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>.
76 <group title="Status">
77 <column name="next_cfg">
78 Sequence number for client to increment. When a client modifies
79 any part of the database configuration and wishes to wait for
80 Open vSwitch to finish applying the changes, it may increment
84 <column name="cur_cfg">
85 Sequence number that Open vSwitch sets to the current value of
86 <ref column="next_cfg"/> after it finishes applying a set of
87 configuration changes.
90 <group title="Statistics">
92 The <code>statistics</code> column contains key-value pairs that
93 report statistics about a system running an Open vSwitch. These are
94 updated periodically (currently, every 5 seconds). Key-value pairs
95 that cannot be determined or that do not apply to a platform are
99 <column name="other_config" key="enable-statistics"
100 type='{"type": "boolean"}'>
101 Statistics are disabled by default to avoid overhead in the common
102 case when statistics gathering is not useful. Set this value to
103 <code>true</code> to enable populating the <ref column="statistics"/>
104 column or to <code>false</code> to explicitly disable it.
107 <column name="statistics" key="cpu"
108 type='{"type": "integer", "minInteger": 1}'>
110 Number of CPU processors, threads, or cores currently online and
111 available to the operating system on which Open vSwitch is running,
112 as an integer. This may be less than the number installed, if some
113 are not online or if they are not available to the operating
117 Open vSwitch userspace processes are not multithreaded, but the
118 Linux kernel-based datapath is.
122 <column name="statistics" key="load_average">
123 A comma-separated list of three floating-point numbers,
124 representing the system load average over the last 1, 5, and 15
125 minutes, respectively.
128 <column name="statistics" key="memory">
130 A comma-separated list of integers, each of which represents a
131 quantity of memory in kilobytes that describes the operating
132 system on which Open vSwitch is running. In respective order,
137 <li>Total amount of RAM allocated to the OS.</li>
138 <li>RAM allocated to the OS that is in use.</li>
139 <li>RAM that can be flushed out to disk or otherwise discarded
140 if that space is needed for another purpose. This number is
141 necessarily less than or equal to the previous value.</li>
142 <li>Total disk space allocated for swap.</li>
143 <li>Swap space currently in use.</li>
147 On Linux, all five values can be determined and are included. On
148 other operating systems, only the first two values can be
149 determined, so the list will only have two values.
153 <column name="statistics" key="process_NAME">
155 One such key-value pair, with <code>NAME</code> replaced by
156 a process name, will exist for each running Open vSwitch
157 daemon process, with <var>name</var> replaced by the
158 daemon's name (e.g. <code>process_ovs-vswitchd</code>). The
159 value is a comma-separated list of integers. The integers
160 represent the following, with memory measured in kilobytes
161 and durations in milliseconds:
165 <li>The process's virtual memory size.</li>
166 <li>The process's resident set size.</li>
167 <li>The amount of user and system CPU time consumed by the
169 <li>The number of times that the process has crashed and been
170 automatically restarted by the monitor.</li>
171 <li>The duration since the process was started.</li>
172 <li>The duration for which the process has been running.</li>
176 The interpretation of some of these values depends on whether the
177 process was started with the <option>--monitor</option>. If it
178 was not, then the crash count will always be 0 and the two
179 durations will always be the same. If <option>--monitor</option>
180 was given, then the crash count may be positive; if it is, the
181 latter duration is the amount of time since the most recent crash
186 There will be one key-value pair for each file in Open vSwitch's
187 ``run directory'' (usually <code>/var/run/openvswitch</code>)
188 whose name ends in <code>.pid</code>, whose contents are a
189 process ID, and which is locked by a running process. The
190 <var>name</var> is taken from the pidfile's name.
194 Currently Open vSwitch is only able to obtain all of the above
195 detail on Linux systems. On other systems, the same key-value
196 pairs will be present but the values will always be the empty
201 <column name="statistics" key="file_systems">
203 A space-separated list of information on local, writable file
204 systems. Each item in the list describes one file system and
205 consists in turn of a comma-separated list of the following:
209 <li>Mount point, e.g. <code>/</code> or <code>/var/log</code>.
210 Any spaces or commas in the mount point are replaced by
212 <li>Total size, in kilobytes, as an integer.</li>
213 <li>Amount of storage in use, in kilobytes, as an integer.</li>
217 This key-value pair is omitted if there are no local, writable
218 file systems or if Open vSwitch cannot obtain the needed
225 <group title="Version Reporting">
227 These columns report the types and versions of the hardware and
228 software running Open vSwitch. We recommend in general that software
229 should test whether specific features are supported instead of relying
230 on version number checks. These values are primarily intended for
231 reporting to human administrators.
234 <column name="ovs_version">
235 The Open vSwitch version number, e.g. <code>1.1.0</code>.
238 <column name="db_version">
240 The database schema version number in the form
241 <code><var>major</var>.<var>minor</var>.<var>tweak</var></code>,
242 e.g. <code>1.2.3</code>. Whenever the database schema is changed in
243 a non-backward compatible way (e.g. deleting a column or a table),
244 <var>major</var> is incremented. When the database schema is changed
245 in a backward compatible way (e.g. adding a new column),
246 <var>minor</var> is incremented. When the database schema is changed
247 cosmetically (e.g. reindenting its syntax), <var>tweak</var> is
252 The schema version is part of the database schema, so it can also be
253 retrieved by fetching the schema using the Open vSwitch database
258 <column name="system_type">
260 An identifier for the type of system on top of which Open vSwitch
261 runs, e.g. <code>XenServer</code> or <code>KVM</code>.
264 System integrators are responsible for choosing and setting an
265 appropriate value for this column.
269 <column name="system_version">
271 The version of the system identified by <ref column="system_type"/>,
272 e.g. <code>5.6.100-39265p</code> on XenServer 5.6.100 build 39265.
275 System integrators are responsible for choosing and setting an
276 appropriate value for this column.
282 <group title="Database Configuration">
284 These columns primarily configure the Open vSwitch database
285 (<code>ovsdb-server</code>), not the Open vSwitch switch
286 (<code>ovs-vswitchd</code>). The OVSDB database also uses the <ref
287 column="ssl"/> settings.
291 The Open vSwitch switch does read the database configuration to
292 determine remote IP addresses to which in-band control should apply.
295 <column name="manager_options">
296 Database clients to which the Open vSwitch database server should
297 connect or to which it should listen, along with options for how these
298 connection should be configured. See the <ref table="Manager"/> table
299 for more information.
303 <group title="Common Columns">
304 The overall purpose of these columns is described under <code>Common
305 Columns</code> at the beginning of this document.
307 <column name="other_config"/>
308 <column name="external_ids"/>
312 <table name="Bridge">
314 Configuration for a bridge within an
315 <ref table="Open_vSwitch"/>.
318 A <ref table="Bridge"/> record represents an Ethernet switch with one or
319 more ``ports,'' which are the <ref table="Port"/> records pointed to by
320 the <ref table="Bridge"/>'s <ref column="ports"/> column.
