with different aggregation keys may not be active
2265 within a given <ref table="Port"/> at the same time.
2269 <group title="Virtual Machine Identifiers">
2271 These key-value pairs specifically apply to an interface that
2272 represents a virtual Ethernet interface connected to a virtual
2273 machine. These key-value pairs should not be present for other types
2274 of interfaces. Keys whose names end in <code>-uuid</code> have
2275 values that uniquely identify the entity in question. For a Citrix
2276 XenServer hypervisor, these values are UUIDs in RFC 4122 format.
2277 Other hypervisors may use other formats.
2280 <column name="external_ids" key="attached-mac">
2281 The MAC address programmed into the ``virtual hardware'' for this
2282 interface, in the form
2283 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>.
2284 For Citrix XenServer, this is the value of the <code>MAC</code> field
2285 in the VIF record for this interface.
2288 <column name="external_ids" key="iface-id">
2289 A system-unique identifier for the interface. On XenServer, this will
2290 commonly be the same as <ref column="external_ids" key="xs-vif-uuid"/>.
2293 <column name="external_ids" key="iface-status"
2294 type='{"type": "string",
2295 "enum": ["set", ["active", "inactive"]]}'>
2297 Hypervisors may sometimes have more than one interface associated
2298 with a given <ref column="external_ids" key="iface-id"/>, only one of
2299 which is actually in use at a given time. For example, in some
2300 circumstances XenServer has both a ``tap'' and a ``vif'' interface
2301 for a single <ref column="external_ids" key="iface-id"/>, but only
2302 uses one of them at a time. A hypervisor that behaves this way must
2303 mark the currently in use interface <code>active</code> and the
2304 others <code>inactive</code>. A hypervisor that never has more than
2305 one interface for a given <ref column="external_ids" key="iface-id"/>
2306 may mark that interface <code>active</code> or omit <ref
2307 column="external_ids" key="iface-status"/> entirely.
2311 During VM migration, a given <ref column="external_ids"
2312 key="iface-id"/> might transiently be marked <code>active</code> on
2313 two different hypervisors. That is, <code>active</code> means that
2314 this <ref column="external_ids" key="iface-id"/> is the active
2315 instance within a single hypervisor, not in a broader scope.
2316 There is one exception: some hypervisors support ``migration'' from a
2317 given hypervisor to itself (most often for test purposes). During
2318 such a ``migration,'' two instances of a single <ref
2319 column="external_ids" key="iface-id"/> might both be briefly marked
2320 <code>active</code> on a single hypervisor.
2324 <column name="external_ids" key="xs-vif-uuid">
2325 The virtual interface associated with this interface.
2328 <column name="external_ids" key="xs-network-uuid">
2329 The virtual network to which this interface is attached.
2332 <column name="external_ids" key="vm-id">
2333 The VM to which this interface belongs. On XenServer, this will be the
2334 same as <ref column="external_ids" key="xs-vm-uuid"/>.
2337 <column name="external_ids" key="xs-vm-uuid">
2338 The VM to which this interface belongs.
2342 <group title="VLAN Splinters">
2344 The ``VLAN splinters'' feature increases Open vSwitch compatibility
2345 with buggy network drivers in old versions of Linux that do not
2346 properly support VLANs when VLAN devices are not used, at some cost
2347 in memory and performance.
2351 When VLAN splinters are enabled on a particular interface, Open vSwitch
2352 creates a VLAN device for each in-use VLAN. For sending traffic tagged
2353 with a VLAN on the interface, it substitutes the VLAN device. Traffic
2354 received on the VLAN device is treated as if it had been received on
2355 the interface on the particular VLAN.
2359 VLAN splinters consider a VLAN to be in use if:
2364 The VLAN is the <ref table="Port" column="tag"/> value in any <ref
2365 table="Port"/> record.
2369 The VLAN is listed within the <ref table="Port" column="trunks"/>
2370 column of the <ref table="Port"/> record of an interface on which
2371 VLAN splinters are enabled.
2373 An empty <ref table="Port" column="trunks"/> does not influence the
2374 in-use VLANs: creating 4,096 VLAN devices is impractical because it
2375 will exceed the current 1,024 port per datapath limit.
2379 An OpenFlow flow within any bridge matches the VLAN.
2384 The same set of in-use VLANs applies to every interface on which VLAN
2385 splinters are enabled. That is, the set is not chosen separately for
2386 each interface but selected once as the union of all in-use VLANs based
2391 It does not make sense to enable VLAN splinters on an interface for an
2392 access port, or on an interface that is not a physical port.
2396 VLAN splinters are deprecated. When broken device drivers are no
2397 longer in widespread use, we will delete this feature.
2400 <column name="other_config" key="enable-vlan-splinters"
2401 type='{"type": "boolean"}'>
2403 Set to <code>true</code> to enable VLAN splinters on this interface.
2404 Defaults to <code>false</code>.
2408 VLAN splinters increase kernel and userspace memory overhead, so do
2409 not use them unless they are needed.
2413 VLAN splinters do not support 802.1p priority tags. Received
2414 priorities will appear to be 0, regardless of their actual values,
2415 and priorities on transmitted packets will also be cleared to 0.
2420 <group title="Common Columns">
2421 The overall purpose of these columns is described under <code>Common
2422 Columns</code> at the beginning of this document.
2424 <column name="other_config"/>
2425 <column name="external_ids"/>
2429 <table name="Flow_Table" title="OpenFlow table configuration">
2430 <p>Configuration for a particular OpenFlow table.</p>
2432 <column name="name">
2433 The table's name. Set this column to change the name that controllers
2434 will receive when they request table statistics, e.g. <code>ovs-ofctl
2435 dump-tables</code>. The name does not affect switch behavior.