323 <group title="Core Features">
325 Bridge identifier. Should be alphanumeric and no more than about 8
326 bytes long. Must be unique among the names of ports, interfaces, and
330 <column name="ports">
331 Ports included in the bridge.
334 <column name="mirrors">
335 Port mirroring configuration.
338 <column name="netflow">
339 NetFlow configuration.
342 <column name="sflow">
346 <column name="flood_vlans">
348 VLAN IDs of VLANs on which MAC address learning should be disabled,
349 so that packets are flooded instead of being sent to specific ports
350 that are believed to contain packets' destination MACs. This should
351 ordinarily be used to disable MAC learning on VLANs used for
352 mirroring (RSPAN VLANs). It may also be useful for debugging.
355 SLB bonding (see the <ref table="Port" column="bond_mode"/> column in
356 the <ref table="Port"/> table) is incompatible with
357 <code>flood_vlans</code>. Consider using another bonding mode or
358 a different type of mirror instead.
363 <group title="OpenFlow Configuration">
364 <column name="controller">
366 OpenFlow controller set. If unset, then no OpenFlow controllers
371 If there are primary controllers, removing all of them clears the
372 flow table. If there are no primary controllers, adding one also
373 clears the flow table. Other changes to the set of controllers, such
374 as adding or removing a service controller, adding another primary
375 controller to supplement an existing primary controller, or removing
376 only one of two primary controllers, have no effect on the flow
381 <column name="flow_tables">
382 Configuration for OpenFlow tables. Each pair maps from an OpenFlow
383 table ID to configuration for that table.
386 <column name="fail_mode">
387 <p>When a controller is configured, it is, ordinarily, responsible
388 for setting up all flows on the switch. Thus, if the connection to
389 the controller fails, no new network connections can be set up.
390 If the connection to the controller stays down long enough,
391 no packets can pass through the switch at all. This setting
392 determines the switch's response to such a situation. It may be set
393 to one of the following:
395 <dt><code>standalone</code></dt>
396 <dd>If no message is received from the controller for three
397 times the inactivity probe interval
398 (see <ref column="inactivity_probe"/>), then Open vSwitch
399 will take over responsibility for setting up flows. In
400 this mode, Open vSwitch causes the bridge to act like an
401 ordinary MAC-learning switch. Open vSwitch will continue
402 to retry connecting to the controller in the background
403 and, when the connection succeeds, it will discontinue its
404 standalone behavior.</dd>
405 <dt><code>secure</code></dt>
406 <dd>Open vSwitch will not set up flows on its own when the
407 controller connection fails or when no controllers are
408 defined. The bridge will continue to retry connecting to
409 any defined controllers forever.</dd>
412 <p>If this value is unset, the default is implementation-specific.</p>
413 <p>When more than one controller is configured,
414 <ref column="fail_mode"/> is considered only when none of the
415 configured controllers can be contacted.</p>
417 Changing <ref column="fail_mode"/> when no primary controllers are
418 configured clears the flow table.
422 <column name="datapath_id">
423 Reports the OpenFlow datapath ID in use. Exactly 16 hex digits.
424 (Setting this column has no useful effect. Set <ref
425 column="other-config" key="datapath-id"/> instead.)
428 <column name="other_config" key="datapath-id">
429 Exactly 16 hex digits to set the OpenFlow datapath ID to a specific
430 value. May not be all-zero.
433 <column name="other_config" key="disable-in-band"
434 type='{"type": "boolean"}'>
435 If set to <code>true</code>, disable in-band control on the bridge
436 regardless of controller and manager settings.
439 <column name="other_config" key="in-band-queue"
440 type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'>
441 A queue ID as a nonnegative integer. This sets the OpenFlow queue ID
442 that will be used by flows set up by in-band control on this bridge.
443 If unset, or if the port used by an in-band control flow does not have
444 QoS configured, or if the port does not have a queue with the specified
445 ID, the default queue is used instead.
449 <group title="Spanning Tree Configuration">
450 The IEEE 802.1D Spanning Tree Protocol (STP) is a network protocol
451 that ensures loop-free topologies. It allows redundant links to
452 be included in the network to provide automatic backup paths if
453 the active links fails.
455 <column name="stp_enable">
456 Enable spanning tree on the bridge. By default, STP is disabled
457 on bridges. Bond, internal, and mirror ports are not supported
458 and will not participate in the spanning tree.
461 <column name="other_config" key="stp-system-id">
462 The bridge's STP identifier (the lower 48 bits of the bridge-id)
464 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>.
465 By default, the identifier is the MAC address of the bridge.
468 <column name="other_config" key="stp-priority"
469 type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'>
470 The bridge's relative priority value for determining the root
471 bridge (the upper 16 bits of the bridge-id). A bridge with the
472 lowest bridge-id is elected the root. By default, the priority
476 <column name="other_config" key="stp-hello-time"
477 type='{"type": "integer", "minInteger": 1, "maxInteger": 10}'>
478 The interval between transmissions of hello messages by
479 designated ports, in seconds. By default the hello interval is
483 <column name="other_config" key="stp-max-age"
484 type='{"type": "integer", "minInteger": 6, "maxInteger": 40}'>
485 The maximum age of the information transmitted by the bridge
486 when it is the root bridge, in seconds. By default, the maximum
490 <column name="other_config" key="stp-forward-delay"
491 type='{"type": "integer", "minInteger": 4, "maxInteger": 30}'>
492 The delay to wait between transitioning root and designated
493 ports to <code>forwarding</code>, in seconds. By default, the
494 forwarding delay is 15 seconds.
498 <group title="Other Features">
499 <column name="datapath_type">
500 Name of datapath provider. The kernel datapath has
501 type <code>system</code>. The userspace datapath has
502 type <code>netdev</code>.
505 <column name="external_ids" key="bridge-id">
506 A unique identifier of the bridge. On Citrix XenServer this will
507 commonly be the same as
508 <ref column="external_ids" key="xs-network-uuids"/>.
511 <column name="external_ids" key="xs-network-uuids">
512 Semicolon-delimited set of universally unique identifier(s) for the
513 network with which this bridge is associated on a Citrix XenServer
514 host. The network identifiers are RFC 4122 UUIDs as displayed by,
515 e.g., <code>xe network-list</code>.
518 <column name="other_config" key="hwaddr">
519 An Ethernet address in the form
520 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>
521 to set the hardware address of the local port and influence the
525 <column name="other_config" key="flow-eviction-threshold"
526 type='{"type": "integer", "minInteger": 0}'>
528 A number of flows as a nonnegative integer. This sets number of
529 flows at which eviction from the kernel flow table will be triggered.
530 If there are a large number of flows then increasing this value to
531 around the number of flows present can result in reduced CPU usage
535 The default is 1000. Values below 100 will be rounded up to 100.
539 <column name="other_config" key="forward-bpdu"
540 type='{"type": "boolean"}'>
541 Option to allow forwarding of BPDU frames when NORMAL action is
542 invoked. Frames with reserved Ethernet addresses (e.g. STP
543 BPDU) will be forwarded when this option is enabled and the
544 switch is not providing that functionality. If STP is enabled
545 on the port, STP BPDUs will never be forwarded. If the Open
546 vSwitch bridge is used to connect different Ethernet networks,
547 and if Open vSwitch node does not run STP, then this option
548 should be enabled. Default is disabled, set to
549 <code>true</code> to enable.
552 <column name="other_config" key="mac-aging-time"
553 type='{"type": "integer", "minInteger": 1}'>
555 The maximum number of seconds to retain a MAC learning entry for
556 which no packets have been seen. The default is currently 300
557 seconds (5 minutes). The value, if specified, is forced into a
558 reasonable range, currently 15 to 3600 seconds.
562 A short MAC aging time allows a network to more quickly detect that a
563 host is no longer connected to a switch port. However, it also makes
564 it more likely that packets will be flooded unnecessarily, when they
565 are addressed to a connected host that rarely transmits packets. To
566 reduce the incidence of unnecessary flooding, use a MAC aging time
567 longer than the maximum interval at which a host will ordinarily
573 <group title="Bridge Status">
575 Status information about bridges.
577 <column name="status">
578 Key-value pairs that report bridge status.
580 <column name="status" key="stp_bridge_id">
582 The bridge-id (in hex) used in spanning tree advertisements.
583 Configuring the bridge-id is described in the
584 <code>stp-system-id</code> and <code>stp-priority</code> keys
585 of the <code>other_config</code> section earlier.
588 <column name="status" key="stp_designated_root">
590 The designated root (in hex) for this spanning tree.
593 <column name="status" key="stp_root_path_cost">
595 The path cost of reaching the designated bridge. A lower
601 <group title="Common Columns">
602 The overall purpose of these columns is described under <code>Common
603 Columns</code> at the beginning of this document.
605 <column name="other_config"/>
606 <column name="external_ids"/>
610 <table name="Port" table="Port or bond configuration.">
611 <p>A port within a <ref table="Bridge"/>.</p>
612 <p>Most commonly, a port has exactly one ``interface,'' pointed to by its
613 <ref column="interfaces"/> column. Such a port logically
614 corresponds to a port on a physical Ethernet switch. A port
615 with more than one interface is a ``bonded port'' (see
616 <ref group="Bonding Configuration"/>).</p>
617 <p>Some properties that one might think as belonging to a port are actually
618 part of the port's <ref table="Interface"/> members.</p>
621 Port name. Should be alphanumeric and no more than about 8
622 bytes long. May be the same as the interface name, for
623 non-bonded ports. Must otherwise be unique among the names of
624 ports, interfaces, and bridges on a host.
627 <column name="interfaces">
628 The port's interfaces. If there is more than one, this is a
632 <group title="VLAN Configuration">
633 <p>Bridge ports support the following types of VLAN configuration:</p>
638 A trunk port carries packets on one or more specified VLANs
639 specified in the <ref column="trunks"/> column (often, on every
640 VLAN). A packet that ingresses on a trunk port is in the VLAN
641 specified in its 802.1Q header, or VLAN 0 if the packet has no
642 802.1Q header. A packet that egresses through a trunk port will
643 have an 802.1Q header if it has a nonzero VLAN ID.
647 Any packet that ingresses on a trunk port tagged with a VLAN that
648 the port does not trunk is dropped.
655 An access port carries packets on exactly one VLAN specified in the
656 <ref column="tag"/> column. Packets egressing on an access port
657 have no 802.1Q header.
661 Any packet with an 802.1Q header with a nonzero VLAN ID that
662 ingresses on an access port is dropped, regardless of whether the
663 VLAN ID in the header is the access port's VLAN ID.
667 <dt>native-tagged</dt>
669 A native-tagged port resembles a trunk port, with the exception that
670 a packet without an 802.1Q header that ingresses on a native-tagged
671 port is in the ``native VLAN'' (specified in the <ref column="tag"/>
675 <dt>native-untagged</dt>
677 A native-untagged port resembles a native-tagged port, with the
678 exception that a packet that egresses on a native-untagged port in
679 the native VLAN will not have an 802.1Q header.
683 A packet will only egress through bridge ports that carry the VLAN of
684 the packet, as described by the rules above.
687 <column name="vlan_mode">
689 The VLAN mode of the port, as described above. When this column is
690 empty, a default mode is selected as follows:
694 If <ref column="tag"/> contains a value, the port is an access
695 port. The <ref column="trunks"/> column should be empty.
698 Otherwise, the port is a trunk port. The <ref column="trunks"/>
699 column value is honored if it is present.
706 For an access port, the port's implicitly tagged VLAN. For a
707 native-tagged or native-untagged port, the port's native VLAN. Must
708 be empty if this is a trunk port.
712 <column name="trunks">
714 For a trunk, native-tagged, or native-untagged port, the 802.1Q VLAN
715 or VLANs that this port trunks; if it is empty, then the port trunks
716 all VLANs. Must be empty if this is an access port.
719 A native-tagged or native-untagged port always trunks its native
720 VLAN, regardless of whether <ref column="trunks"/> includes that
725 <column name="other_config" key="priority-tags"
726 type='{"type": "boolean"}'>
728 An 802.1Q header contains two important pieces of information: a VLAN
729 ID and a priority. A frame with a zero VLAN ID, called a
730 ``priority-tagged'' frame, is supposed to be treated the same way as
731 a frame without an 802.1Q header at all (except for the priority).
735 However, some network elements ignore any frame that has 802.1Q
736 header at all, even when the VLAN ID is zero. Therefore, by default
737 Open vSwitch does not output priority-tagged frames, instead omitting
738 the 802.1Q header entirely if the VLAN ID is zero. Set this key to
739 <code>true</code> to enable priority-tagged frames on a port.
743 Regardless of this setting, Open vSwitch omits the 802.1Q header on
744 output if both the VLAN ID and priority would be zero.
748 All frames output to native-tagged ports have a nonzero VLAN ID, so
749 this setting is not meaningful on native-tagged ports.
754 <group title="Bonding Configuration">
755 <p>A port that has more than one interface is a ``bonded port.'' Bonding
756 allows for load balancing and fail-over. Some kinds of bonding will
757 work with any kind of upstream switch:</p>
760 <dt><code>balance-slb</code></dt>
762 Balances flows among slaves based on source MAC address and output
763 VLAN, with periodic rebalancing as traffic patterns change.
766 <dt><code>active-backup</code></dt>
768 Assigns all flows to one slave, failing over to a backup slave when
769 the active slave is disabled.
774 The following modes require the upstream switch to support 802.3ad with
775 successful LACP negotiation:
779 <dt><code>balance-tcp</code></dt>
781 Balances flows among slaves based on L2, L3, and L4 protocol
782 information such as destination MAC address, IP address, and TCP
786 <dt><code>stable</code></dt>
788 <p>Attempts to always assign a given flow to the same slave
789 consistently. In an effort to maintain stability, no load
790 balancing is done. Uses a similar hashing strategy to
791 <code>balance-tcp</code>, always taking into account L3 and L4
792 fields even if LACP negotiations are unsuccessful. </p>
793 <p>Slave selection decisions are made based on <ref table="Interface"
794 column="other_config" key="bond-stable-id"/> if set. Otherwise,
795 OpenFlow port number is used. Decisions are consistent across all
796 <code>ovs-vswitchd</code> instances with equivalent
797 <ref table="Interface" column="other_config" key="bond-stable-id"/>
802 <p>These columns apply only to bonded ports. Their values are
803 otherwise ignored.</p>
805 <column name="bond_mode">
806 <p>The type of bonding used for a bonded port. Defaults to
807 <code>active-backup</code> if unset.