2438 <column name="flow_limit">
2439 If set, limits the number of flows that may be added to the table. Open
2440 vSwitch may limit the number of flows in a table for other reasons,
2441 e.g. due to hardware limitations or for resource availability or
2442 performance reasons.
2445 <column name="overflow_policy">
2447 Controls the switch's behavior when an OpenFlow flow table modification
2448 request would add flows in excess of <ref column="flow_limit"/>. The
2449 supported values are:
2453 <dt><code>refuse</code></dt>
2455 Refuse to add the flow or flows. This is also the default policy
2456 when <ref column="overflow_policy"/> is unset.
2459 <dt><code>evict</code></dt>
2461 Delete the flow that will expire soonest. See <ref column="groups"/>
2467 <column name="groups">
2469 When <ref column="overflow_policy"/> is <code>evict</code>, this
2470 controls how flows are chosen for eviction when the flow table would
2471 otherwise exceed <ref column="flow_limit"/> flows. Its value is a set
2472 of NXM fields or sub-fields, each of which takes one of the forms
2473 <code><var>field</var>[]</code> or
2474 <code><var>field</var>[<var>start</var>..<var>end</var>]</code>,
2475 e.g. <code>NXM_OF_IN_PORT[]</code>. Please see
2476 <code>nicira-ext.h</code> for a complete list of NXM field names.
2480 When a flow must be evicted due to overflow, the flow to evict is
2481 chosen through an approximation of the following algorithm:
2486 Divide the flows in the table into groups based on the values of the
2487 specified fields or subfields, so that all of the flows in a given
2488 group have the same values for those fields. If a flow does not
2489 specify a given field, that field's value is treated as 0.
2493 Consider the flows in the largest group, that is, the group that
2494 contains the greatest number of flows. If two or more groups all
2495 have the same largest number of flows, consider the flows in all of
2500 Among the flows under consideration, choose the flow that expires
2501 soonest for eviction.
2506 The eviction process only considers flows that have an idle timeout or
2507 a hard timeout. That is, eviction never deletes permanent flows.
2508 (Permanent flows do count against <ref column="flow_limit"/>.)
2512 Open vSwitch ignores any invalid or unknown field specifications.
2516 When <ref column="overflow_policy"/> is not <code>evict</code>, this
2517 column has no effect.
2522 <table name="QoS" title="Quality of Service configuration">
2523 <p>Quality of Service (QoS) configuration for each Port that
2526 <column name="type">
2527 <p>The type of QoS to implement. The currently defined types are
2530 <dt><code>linux-htb</code></dt>
2532 Linux ``hierarchy token bucket'' classifier. See tc-htb(8) (also at
2533 <code>http://linux.die.net/man/8/tc-htb</code>) and the HTB manual
2534 (<code>http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm</code>)
2535 for information on how this classifier works and how to configure it.
2539 <dt><code>linux-hfsc</code></dt>
2541 Linux "Hierarchical Fair Service Curve" classifier.
2542 See <code>http://linux-ip.net/articles/hfsc.en/</code> for
2543 information on how this classifier works.
2548 <column name="queues">
2549 <p>A map from queue numbers to <ref table="Queue"/> records. The
2550 supported range of queue numbers depend on <ref column="type"/>. The
2551 queue numbers are the same as the <code>queue_id</code> used in
2552 OpenFlow in <code>struct ofp_action_enqueue</code> and other
2556 Queue 0 is the ``default queue.'' It is used by OpenFlow output
2557 actions when no specific queue has been set. When no configuration for
2558 queue 0 is present, it is automatically configured as if a <ref
2559 table="Queue"/> record with empty <ref table="Queue" column="dscp"/>
2560 and <ref table="Queue" column="other_config"/> columns had been
2562 (Before version 1.6, Open vSwitch would leave queue 0 unconfigured in
2563 this case. With some queuing disciplines, this dropped all packets
2564 destined for the default queue.)
2568 <group title="Configuration for linux-htb and linux-hfsc">
2570 The <code>linux-htb</code> and <code>linux-hfsc</code> classes support
2571 the following key-value pair:
2574 <column name="other_config" key="max-rate" type='{"type": "integer"}'>
2575 Maximum rate shared by all queued traffic, in bit/s. Optional. If not
2576 specified, for physical interfaces, the default is the link rate. For
2577 other interfaces or if the link rate cannot be determined, the default
2578 is currently 100 Mbps.
2582 <group title="Common Columns">
2583 The overall purpose of these columns is described under <code>Common
2584 Columns</code> at the beginning of this document.
2586 <column name="other_config"/>
2587 <column name="external_ids"/>
2591 <table name="Queue" title="QoS output queue.">
2592 <p>A configuration for a port output queue, used in configuring Quality of
2593 Service (QoS) features. May be referenced by <ref column="queues"
2594 table="QoS"/> column in <ref table="QoS"/> table.</p>
2596 <column name="dscp">
2597 If set, Open vSwitch will mark all traffic egressing this
2598 <ref table="Queue"/> with the given DSCP bits. Traffic egressing the
2599 default <ref table="Queue"/> is only marked if it was explicitly selected
2600 as the <ref table="Queue"/> at the time the packet was output. If unset,
2601 the DSCP bits of traffic egressing this <ref table="Queue"/> will remain
2605 <group title="Configuration for linux-htb QoS">
2607 <ref table="QoS"/> <ref table="QoS" column="type"/>
2608 <code>linux-htb</code> may use <code>queue_id</code>s less than 61440.