811 <column name="other_config" key="bond-hash-basis"
812 type='{"type": "integer"}'>
813 An integer hashed along with flows when choosing output slaves in load
814 balanced bonds. When changed, all flows will be assigned different
815 hash values possibly causing slave selection decisions to change. Does
816 not affect bonding modes which do not employ load balancing such as
817 <code>active-backup</code>.
820 <group title="Link Failure Detection">
822 An important part of link bonding is detecting that links are down so
823 that they may be disabled. These settings determine how Open vSwitch
824 detects link failure.
827 <column name="other_config" key="bond-detect-mode"
828 type='{"type": "string", "enum": ["set", ["carrier", "miimon"]]}'>
829 The means used to detect link failures. Defaults to
830 <code>carrier</code> which uses each interface's carrier to detect
831 failures. When set to <code>miimon</code>, will check for failures
832 by polling each interface's MII.
835 <column name="other_config" key="bond-miimon-interval"
836 type='{"type": "integer"}'>
837 The interval, in milliseconds, between successive attempts to poll
838 each interface's MII. Relevant only when <ref column="other_config"
839 key="bond-detect-mode"/> is <code>miimon</code>.
842 <column name="bond_updelay">
844 The number of milliseconds for which carrier must stay up on an
845 interface before the interface is considered to be up. Specify
846 <code>0</code> to enable the interface immediately.
850 This setting is honored only when at least one bonded interface is
851 already enabled. When no interfaces are enabled, then the first
852 bond interface to come up is enabled immediately.
856 <column name="bond_downdelay">
857 The number of milliseconds for which carrier must stay down on an
858 interface before the interface is considered to be down. Specify
859 <code>0</code> to disable the interface immediately.
863 <group title="LACP Configuration">
865 LACP, the Link Aggregation Control Protocol, is an IEEE standard that
866 allows switches to automatically detect that they are connected by
867 multiple links and aggregate across those links. These settings
868 control LACP behavior.
872 Configures LACP on this port. LACP allows directly connected
873 switches to negotiate which links may be bonded. LACP may be enabled
874 on non-bonded ports for the benefit of any switches they may be
875 connected to. <code>active</code> ports are allowed to initiate LACP
876 negotiations. <code>passive</code> ports are allowed to participate
877 in LACP negotiations initiated by a remote switch, but not allowed to
878 initiate such negotiations themselves. If LACP is enabled on a port
879 whose partner switch does not support LACP, the bond will be
880 disabled. Defaults to <code>off</code> if unset.
883 <column name="other_config" key="lacp-system-id">
884 The LACP system ID of this <ref table="Port"/>. The system ID of a
885 LACP bond is used to identify itself to its partners. Must be a
886 nonzero MAC address. Defaults to the bridge Ethernet address if
890 <column name="other_config" key="lacp-system-priority"
891 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
892 The LACP system priority of this <ref table="Port"/>. In LACP
893 negotiations, link status decisions are made by the system with the
894 numerically lower priority.
897 <column name="other_config" key="lacp-time"
898 type='{"type": "string", "enum": ["set", ["fast", "slow"]]}'>
900 The LACP timing which should be used on this <ref table="Port"/>.
901 By default <code>slow</code> is used. When configured to be
902 <code>fast</code> LACP heartbeats are requested at a rate of once
903 per second causing connectivity problems to be detected more
904 quickly. In <code>slow</code> mode, heartbeats are requested at a
905 rate of once every 30 seconds.
910 <group title="SLB Configuration">
912 These settings control behavior when a bond is in
913 <code>balance-slb</code> mode, regardless of whether the bond was
914 intentionally configured in SLB mode or it fell back to SLB mode
915 because LACP negotiation failed.
918 <column name="other_config" key="bond-rebalance-interval"
919 type='{"type": "integer", "minInteger": 0, "maxInteger": 10000}'>
920 For a load balanced bonded port, the number of milliseconds between
921 successive attempts to rebalance the bond, that is, to move flows
922 from one interface on the bond to another in an attempt to keep usage
923 of each interface roughly equal. If zero, load balancing is disabled
924 on the bond (carrier status changes still cause flows to move). If
925 less than 1000ms, the rebalance interval will be 1000ms.
929 <column name="bond_fake_iface">
930 For a bonded port, whether to create a fake internal interface with the
931 name of the port. Use only for compatibility with legacy software that
936 <group title="Spanning Tree Configuration">
937 <column name="other_config" key="stp-enable"
938 type='{"type": "boolean"}'>
939 If spanning tree is enabled on the bridge, member ports are
940 enabled by default (with the exception of bond, internal, and
941 mirror ports which do not work with STP). If this column's
942 value is <code>false</code> spanning tree is disabled on the
946 <column name="other_config" key="stp-port-num"
947 type='{"type": "integer", "minInteger": 1, "maxInteger": 255}'>
948 The port number used for the lower 8 bits of the port-id. By
949 default, the numbers will be assigned automatically. If any
950 port's number is manually configured on a bridge, then they
954 <column name="other_config" key="stp-port-priority"
955 type='{"type": "integer", "minInteger": 0, "maxInteger": 255}'>
956 The port's relative priority value for determining the root
957 port (the upper 8 bits of the port-id). A port with a lower
958 port-id will be chosen as the root port. By default, the
962 <column name="other_config" key="stp-path-cost"
963 type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'>
964 Spanning tree path cost for the port. A lower number indicates
965 a faster link. By default, the cost is based on the maximum
970 <group title="Other Features">
972 Quality of Service configuration for this port.
976 The MAC address to use for this port for the purpose of choosing the
977 bridge's MAC address. This column does not necessarily reflect the
978 port's actual MAC address, nor will setting it change the port's actual
982 <column name="fake_bridge">
983 Does this port represent a sub-bridge for its tagged VLAN within the
984 Bridge? See ovs-vsctl(8) for more information.
987 <column name="external_ids" key="fake-bridge-id-*">
988 External IDs for a fake bridge (see the <ref column="fake_bridge"/>
989 column) are defined by prefixing a <ref table="Bridge"/> <ref
990 table="Bridge" column="external_ids"/> key with
991 <code>fake-bridge-</code>,
992 e.g. <code>fake-bridge-xs-network-uuids</code>.
996 <group title="Port Status">
998 Status information about ports attached to bridges.
1000 <column name="status">
1001 Key-value pairs that report port status.
1003 <column name="status" key="stp_port_id">
1005 The port-id (in hex) used in spanning tree advertisements for
1006 this port. Configuring the port-id is described in the
1007 <code>stp-port-num</code> and <code>stp-port-priority</code>
1008 keys of the <code>other_config</code> section earlier.
1011 <column name="status" key="stp_state"
1012 type='{"type": "string", "enum": ["set",
1013 ["disabled", "listening", "learning",
1014 "forwarding", "blocking"]]}'>
1016 STP state of the port.
1019 <column name="status" key="stp_sec_in_state"
1020 type='{"type": "integer", "minInteger": 0}'>
1022 The amount of time (in seconds) port has been in the current
1026 <column name="status" key="stp_role"
1027 type='{"type": "string", "enum": ["set",
1028 ["root", "designated", "alternate"]]}'>
1030 STP role of the port.
1035 <group title="Port Statistics">
1037 Key-value pairs that report port statistics.
1039 <group title="Statistics: STP transmit and receive counters">
1040 <column name="statistics" key="stp_tx_count">
1041 Number of STP BPDUs sent on this port by the spanning
1044 <column name="statistics" key="stp_rx_count">
1045 Number of STP BPDUs received on this port and accepted by the
1046 spanning tree library.
1048 <column name="statistics" key="stp_error_count">
1049 Number of bad STP BPDUs received on this port. Bad BPDUs
1050 include runt packets and those with an unexpected protocol ID.
1055 <group title="Common Columns">
1056 The overall purpose of these columns is described under <code>Common
1057 Columns</code> at the beginning of this document.
1059 <column name="other_config"/>
1060 <column name="external_ids"/>
1064 <table name="Interface" title="One physical network device in a Port.">
1065 An interface within a <ref table="Port"/>.
1067 <group title="Core Features">
1068 <column name="name">
1069 Interface name. Should be alphanumeric and no more than about 8 bytes
1070 long. May be the same as the port name, for non-bonded ports. Must
1071 otherwise be unique among the names of ports, interfaces, and bridges
1076 <p>Ethernet address to set for this interface. If unset then the
1077 default MAC address is used:</p>
1079 <li>For the local interface, the default is the lowest-numbered MAC
1080 address among the other bridge ports, either the value of the
1081 <ref table="Port" column="mac"/> in its <ref table="Port"/> record,
1082 if set, or its actual MAC (for bonded ports, the MAC of its slave
1083 whose name is first in alphabetical order). Internal ports and
1084 bridge ports that are used as port mirroring destinations (see the
1085 <ref table="Mirror"/> table) are ignored.</li>
1086 <li>For other internal interfaces, the default MAC is randomly
1088 <li>External interfaces typically have a MAC address associated with
1089 their hardware.</li>
1091 <p>Some interfaces may not have a software-controllable MAC
1095 <column name="ofport">
1096 <p>OpenFlow port number for this interface. Unlike most columns, this
1097 column's value should be set only by Open vSwitch itself. Other
1098 clients should set this column to an empty set (the default) when
1099 creating an <ref table="Interface"/>.</p>
1100 <p>Open vSwitch populates this column when the port number becomes
1101 known. If the interface is successfully added,
1102 <ref column="ofport"/> will be set to a number between 1 and 65535
1103 (generally either in the range 1 to 65279, inclusive, or 65534, the
1104 port number for the OpenFlow ``local port''). If the interface
1105 cannot be added then Open vSwitch sets this column
1110 <group title="System-Specific Details">
1111 <column name="type">
1113 The interface type, one of:
1117 <dt><code>system</code></dt>
1118 <dd>An ordinary network device, e.g. <code>eth0</code> on Linux.
1119 Sometimes referred to as ``external interfaces'' since they are
1120 generally connected to hardware external to that on which the Open
1121 vSwitch is running. The empty string is a synonym for
1122 <code>system</code>.</dd>
1124 <dt><code>internal</code></dt>
1125 <dd>A simulated network device that sends and receives traffic. An
1126 internal interface whose <ref column="name"/> is the same as its
1127 bridge's <ref table="Open_vSwitch" column="name"/> is called the
1128 ``local interface.'' It does not make sense to bond an internal
1129 interface, so the terms ``port'' and ``interface'' are often used
1130 imprecisely for internal interfaces.</dd>
1132 <dt><code>tap</code></dt>
1133 <dd>A TUN/TAP device managed by Open vSwitch.</dd>
1135 <dt><code>gre</code></dt>
1137 An Ethernet over RFC 2890 Generic Routing Encapsulation over IPv4
1138 tunnel. See <ref group="Tunnel Options"/> for information on
1139 configuring GRE tunnels.
1142 <dt><code>ipsec_gre</code></dt>
1144 An Ethernet over RFC 2890 Generic Routing Encapsulation over IPv4
1148 <dt><code>capwap</code></dt>
1150 An Ethernet tunnel over the UDP transport portion of CAPWAP (RFC
1151 5415). This allows interoperability with certain switches that do
1152 not support GRE. Only the tunneling component of the protocol is
1153 implemented. UDP ports 58881 and 58882 are used as the source and
1154 destination ports respectively. CAPWAP is currently supported only
1155 with the Linux kernel datapath with kernel version 2.6.26 or later.
1158 <dt><code>patch</code></dt>
1160 A pair of virtual devices that act as a patch cable.
1163 <dt><code>null</code></dt>
1164 <dd>An ignored interface.</dd>
1169 <group title="Tunnel Options">
1171 These options apply to interfaces with <ref column="type"/> of
1172 <code>gre</code>, <code>ipsec_gre</code>, and <code>capwap</code>.
1176 Each tunnel must be uniquely identified by the combination of <ref
1177 column="type"/>, <ref column="options" key="remote_ip"/>, <ref
1178 column="options" key="local_ip"/>, and <ref column="options"
1179 key="in_key"/>. If two ports are defined that are the same except one
1180 has an optional identifier and the other does not, the more specific
1181 one is matched first. <ref column="options" key="in_key"/> is
1182 considered more specific than <ref column="options" key="local_ip"/> if
1183 a port defines one and another port defines the other.
1186 <column name="options" key="remote_ip">
1188 Required. The tunnel endpoint. Unicast and multicast endpoints are
1193 When a multicast endpoint is specified, a routing table lookup occurs
1194 only when the tunnel is created. Following a routing change, delete
1195 and then re-create the tunnel to force a new routing table lookup.
1199 <column name="options" key="local_ip">
1200 Optional. The destination IP that received packets must match.
1201 Default is to match all addresses. Must be omitted when <ref
1202 column="options" key="remote_ip"/> is a multicast address.
1205 <column name="options" key="in_key">
1206 <p>Optional. The key that received packets must contain, one of:</p>
1210 <code>0</code>. The tunnel receives packets with no key or with a
1211 key of 0. This is equivalent to specifying no <ref column="options"
1212 key="in_key"/> at all.
1215 A positive 32-bit (for GRE) or 64-bit (for CAPWAP) number. The
1216 tunnel receives only packets with the specified key.
1219 The word <code>flow</code>. The tunnel accepts packets with any
1220 key. The key will be placed in the <code>tun_id</code> field for
1221 matching in the flow table. The <code>ovs-ofctl</code> manual page
1222 contains additional information about matching fields in OpenFlow
1231 <column name="options" key="out_key">
1232 <p>Optional. The key to be set on outgoing packets, one of:</p>
1236 <code>0</code>. Packets sent through the tunnel will have no key.