2609 It has the following key-value pairs defined.
2612 <column name="other_config" key="min-rate"
2613 type='{"type": "integer", "minInteger": 1}'>
2614 Minimum guaranteed bandwidth, in bit/s.
2617 <column name="other_config" key="max-rate"
2618 type='{"type": "integer", "minInteger": 1}'>
2619 Maximum allowed bandwidth, in bit/s. Optional. If specified, the
2620 queue's rate will not be allowed to exceed the specified value, even
2621 if excess bandwidth is available. If unspecified, defaults to no
2625 <column name="other_config" key="burst"
2626 type='{"type": "integer", "minInteger": 1}'>
2627 Burst size, in bits. This is the maximum amount of ``credits'' that a
2628 queue can accumulate while it is idle. Optional. Details of the
2629 <code>linux-htb</code> implementation require a minimum burst size, so
2630 a too-small <code>burst</code> will be silently ignored.
2633 <column name="other_config" key="priority"
2634 type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'>
2635 A queue with a smaller <code>priority</code> will receive all the
2636 excess bandwidth that it can use before a queue with a larger value
2637 receives any. Specific priority values are unimportant; only relative
2638 ordering matters. Defaults to 0 if unspecified.
2642 <group title="Configuration for linux-hfsc QoS">
2644 <ref table="QoS"/> <ref table="QoS" column="type"/>
2645 <code>linux-hfsc</code> may use <code>queue_id</code>s less than 61440.
2646 It has the following key-value pairs defined.
2649 <column name="other_config" key="min-rate"
2650 type='{"type": "integer", "minInteger": 1}'>
2651 Minimum guaranteed bandwidth, in bit/s.
2654 <column name="other_config" key="max-rate"
2655 type='{"type": "integer", "minInteger": 1}'>
2656 Maximum allowed bandwidth, in bit/s. Optional. If specified, the
2657 queue's rate will not be allowed to exceed the specified value, even if
2658 excess bandwidth is available. If unspecified, defaults to no
2663 <group title="Common Columns">
2664 The overall purpose of these columns is described under <code>Common
2665 Columns</code> at the beginning of this document.
2667 <column name="other_config"/>
2668 <column name="external_ids"/>
2672 <table name="Mirror" title="Port mirroring.">
2673 <p>A port mirror within a <ref table="Bridge"/>.</p>
2674 <p>A port mirror configures a bridge to send selected frames to special
2675 ``mirrored'' ports, in addition to their normal destinations. Mirroring
2676 traffic may also be referred to as SPAN or RSPAN, depending on how
2677 the mirrored traffic is sent.</p>
2679 <column name="name">
2680 Arbitrary identifier for the <ref table="Mirror"/>.
2683 <group title="Selecting Packets for Mirroring">
2685 To be selected for mirroring, a given packet must enter or leave the
2686 bridge through a selected port and it must also be in one of the
2690 <column name="select_all">
2691 If true, every packet arriving or departing on any port is
2692 selected for mirroring.
2695 <column name="select_dst_port">
2696 Ports on which departing packets are selected for mirroring.
2699 <column name="select_src_port">
2700 Ports on which arriving packets are selected for mirroring.
2703 <column name="select_vlan">
2704 VLANs on which packets are selected for mirroring. An empty set
2705 selects packets on all VLANs.
2709 <group title="Mirroring Destination Configuration">
2711 These columns are mutually exclusive. Exactly one of them must be
2715 <column name="output_port">
2716 <p>Output port for selected packets, if nonempty.</p>
2717 <p>Specifying a port for mirror output reserves that port exclusively
2718 for mirroring. No frames other than those selected for mirroring
2720 will be forwarded to the port, and any frames received on the port
2721 will be discarded.</p>
2723 The output port may be any kind of port supported by Open vSwitch.
2724 It may be, for example, a physical port (sometimes called SPAN) or a
2729 <column name="output_vlan">
2730 <p>Output VLAN for selected packets, if nonempty.</p>
2731 <p>The frames will be sent out all ports that trunk
2732 <ref column="output_vlan"/>, as well as any ports with implicit VLAN
2733 <ref column="output_vlan"/>. When a mirrored frame is sent out a
2734 trunk port, the frame's VLAN tag will be set to
2735 <ref column="output_vlan"/>, replacing any existing tag; when it is
2736 sent out an implicit VLAN port, the frame will not be tagged. This
2737 type of mirroring is sometimes called RSPAN.</p>
2739 See the documentation for
2740 <ref column="other_config" key="forward-bpdu"/> in the
2741 <ref table="Interface"/> table for a list of destination MAC
2742 addresses which will not be mirrored to a VLAN to avoid confusing
2743 switches that interpret the protocols that they represent.
2745 <p><em>Please note:</em> Mirroring to a VLAN can disrupt a network that
2746 contains unmanaged switches. Consider an unmanaged physical switch
2747 with two ports: port 1, connected to an end host, and port 2,
2748 connected to an Open vSwitch configured to mirror received packets
2749 into VLAN 123 on port 2. Suppose that the end host sends a packet on
2750 port 1 that the physical switch forwards to port 2. The Open vSwitch
2751 forwards this packet to its destination and then reflects it back on
2752 port 2 in VLAN 123. This reflected packet causes the unmanaged
2753 physical switch to replace the MAC learning table entry, which
2754 correctly pointed to port 1, with one that incorrectly points to port
2755 2. Afterward, the physical switch will direct packets destined for
2756 the end host to the Open vSwitch on port 2, instead of to the end
2757 host on port 1, disrupting connectivity. If mirroring to a VLAN is
2758 desired in this scenario, then the physical switch must be replaced
2759 by one that learns Ethernet addresses on a per-VLAN basis. In
2760 addition, learning should be disabled on the VLAN containing mirrored
2761 traffic. If this is not done then intermediate switches will learn
2762 the MAC address of each end host from the mirrored traffic. If
2763 packets being sent to that end host are also mirrored, then they will
2764 be dropped since the switch will attempt to send them out the input
2765 port. Disabling learning for the VLAN will cause the switch to
2766 correctly send the packet out all ports configured for that VLAN. If
2767 Open vSwitch is being used as an intermediate switch, learning can be
2768 disabled by adding the mirrored VLAN to <ref column="flood_vlans"/>
2769 in the appropriate <ref table="Bridge"/> table or tables.</p>
2771 Mirroring to a GRE tunnel has fewer caveats than mirroring to a
2772 VLAN and should generally be preferred.
2777 <group title="Statistics: Mirror counters">
2779 Key-value pairs that report mirror statistics.
2781 <column name="statistics" key="tx_packets">
2782 Number of packets transmitted through this mirror.
2784 <column name="statistics" key="tx_bytes">
2785 Number of bytes transmitted through this mirror.
2789 <group title="Common Columns">
2790 The overall purpose of these columns is described under <code>Common
2791 Columns</code> at the beginning of this document.
2793 <column name="external_ids"/>
2797 <table name="Controller" title="OpenFlow controller configuration.">
2798 <p>An OpenFlow controller.</p>
2801 Open vSwitch supports two kinds of OpenFlow controllers:
2805 <dt>Primary controllers</dt>
2808 This is the kind of controller envisioned by the OpenFlow 1.0
2809 specification. Usually, a primary controller implements a network
2810 policy by taking charge of the switch's flow table.
2814 Open vSwitch initiates and maintains persistent connections to
2815 primary controllers, retrying the connection each time it fails or
2816 drops. The <ref table="Bridge" column="fail_mode"/> column in the
2817 <ref table="Bridge"/> table applies to primary controllers.
2821 Open vSwitch permits a bridge to have any number of primary
2822 controllers. When multiple controllers are configured, Open
2823 vSwitch connects to all of them simultaneously. Because
2824 OpenFlow 1.0 does not specify how multiple controllers
2825 coordinate in interacting with a single switch, more than
2826 one primary controller should be specified only if the
2827 controllers are themselves designed to coordinate with each
2828 other. (The Nicira-defined <code>NXT_ROLE</code> OpenFlow
2829 vendor extension may be useful for this.)
2832 <dt>Service controllers</dt>
2835 These kinds of OpenFlow controller connections are intended for
2836 occasional support and maintenance use, e.g. with
2837 <code>ovs-ofctl</code>. Usually a service controller connects only
2838 briefly to inspect or modify some of a switch's state.
2842 Open vSwitch listens for incoming connections from service
2843 controllers. The service controllers initiate and, if necessary,
2844 maintain the connections from their end. The <ref table="Bridge"
2845 column="fail_mode"/> column in the <ref table="Bridge"/> table does
2846 not apply to service controllers.
2850 Open vSwitch supports configuring any number of service controllers.
2856 The <ref column="target"/> determines the type of controller.
2859 <group title="Core Features">
2860 <column name="target">
2861 <p>Connection method for controller.</p>
2863 The following connection methods are currently supported for primary
2867 <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2869 <p>The specified SSL <var>port</var> on the host at the
2870 given <var>ip</var>, which must be expressed as an IP
2871 address (not a DNS name). The <ref table="Open_vSwitch"
2872 column="ssl"/> column in the <ref table="Open_vSwitch"/>
2873 table must point to a valid SSL configuration when this form
2875 <p>If <var>port</var> is not specified, it currently
2876 defaults to 6633. In the future, the default will change to
2877 6653, which is the IANA-defined value.</p>
2878 <p>SSL support is an optional feature that is not always built as
2879 part of Open vSwitch.</p>
2881 <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2883 <p>The specified TCP <var>port</var> on the host at the
2884 given <var>ip</var>, which must be expressed as an IP
2885 address (not a DNS name).</p>
2886 <p>If <var>port</var> is not specified, it currently
2887 defaults to 6633. In the future, the default will change to
2888 6653, which is the IANA-defined value.</p>
2892 The following connection methods are currently supported for service
2896 <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2898 <p> Listens for SSL connections on the specified TCP
2899 <var>port</var>. If <var>ip</var>, which must be expressed
2900 as an IP address (not a DNS name), is specified, then
2901 connections are restricted to the specified local IP
2902 address. The <ref table="Open_vSwitch" column="ssl"/>
2903 column in the <ref table="Open_vSwitch"/> table must point
2904 to a valid SSL configuration when this form is used.</p>
2905 <p>If <var>port</var> is not specified, it currently
2906 defaults to 6633. In the future, the default will change to
2907 6653, which is the IANA-defined value.</p>
2908 <p>SSL support is an optional feature that is not always built as
2909 part of Open vSwitch.</p>
2911 <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2913 <p>Listens for connections on the specified TCP
2914 <var>port</var>. If <var>ip</var>, which must be expressed
2915 as an IP address (not a DNS name), is specified, then
2916 connections are restricted to the specified local IP
2918 <p>If <var>port</var> is not specified, it currently
2919 defaults to 6633. In the future, the default will change to
2920 6653, which is the IANA-defined value.</p>
2923 <p>When multiple controllers are configured for a single bridge, the
2924 <ref column="target"/> values must be unique. Duplicate
2925 <ref column="target"/> values yield unspecified results.</p>
2928 <column name="connection_mode">
2929 <p>If it is specified, this setting must be one of the following
2930 strings that describes how Open vSwitch contacts this OpenFlow
2931 controller over the network:</p>
2934 <dt><code>in-band</code></dt>
2935 <dd>In this mode, this controller's OpenFlow traffic travels over the
2936 bridge associated with the controller. With this setting, Open
2937 vSwitch allows traffic to and from the controller regardless of the
2938 contents of the OpenFlow flow table. (Otherwise, Open vSwitch
2939 would never be able to connect to the controller, because it did
2940 not have a flow to enable it.) This is the most common connection
2941 mode because it is not necessary to maintain two independent
2943 <dt><code>out-of-band</code></dt>
2944 <dd>In this mode, OpenFlow traffic uses a control network separate
2945 from the bridge associated with this controller, that is, the
2946 bridge does not use any of its own network devices to communicate
2947 with the controller. The control network must be configured
2948 separately, before or after <code>ovs-vswitchd</code> is started.