1237 This is equivalent to specifying no <ref column="options"
1238 key="out_key"/> at all.
1241 A positive 32-bit (for GRE) or 64-bit (for CAPWAP) number. Packets
1242 sent through the tunnel will have the specified key.
1245 The word <code>flow</code>. Packets sent through the tunnel will
1246 have the key set using the <code>set_tunnel</code> Nicira OpenFlow
1247 vendor extension (0 is used in the absence of an action). The
1248 <code>ovs-ofctl</code> manual page contains additional information
1249 about the Nicira OpenFlow vendor extensions.
1254 <column name="options" key="key">
1255 Optional. Shorthand to set <code>in_key</code> and
1256 <code>out_key</code> at the same time.
1259 <column name="options" key="tos">
1260 Optional. The value of the ToS bits to be set on the encapsulating
1261 packet. It may also be the word <code>inherit</code>, in which case
1262 the ToS will be copied from the inner packet if it is IPv4 or IPv6
1263 (otherwise it will be 0). The ECN fields are always inherited.
1267 <column name="options" key="ttl">
1268 Optional. The TTL to be set on the encapsulating packet. It may also
1269 be the word <code>inherit</code>, in which case the TTL will be copied
1270 from the inner packet if it is IPv4 or IPv6 (otherwise it will be the
1271 system default, typically 64). Default is the system default TTL.
1274 <column name="options" key="df_inherit" type='{"type": "boolean"}'>
1275 Optional. If enabled, the Don't Fragment bit will be copied from the
1276 inner IP headers (those of the encapsulated traffic) to the outer
1277 (tunnel) headers. Default is disabled; set to <code>true</code> to
1281 <column name="options" key="df_default"
1282 type='{"type": "boolean"}'>
1283 Optional. If enabled, the Don't Fragment bit will be set by default on
1284 tunnel headers if the <code>df_inherit</code> option is not set, or if
1285 the encapsulated packet is not IP. Default is enabled; set to
1286 <code>false</code> to disable.
1289 <column name="options" key="pmtud" type='{"type": "boolean"}'>
1290 Optional. Enable tunnel path MTU discovery. If enabled ``ICMP
1291 Destination Unreachable - Fragmentation Needed'' messages will be
1292 generated for IPv4 packets with the DF bit set and IPv6 packets above
1293 the minimum MTU if the packet size exceeds the path MTU minus the size
1294 of the tunnel headers. Note that this option causes behavior that is
1295 typically reserved for routers and therefore is not entirely in
1296 compliance with the IEEE 802.1D specification for bridges. Default is
1297 enabled; set to <code>false</code> to disable.
1300 <group title="Tunnel Options: gre only">
1302 Only <code>gre</code> interfaces support these options.
1305 <column name="options" key="header_cache" type='{"type": "boolean"}'>
1306 Enable caching of tunnel headers and the output path. This can lead
1307 to a significant performance increase without changing behavior. In
1308 general it should not be necessary to adjust this setting. However,
1309 the caching can bypass certain components of the IP stack (such as
1310 <code>iptables</code>) and it may be useful to disable it if these
1311 features are required or as a debugging measure. Default is enabled,
1312 set to <code>false</code> to disable.
1316 <group title="Tunnel Options: gre and ipsec_gre only">
1318 Only <code>gre</code> and <code>ipsec_gre</code> interfaces support
1322 <column name="options" key="csum" type='{"type": "boolean"}'>
1324 Optional. Compute GRE checksums on outgoing packets. Default is
1325 disabled, set to <code>true</code> to enable. Checksums present on
1326 incoming packets will be validated regardless of this setting.
1330 GRE checksums impose a significant performance penalty because they
1331 cover the entire packet. The encapsulated L3, L4, and L7 packet
1332 contents typically have their own checksums, so this additional
1333 checksum only adds value for the GRE and encapsulated L2 headers.
1337 This option is supported for <code>ipsec_gre</code>, but not useful
1338 because GRE checksums are weaker than, and redundant with, IPsec
1339 payload authentication.
1344 <group title="Tunnel Options: ipsec_gre only">
1346 Only <code>ipsec_gre</code> interfaces support these options.
1349 <column name="options" key="peer_cert">
1350 Required for certificate authentication. A string containing the
1351 peer's certificate in PEM format. Additionally the host's
1352 certificate must be specified with the <code>certificate</code>
1356 <column name="options" key="certificate">
1357 Required for certificate authentication. The name of a PEM file
1358 containing a certificate that will be presented to the peer during
1362 <column name="options" key="private_key">
1363 Optional for certificate authentication. The name of a PEM file
1364 containing the private key associated with <code>certificate</code>.
1365 If <code>certificate</code> contains the private key, this option may
1369 <column name="options" key="psk">
1370 Required for pre-shared key authentication. Specifies a pre-shared
1371 key for authentication that must be identical on both sides of the
1377 <group title="Patch Options">
1379 Only <code>patch</code> interfaces support these options.
1382 <column name="options" key="peer">
1383 The <ref column="name"/> of the <ref table="Interface"/> for the other
1384 side of the patch. The named <ref table="Interface"/>'s own
1385 <code>peer</code> option must specify this <ref table="Interface"/>'s
1386 name. That is, the two patch interfaces must have reversed <ref
1387 column="name"/> and <code>peer</code> values.
1391 <group title="Interface Status">
1393 Status information about interfaces attached to bridges, updated every
1394 5 seconds. Not all interfaces have all of these properties; virtual
1395 interfaces don't have a link speed, for example. Non-applicable
1396 columns will have empty values.
1398 <column name="admin_state">
1400 The administrative state of the physical network link.
1404 <column name="link_state">
1406 The observed state of the physical network link. This is ordinarily
1407 the link's carrier status. If the interface's <ref table="Port"/> is
1408 a bond configured for miimon monitoring, it is instead the network
1409 link's miimon status.
1413 <column name="link_resets">
1415 The number of times Open vSwitch has observed the
1416 <ref column="link_state"/> of this <ref table="Interface"/> change.
1420 <column name="link_speed">
1422 The negotiated speed of the physical network link.
1423 Valid values are positive integers greater than 0.
1427 <column name="duplex">
1429 The duplex mode of the physical network link.
1435 The MTU (maximum transmission unit); i.e. the largest
1436 amount of data that can fit into a single Ethernet frame.
1437 The standard Ethernet MTU is 1500 bytes. Some physical media
1438 and many kinds of virtual interfaces can be configured with
1442 This column will be empty for an interface that does not
1443 have an MTU as, for example, some kinds of tunnels do not.
1447 <column name="lacp_current">
1448 Boolean value indicating LACP status for this interface. If true, this
1449 interface has current LACP information about its LACP partner. This
1450 information may be used to monitor the health of interfaces in a LACP
1451 enabled port. This column will be empty if LACP is not enabled.
1454 <column name="status">
1455 Key-value pairs that report port status. Supported status values are
1456 <ref column="type"/>-dependent; some interfaces may not have a valid
1457 <ref column="status" key="driver_name"/>, for example.
1460 <column name="status" key="driver_name">
1461 The name of the device driver controlling the network adapter.
1464 <column name="status" key="driver_version">
1465 The version string of the device driver controlling the network
1469 <column name="status" key="firmware_version">
1470 The version string of the network adapter's firmware, if available.
1473 <column name="status" key="source_ip">
1474 The source IP address used for an IPv4 tunnel end-point, such as
1475 <code>gre</code> or <code>capwap</code>.
1478 <column name="status" key="tunnel_egress_iface">
1479 Egress interface for tunnels. Currently only relevant for GRE and
1480 CAPWAP tunnels. On Linux systems, this column will show the name of
1481 the interface which is responsible for routing traffic destined for the
1482 configured <ref column="options" key="remote_ip"/>. This could be an
1483 internal interface such as a bridge port.
1486 <column name="status" key="tunnel_egress_iface_carrier"
1487 type='{"type": "string", "enum": ["set", ["down", "up"]]}'>
1488 Whether carrier is detected on <ref column="status"
1489 key="tunnel_egress_iface"/>.
1493 <group title="Statistics">
1495 Key-value pairs that report interface statistics. The current
1496 implementation updates these counters periodically. Future
1497 implementations may update them when an interface is created, when they
1498 are queried (e.g. using an OVSDB <code>select</code> operation), and
1499 just before an interface is deleted due to virtual interface hot-unplug
1500 or VM shutdown, and perhaps at other times, but not on any regular
1504 These are the same statistics reported by OpenFlow in its <code>struct
1505 ofp_port_stats</code> structure. If an interface does not support a
1506 given statistic, then that pair is omitted.
1508 <group title="Statistics: Successful transmit and receive counters">
1509 <column name="statistics" key="rx_packets">
1510 Number of received packets.
1512 <column name="statistics" key="rx_bytes">
1513 Number of received bytes.
1515 <column name="statistics" key="tx_packets">
1516 Number of transmitted packets.
1518 <column name="statistics" key="tx_bytes">
1519 Number of transmitted bytes.
1522 <group title="Statistics: Receive errors">
1523 <column name="statistics" key="rx_dropped">
1524 Number of packets dropped by RX.
1526 <column name="statistics" key="rx_frame_err">
1527 Number of frame alignment errors.
1529 <column name="statistics" key="rx_over_err">
1530 Number of packets with RX overrun.
1532 <column name="statistics" key="rx_crc_err">
1533 Number of CRC errors.
1535 <column name="statistics" key="rx_errors">
1536 Total number of receive errors, greater than or equal to the sum of
1540 <group title="Statistics: Transmit errors">
1541 <column name="statistics" key="tx_dropped">
1542 Number of packets dropped by TX.
1544 <column name="statistics" key="collisions">
1545 Number of collisions.
1547 <column name="statistics" key="tx_errors">
1548 Total number of transmit errors, greater than or equal to the sum of
1554 <group title="Ingress Policing">
1556 These settings control ingress policing for packets received on this
1557 interface. On a physical interface, this limits the rate at which
1558 traffic is allowed into the system from the outside; on a virtual
1559 interface (one connected to a virtual machine), this limits the rate at
1560 which the VM is able to transmit.
1563 Policing is a simple form of quality-of-service that simply drops
1564 packets received in excess of the configured rate. Due to its
1565 simplicity, policing is usually less accurate and less effective than
1566 egress QoS (which is configured using the <ref table="QoS"/> and <ref
1567 table="Queue"/> tables).
1570 Policing is currently implemented only on Linux. The Linux
1571 implementation uses a simple ``token bucket'' approach:
1575 The size of the bucket corresponds to <ref
1576 column="ingress_policing_burst"/>. Initially the bucket is full.
1579 Whenever a packet is received, its size (converted to tokens) is
1580 compared to the number of tokens currently in the bucket. If the
1581 required number of tokens are available, they are removed and the
1582 packet is forwarded. Otherwise, the packet is dropped.
1585 Whenever it is not full, the bucket is refilled with tokens at the
1586 rate specified by <ref column="ingress_policing_rate"/>.
1590 Policing interacts badly with some network protocols, and especially
1591 with fragmented IP packets. Suppose that there is enough network
1592 activity to keep the bucket nearly empty all the time. Then this token
1593 bucket algorithm will forward a single packet every so often, with the
1594 period depending on packet size and on the configured rate. All of the
1595 fragments of an IP packets are normally transmitted back-to-back, as a
1596 group. In such a situation, therefore, only one of these fragments
1597 will be forwarded and the rest will be dropped. IP does not provide
1598 any way for the intended recipient to ask for only the remaining
1599 fragments. In such a case there are two likely possibilities for what
1600 will happen next: either all of the fragments will eventually be
1601 retransmitted (as TCP will do), in which case the same problem will
1602 recur, or the sender will not realize that its packet has been dropped
1603 and data will simply be lost (as some UDP-based protocols will do).
1604 Either way, it is possible that no forward progress will ever occur.
1606 <column name="ingress_policing_rate">
1608 Maximum rate for data received on this interface, in kbps. Data
1609 received faster than this rate is dropped. Set to <code>0</code>
1610 (the default) to disable policing.
1614 <column name="ingress_policing_burst">
1615 <p>Maximum burst size for data received on this interface, in kb. The
1616 default burst size if set to <code>0</code> is 1000 kb. This value
1617 has no effect if <ref column="ingress_policing_rate"/>
1618 is <code>0</code>.</p>
1620 Specifying a larger burst size lets the algorithm be more forgiving,
1621 which is important for protocols like TCP that react severely to
1622 dropped packets. The burst size should be at least the size of the
1623 interface's MTU. Specifying a value that is numerically at least as
1624 large as 10% of <ref column="ingress_policing_rate"/> helps TCP come
1625 closer to achieving the full rate.
1630 <group title="Connectivity Fault Management">
1632 802.1ag Connectivity Fault Management (CFM) allows a group of
1633 Maintenance Points (MPs) called a Maintenance Association (MA) to
1634 detect connectivity problems with each other. MPs within a MA should
1635 have complete and exclusive interconnectivity. This is verified by
1636 occasionally broadcasting Continuity Check Messages (CCMs) at a
1637 configurable transmission interval.
1641 According to the 802.1ag specification, each Maintenance Point should
1642 be configured out-of-band with a list of Remote Maintenance Points it
1643 should have connectivity to. Open vSwitch differs from the
1644 specification in this area. It simply assumes the link is faulted if
1645 no Remote Maintenance Points are reachable, and considers it not
1649 <column name="cfm_mpid">
1650 A Maintenance Point ID (MPID) uniquely identifies each endpoint within
1651 a Maintenance Association. The MPID is used to identify this endpoint
1652 to other Maintenance Points in the MA. Each end of a link being
1653 monitored should have a different MPID. Must be configured to enable
1654 CFM on this <ref table="Interface"/>.
1657 <column name="cfm_fault">
1659 Indicates a connectivity fault triggered by an inability to receive
1660 heartbeats from any remote endpoint. When a fault is triggered on
1661 <ref table="Interface"/>s participating in bonds, they will be
1665 Faults can be triggered for several reasons. Most importantly they
1666 are triggered when no CCMs are received for a period of 3.5 times the
1667 transmission interval. Faults are also triggered when any CCMs
1668 indicate that a Remote Maintenance Point is not receiving CCMs but
1669 able to send them. Finally, a fault is triggered if a CCM is
1670 received which indicates unexpected configuration. Notably, this
1671 case arises when a CCM is received which advertises the local MPID.