2952 <p>If not specified, the default is implementation-specific.</p>
2956 <group title="Controller Failure Detection and Handling">
2957 <column name="max_backoff">
2958 Maximum number of milliseconds to wait between connection attempts.
2959 Default is implementation-specific.
2962 <column name="inactivity_probe">
2963 Maximum number of milliseconds of idle time on connection to
2964 controller before sending an inactivity probe message. If Open
2965 vSwitch does not communicate with the controller for the specified
2966 number of seconds, it will send a probe. If a response is not
2967 received for the same additional amount of time, Open vSwitch
2968 assumes the connection has been broken and attempts to reconnect.
2969 Default is implementation-specific. A value of 0 disables
2974 <group title="Asynchronous Message Configuration">
2976 OpenFlow switches send certain messages to controllers spontanenously,
2977 that is, not in response to any request from the controller. These
2978 messages are called ``asynchronous messages.'' These columns allow
2979 asynchronous messages to be limited or disabled to ensure the best use
2980 of network resources.
2983 <column name="enable_async_messages">
2984 The OpenFlow protocol enables asynchronous messages at time of
2985 connection establishment, which means that a controller can receive
2986 asynchronous messages, potentially many of them, even if it turns them
2987 off immediately after connecting. Set this column to
2988 <code>false</code> to change Open vSwitch behavior to disable, by
2989 default, all asynchronous messages. The controller can use the
2990 <code>NXT_SET_ASYNC_CONFIG</code> Nicira extension to OpenFlow to turn
2991 on any messages that it does want to receive, if any.
2994 <column name="controller_rate_limit">
2996 The maximum rate at which the switch will forward packets to the
2997 OpenFlow controller, in packets per second. This feature prevents a
2998 single bridge from overwhelming the controller. If not specified,
2999 the default is implementation-specific.
3003 In addition, when a high rate triggers rate-limiting, Open vSwitch
3004 queues controller packets for each port and transmits them to the
3005 controller at the configured rate. The <ref
3006 column="controller_burst_limit"/> value limits the number of queued
3007 packets. Ports on a bridge share the packet queue fairly.
3011 Open vSwitch maintains two such packet rate-limiters per bridge: one
3012 for packets sent up to the controller because they do not correspond
3013 to any flow, and the other for packets sent up to the controller by
3014 request through flow actions. When both rate-limiters are filled with
3015 packets, the actual rate that packets are sent to the controller is
3016 up to twice the specified rate.
3020 <column name="controller_burst_limit">
3021 In conjunction with <ref column="controller_rate_limit"/>,
3022 the maximum number of unused packet credits that the bridge will
3023 allow to accumulate, in packets. If not specified, the default
3024 is implementation-specific.
3028 <group title="Additional In-Band Configuration">
3029 <p>These values are considered only in in-band control mode (see
3030 <ref column="connection_mode"/>).</p>
3032 <p>When multiple controllers are configured on a single bridge, there
3033 should be only one set of unique values in these columns. If different
3034 values are set for these columns in different controllers, the effect
3037 <column name="local_ip">
3038 The IP address to configure on the local port,
3039 e.g. <code>192.168.0.123</code>. If this value is unset, then
3040 <ref column="local_netmask"/> and <ref column="local_gateway"/> are
3044 <column name="local_netmask">
3045 The IP netmask to configure on the local port,
3046 e.g. <code>255.255.255.0</code>. If <ref column="local_ip"/> is set
3047 but this value is unset, then the default is chosen based on whether
3048 the IP address is class A, B, or C.
3051 <column name="local_gateway">
3052 The IP address of the gateway to configure on the local port, as a
3053 string, e.g. <code>192.168.0.1</code>. Leave this column unset if
3054 this network has no gateway.
3058 <group title="Controller Status">
3059 <column name="is_connected">
3060 <code>true</code> if currently connected to this controller,
3061 <code>false</code> otherwise.
3065 type='{"type": "string", "enum": ["set", ["other", "master", "slave"]]}'>
3066 <p>The level of authority this controller has on the associated
3067 bridge. Possible values are:</p>
3069 <dt><code>other</code></dt>
3070 <dd>Allows the controller access to all OpenFlow features.</dd>
3071 <dt><code>master</code></dt>
3072 <dd>Equivalent to <code>other</code>, except that there may be at
3073 most one master controller at a time. When a controller configures
3074 itself as <code>master</code>, any existing master is demoted to
3075 the <code>slave</code>role.</dd>
3076 <dt><code>slave</code></dt>
3077 <dd>Allows the controller read-only access to OpenFlow features.
3078 Attempts to modify the flow table will be rejected with an
3079 error. Slave controllers do not receive OFPT_PACKET_IN or
3080 OFPT_FLOW_REMOVED messages, but they do receive OFPT_PORT_STATUS
3085 <column name="status" key="last_error">
3086 A human-readable description of the last error on the connection
3087 to the controller; i.e. <code>strerror(errno)</code>. This key
3088 will exist only if an error has occurred.
3091 <column name="status" key="state"
3092 type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'>
3094 The state of the connection to the controller:
3097 <dt><code>VOID</code></dt>
3098 <dd>Connection is disabled.</dd>
3100 <dt><code>BACKOFF</code></dt>
3101 <dd>Attempting to reconnect at an increasing period.</dd>
3103 <dt><code>CONNECTING</code></dt>
3104 <dd>Attempting to connect.</dd>
3106 <dt><code>ACTIVE</code></dt>
3107 <dd>Connected, remote host responsive.</dd>
3109 <dt><code>IDLE</code></dt>
3110 <dd>Connection is idle. Waiting for response to keep-alive.</dd>
3113 These values may change in the future. They are provided only for
3118 <column name="status" key="sec_since_connect"
3119 type='{"type": "integer", "minInteger": 0}'>
3120 The amount of time since this controller last successfully connected to
3121 the switch (in seconds). Value is empty if controller has never
3122 successfully connected.
3125 <column name="status" key="sec_since_disconnect"
3126 type='{"type": "integer", "minInteger": 1}'>
3127 The amount of time since this controller last disconnected from
3128 the switch (in seconds). Value is empty if controller has never
3133 <group title="Connection Parameters">
3135 Additional configuration for a connection between the controller
3136 and the Open vSwitch.
3139 <column name="other_config" key="dscp"
3140 type='{"type": "integer"}'>
3141 The Differentiated Service Code Point (DSCP) is specified using 6 bits
3142 in the Type of Service (TOS) field in the IP header. DSCP provides a
3143 mechanism to classify the network traffic and provide Quality of
3144 Service (QoS) on IP networks.
3146 The DSCP value specified here is used when establishing the connection
3147 between the controller and the Open vSwitch. If no value is specified,
3148 a default value of 48 is chosen. Valid DSCP values must be in the
3154 <group title="Common Columns">
3155 The overall purpose of these columns is described under <code>Common
3156 Columns</code> at the beginning of this document.
3158 <column name="external_ids"/>
3159 <column name="other_config"/>
3163 <table name="Manager" title="OVSDB management connection.">
3165 Configuration for a database connection to an Open vSwitch database
3170 This table primarily configures the Open vSwitch database
3171 (<code>ovsdb-server</code>), not the Open vSwitch switch
3172 (<code>ovs-vswitchd</code>). The switch does read the table to determine
3173 what connections should be treated as in-band.
3177 The Open vSwitch database server can initiate and maintain active
3178 connections to remote clients. It can also listen for database
3182 <group title="Core Features">
3183 <column name="target">
3184 <p>Connection method for managers.</p>
3186 The following connection methods are currently supported:
3189 <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
3192 The specified SSL <var>port</var> on the host at the given
3193 <var>ip</var>, which must be expressed as an IP address
3194 (not a DNS name). The <ref table="Open_vSwitch"
3195 column="ssl"/> column in the <ref table="Open_vSwitch"/>
3196 table must point to a valid SSL configuration when this
3200 If <var>port</var> is not specified, it currently defaults
3201 to 6632. In the future, the default will change to 6640,
3202 which is the IANA-defined value.
3205 SSL support is an optional feature that is not always
3206 built as part of Open vSwitch.
3210 <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
3213 The specified TCP <var>port</var> on the host at the given
3214 <var>ip</var>, which must be expressed as an IP address
3218 If <var>port</var> is not specified, it currently defaults
3219 to 6632. In the future, the default will change to 6640,
3220 which is the IANA-defined value.
3223 <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
3226 Listens for SSL connections on the specified TCP
3227 <var>port</var>. Specify 0 for <var>port</var> to have
3228 the kernel automatically choose an available port. If
3229 <var>ip</var>, which must be expressed as an IP address
3230 (not a DNS name), is specified, then connections are
3231 restricted to the specified local IP address. The <ref
3232 table="Open_vSwitch" column="ssl"/> column in the <ref
3233 table="Open_vSwitch"/> table must point to a valid SSL
3234 configuration when this form is used.
3237 If <var>port</var> is not specified, it currently defaults
3238 to 6632. In the future, the default will change to 6640,
3239 which is the IANA-defined value.
3242 SSL support is an optional feature that is not always built as
3243 part of Open vSwitch.
3246 <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
3249 Listens for connections on the specified TCP
3250 <var>port</var>. Specify 0 for <var>port</var> to have
3251 the kernel automatically choose an available port. If
3252 <var>ip</var>, which must be expressed as an IP address
3253 (not a DNS name), is specified, then connections are
3254 restricted to the specified local IP address.
3257 If <var>port</var> is not specified, it currently defaults
3258 to 6632. In the future, the default will change to 6640,
3259 which is the IANA-defined value.
3263 <p>When multiple managers are configured, the <ref column="target"/>
3264 values must be unique. Duplicate <ref column="target"/> values yield
3265 unspecified results.</p>
3268 <column name="connection_mode">
3270 If it is specified, this setting must be one of the following strings
3271 that describes how Open vSwitch contacts this OVSDB client over the
3276 <dt><code>in-band</code></dt>
3278 In this mode, this connection's traffic travels over a bridge
3279 managed by Open vSwitch. With this setting, Open vSwitch allows
3280 traffic to and from the client regardless of the contents of the
3281 OpenFlow flow table. (Otherwise, Open vSwitch would never be able
3282 to connect to the client, because it did not have a flow to enable
3283 it.) This is the most common connection mode because it is not
3284 necessary to maintain two independent networks.
3286 <dt><code>out-of-band</code></dt>
3288 In this mode, the client's traffic uses a control network separate
3289 from that managed by Open vSwitch, that is, Open vSwitch does not
3290 use any of its own network devices to communicate with the client.
3291 The control network must be configured separately, before or after
3292 <code>ovs-vswitchd</code> is started.
3297 If not specified, the default is implementation-specific.
3302 <group title="Client Failure Detection and Handling">
3303 <column name="max_backoff">
3304 Maximum number of milliseconds to wait between connection attempts.
3305 Default is implementation-specific.
3308 <column name="inactivity_probe">
3309 Maximum number of milliseconds of idle time on connection to the client
3310 before sending an inactivity probe message. If Open vSwitch does not
3311 communicate with the client for the specified number of seconds, it
3312 will send a probe. If a response is not received for the same
3313 additional amount of time, Open vSwitch assumes the connection has been
3314 broken and attempts to reconnect. Default is implementation-specific.
3315 A value of 0 disables inactivity probes.
3319 <group title="Status">
3320 <column name="is_connected">
3321 <code>true</code> if currently connected to this manager,
3322 <code>false</code> otherwise.
3325 <column name="status" key="last_error">
3326 A human-readable description of the last error on the connection
3327 to the manager; i.e. <code>strerror(errno)</code>. This key
3328 will exist only if an error has occurred.
3331 <column name="status" key="state"
3332 type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'>
3334 The state of the connection to the manager:
3337 <dt><code>VOID</code></dt>
3338 <dd>Connection is disabled.</dd>
3340 <dt><code>BACKOFF</code></dt>
3341 <dd>Attempting to reconnect at an increasing period.</dd>
3343 <dt><code>CONNECTING</code></dt>
3344 <dd>Attempting to connect.</dd>
3346 <dt><code>ACTIVE</code></dt>
3347 <dd>Connected, remote host responsive.</dd>
3349 <dt><code>IDLE</code></dt>
3350 <dd>Connection is idle. Waiting for response to keep-alive.</dd>
3353 These values may change in the future. They are provided only for
3358 <column name="status" key="sec_since_connect"
3359 type='{"type": "integer", "minInteger": 0}'>
3360 The amount of time since this manager last successfully connected
3361 to the database (in seconds). Value is empty if manager has never
3362 successfully connected.
3365 <column name="status" key="sec_since_disconnect"
3366 type='{"type": "integer", "minInteger": 0}'>
3367 The amount of time since this manager last disconnected from the
3368 database (in seconds). Value is empty if manager has never
3372 <column name="status" key="locks_held">
3373 Space-separated list of the names of OVSDB locks that the connection
3374 holds. Omitted if the connection does not hold any locks.
3377 <column name="status" key="locks_waiting">
3378 Space-separated list of the names of OVSDB locks that the connection is
3379 currently waiting to acquire. Omitted if the connection is not waiting
3383 <column name="status" key="locks_lost">
3384 Space-separated list of the names of OVSDB locks that the connection
3385 has had stolen by another OVSDB client. Omitted if no locks have been
3386 stolen from this connection.
3389 <column name="status" key="n_connections"
3390 type='{"type": "integer", "minInteger": 2}'>
3392 When <ref column="target"/> specifies a connection method that
3393 listens for inbound connections (e.g. <code>ptcp:</code> or
3394 <code>pssl:</code>) and more than one connection is actually active,
3395 the value is the number of active connections. Otherwise, this
3396 key-value pair is omitted.
3399 When multiple connections are active, status columns and key-value
3400 pairs (other than this one) report the status of one arbitrarily
3405 <column name="status" key="bound_port" type='{"type": "integer"}'>
3406 When <ref column="target"/> is <code>ptcp:</code> or
3407 <code>pssl:</code>, this is the TCP port on which the OVSDB server is
3408 listening. (This is is particularly useful when <ref
3409 column="target"/> specifies a port of 0, allowing the kernel to
3410 choose any available port.)
3414 <group title="Connection Parameters">
3416 Additional configuration for a connection between the manager
3417 and the Open vSwitch Database.
3420 <column name="other_config" key="dscp"
3421 type='{"type": "integer"}'>
3422 The Differentiated Service Code Point (DSCP) is specified using 6 bits
3423 in the Type of Service (TOS) field in the IP header. DSCP provides a
3424 mechanism to classify the network traffic and provide Quality of
3425 Service (QoS) on IP networks.
3427 The DSCP value specified here is used when establishing the connection
3428 between the manager and the Open vSwitch. If no value is specified, a
3429 default value of 48 is chosen. Valid DSCP values must be in the range
3434 <group title="Common Columns">
3435 The overall purpose of these columns is described under <code>Common
3436 Columns</code> at the beginning of this document.
3438 <column name="external_ids"/>
3439 <column name="other_config"/>
3443 <table name="NetFlow">
3444 A NetFlow target. NetFlow is a protocol that exports a number of
3445 details about terminating IP flows, such as the principals involved
3448 <column name="targets">
3449 NetFlow targets in the form
3450 <code><var>ip</var>:<var>port</var></code>. The <var>ip</var>
3451 must be specified numerically, not as a DNS name.
3454 <column name="engine_id">
3455 Engine ID to use in NetFlow messages. Defaults to datapath index
3459 <column name="engine_type">
3460 Engine type to use in NetFlow messages. Defaults to datapath
3461 index if not specified.
3464 <column name="active_timeout">
3465 The interval at which NetFlow records are sent for flows that are
3466 still active, in seconds. A value of <code>0</code> requests the
3467 default timeout (currently 600 seconds); a value of <code>-1</code>
3468 disables active timeouts.
3471 <column name="add_id_to_interface">
3472 <p>If this column's value is <code>false</code>, the ingress and egress
3473 interface fields of NetFlow flow records are derived from OpenFlow port
3474 numbers. When it is <code>true</code>, the 7 most significant bits of
3475 these fields will be replaced by the least significant 7 bits of the
3476 engine id. This is useful because many NetFlow collectors do not
3477 expect multiple switches to be sending messages from the same host, so
3478 they do not store the engine information which could be used to
3479 disambiguate the traffic.</p>
3480 <p>When this option is enabled, a maximum of 508 ports are supported.</p>
3483 <group title="Common Columns">
3484 The overall purpose of these columns is described under <code>Common
3485 Columns</code> at the beginning of this document.
3487 <column name="external_ids"/>
3492 SSL configuration for an Open_vSwitch.
3494 <column name="private_key">
3495 Name of a PEM file containing the private key used as the switch's
3496 identity for SSL connections to the controller.
3499 <column name="certificate">
3500 Name of a PEM file containing a certificate, signed by the
3501 certificate authority (CA) used by the controller and manager,
3502 that certifies the switch's private key, identifying a trustworthy
3506 <column name="ca_cert">
3507 Name of a PEM file containing the CA certificate used to verify
3508 that the switch is connected to a trustworthy controller.
3511 <column name="bootstrap_ca_cert">
3512 If set to <code>true</code>, then Open vSwitch will attempt to
3513 obtain the CA certificate from the controller on its first SSL
3514 connection and save it to the named PEM file. If it is successful,
3515 it will immediately drop the connection and reconnect, and from then
3516 on all SSL connections must be authenticated by a certificate signed
3517 by the CA certificate thus obtained. <em>This option exposes the
3518 SSL connection to a man-in-the-middle attack obtaining the initial
3519 CA certificate.</em> It may still be useful for bootstrapping.
3522 <group title="Common Columns">
3523 The overall purpose of these columns is described under <code>Common
3524 Columns</code> at the beginning of this document.
3526 <column name="external_ids"/>
3530 <table name="sFlow">
3531 <p>A set of sFlow(R) targets. sFlow is a protocol for remote
3532 monitoring of switches.</p>
3534 <column name="agent">
3535 Name of the network device whose IP address should be reported as the
3536 ``agent address'' to collectors. If not specified, the agent device is
3537 figured from the first target address and the routing table. If the
3538 routing table does not contain a route to the target, the IP address
3539 defaults to the <ref table="Controller" column="local_ip"/> in the
3540 collector's <ref table="Controller"/>. If an agent IP address cannot be
3541 determined any of these ways, sFlow is disabled.
3544 <column name="header">
3545 Number of bytes of a sampled packet to send to the collector.
3546 If not specified, the default is 128 bytes.
3549 <column name="polling">
3550 Polling rate in seconds to send port statistics to the collector.
3551 If not specified, defaults to 30 seconds.
3554 <column name="sampling">
3555 Rate at which packets should be sampled and sent to the collector.
3556 If not specified, defaults to 400, which means one out of 400
3557 packets, on average, will be sent to the collector.
3560 <column name="targets">
3561 sFlow targets in the form
3562 <code><var>ip</var>:<var>port</var></code>.
3565 <group title="Common Columns">
3566 The overall purpose of these columns is described under <code>Common
3567 Columns</code> at the beginning of this document.
3569 <column name="external_ids"/>
3573 <table name="IPFIX">
3574 <p>A set of IPFIX collectors. IPFIX is a protocol that exports a
3575 number of details about flows.</p>
3577 <column name="targets">
3578 IPFIX target collectors in the form
3579 <code><var>ip</var>:<var>port</var></code>.
3582 <column name="sampling">
3583 For per-bridge packet sampling, i.e. when this row is referenced
3584 from a <ref table="Bridge"/>, the rate at which packets should
3585 be sampled and sent to each target collector. If not specified,
3586 defaults to 400, which means one out of 400 packets, on average,
3587 will be sent to each target collector. Ignored for per-flow
3588 sampling, i.e. when this row is referenced from a <ref
3589 table="Flow_Sample_Collector_Set"/>.
3592 <column name="obs_domain_id">
3593 For per-bridge packet sampling, i.e. when this row is referenced
3594 from a <ref table="Bridge"/>, the IPFIX Observation Domain ID
3595 sent in each IPFIX packet. If not specified, defaults to 0.
3596 Ignored for per-flow sampling, i.e. when this row is referenced
3597 from a <ref table="Flow_Sample_Collector_Set"/>.
3600 <column name="obs_point_id">
3601 For per-bridge packet sampling, i.e. when this row is referenced
3602 from a <ref table="Bridge"/>, the IPFIX Observation Point ID
3603 sent in each IPFIX flow record. If not specified, defaults to
3604 0. Ignored for per-flow sampling, i.e. when this row is
3605 referenced from a <ref table="Flow_Sample_Collector_Set"/>.
3608 <column name="cache_active_timeout">
3609 The maximum period in seconds for which an IPFIX flow record is
3610 cached and aggregated before being sent. If not specified,
3611 defaults to 0. If 0, caching is disabled.
3614 <column name="cache_max_flows">
3615 The maximum number of IPFIX flow records that can be cached at a
3616 time. If not specified, defaults to 0. If 0, caching is
3620 <group title="Common Columns">
3621 The overall purpose of these columns is described under <code>Common
3622 Columns</code> at the beginning of this document.
3624 <column name="external_ids"/>
3628 <table name="Flow_Sample_Collector_Set">
3629 <p>A set of IPFIX collectors of packet samples generated by
3630 OpenFlow <code>sample</code> actions.</p>
3633 The ID of this collector set, unique among the bridge's
3634 collector sets, to be used as the <code>collector_set_id</code>
3635 in OpenFlow <code>sample</code> actions.
3638 <column name="bridge">
3639 The bridge into which OpenFlow <code>sample</code> actions can
3640 be added to send packet samples to this set of IPFIX collectors.
3643 <column name="ipfix">
3644 Configuration of the set of IPFIX collectors to send one flow
3645 record per sampled packet to.
3648 <group title="Common Columns">
3649 The overall purpose of these columns is described under <code>Common
3650 Columns</code> at the beginning of this document.
3652 <column name="external_ids"/>