1675 <column name="cfm_fault_status" key="recv">
1676 Indicates a CFM fault was triggered due to a lack of CCMs received on
1677 the <ref table="Interface"/>.
1680 <column name="cfm_fault_status" key="rdi">
1681 Indicates a CFM fault was triggered due to the reception of a CCM with
1682 the RDI bit flagged. Endpoints set the RDI bit in their CCMs when they
1683 are not receiving CCMs themselves. This typically indicates a
1684 unidirectional connectivity failure.
1687 <column name="cfm_fault_status" key="maid">
1688 Indicates a CFM fault was triggered due to the reception of a CCM with
1689 a MAID other than the one Open vSwitch uses. CFM broadcasts are tagged
1690 with an identification number in addition to the MPID called the MAID.
1691 Open vSwitch only supports receiving CCM broadcasts tagged with the
1692 MAID it uses internally.
1695 <column name="cfm_fault_status" key="loopback">
1696 Indicates a CFM fault was triggered due to the reception of a CCM
1697 advertising the same MPID configured in the <ref column="cfm_mpid"/>
1698 column of this <ref table="Interface"/>. This may indicate a loop in
1702 <column name="cfm_fault_status" key="overflow">
1703 Indicates a CFM fault was triggered because the CFM module received
1704 CCMs from more remote endpoints than it can keep track of.
1707 <column name="cfm_fault_status" key="override">
1708 Indicates a CFM fault was manually triggered by an administrator using
1709 an <code>ovs-appctl</code> command.
1712 <column name="cfm_fault_status" key="interval">
1713 Indicates a CFM fault was triggered due to the reception of a CCM
1714 frame having an invalid interval.
1717 <column name="cfm_fault_status" key="sequence">
1718 Indicates a CFM fault was triggered because the CFM module received
1719 a CCM frame with a sequence number that it was not expecting.
1722 <column name="cfm_health">
1724 Indicates the health of the interface as a percentage of CCM frames
1725 received over 21 <ref column="other_config" key="cfm_interval"/>s.
1726 The health of an interface is undefined if it is communicating with
1727 more than one <ref column="cfm_remote_mpids"/>. It reduces if
1728 healthy heartbeats are not received at the expected rate, and
1729 gradually improves as healthy heartbeats are received at the desired
1730 rate. Every 21 <ref column="other_config" key="cfm_interval"/>s, the
1731 health of the interface is refreshed.
1734 As mentioned above, the faults can be triggered for several reasons.
1735 The link health will deteriorate even if heartbeats are received but
1736 they are reported to be unhealthy. An unhealthy heartbeat in this
1737 context is a heartbeat for which either some fault is set or is out
1738 of sequence. The interface health can be 100 only on receiving
1739 healthy heartbeats at the desired rate.
1743 <column name="cfm_remote_mpids">
1744 When CFM is properly configured, Open vSwitch will occasionally
1745 receive CCM broadcasts. These broadcasts contain the MPID of the
1746 sending Maintenance Point. The list of MPIDs from which this
1747 <ref table="Interface"/> is receiving broadcasts from is regularly
1748 collected and written to this column.
1751 <column name="other_config" key="cfm_interval"
1752 type='{"type": "integer"}'>
1753 The interval, in milliseconds, between transmissions of CFM heartbeats.
1754 Three missed heartbeat receptions indicate a connectivity fault.
1758 <column name="other_config" key="cfm_extended"
1759 type='{"type": "boolean"}'>
1760 When <code>true</code>, the CFM module operates in extended mode. This
1761 causes it to use a nonstandard destination address to avoid conflicting
1762 with compliant implementations which may be running concurrently on the
1763 network. Furthermore, extended mode increases the accuracy of the
1764 <code>cfm_interval</code> configuration parameter by breaking wire
1765 compatibility with 802.1ag compliant implementations. Defaults to
1768 <column name="other_config" key="cfm_opstate"
1769 type='{"type": "string", "enum": ["set", ["down", "up"]]}'>
1770 When <code>down</code>, the CFM module marks all CCMs it generates as
1771 operationally down without triggering a fault. This allows remote
1772 maintenance points to choose not to forward traffic to the
1773 <ref table="Interface"/> on which this CFM module is running.
1774 Currently, in Open vSwitch, the opdown bit of CCMs affects
1775 <ref table="Interface"/>s participating in bonds, and the bundle
1776 OpenFlow action. This setting is ignored when CFM is not in extended
1777 mode. Defaults to <code>up</code>.
1780 <column name="other_config" key="cfm_ccm_vlan"
1781 type='{"type": "integer", "minInteger": 1, "maxInteger": 4095}'>
1782 When set, the CFM module will apply a VLAN tag to all CCMs it generates
1783 with the given value. May be the string <code>random</code> in which
1784 case each CCM will be tagged with a different randomly generated VLAN.
1787 <column name="other_config" key="cfm_ccm_pcp"
1788 type='{"type": "integer", "minInteger": 1, "maxInteger": 7}'>
1789 When set, the CFM module will apply a VLAN tag to all CCMs it generates
1790 with the given PCP value. The VLAN ID of the tag is governed by the
1791 value of <ref column="other_config" key="cfm_ccm_vlan"/>. If
1792 <ref column="other_config" key="cfm_ccm_vlan"/> is unset, a VLAN ID of
1798 <group title="Bonding Configuration">
1799 <column name="other_config" key="bond-stable-id"
1800 type='{"type": "integer", "minInteger": 1}'>
1801 Used in <code>stable</code> bond mode to make slave
1802 selection decisions. Allocating <ref column="other_config"
1803 key="bond-stable-id"/> values consistently across interfaces
1804 participating in a bond will guarantee consistent slave selection
1805 decisions across <code>ovs-vswitchd</code> instances when using
1806 <code>stable</code> bonding mode.
1809 <column name="other_config" key="lacp-port-id"
1810 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
1811 The LACP port ID of this <ref table="Interface"/>. Port IDs are
1812 used in LACP negotiations to identify individual ports
1813 participating in a bond.
1816 <column name="other_config" key="lacp-port-priority"
1817 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
1818 The LACP port priority of this <ref table="Interface"/>. In LACP
1819 negotiations <ref table="Interface"/>s with numerically lower
1820 priorities are preferred for aggregation.
1823 <column name="other_config" key="lacp-aggregation-key"
1824 type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'>
1825 The LACP aggregation key of this <ref table="Interface"/>. <ref
1826 table="Interface"/>s with different aggregation keys may not be active
1827 within a given <ref table="Port"/> at the same time.
1831 <group title="Virtual Machine Identifiers">
1833 These key-value pairs specifically apply to an interface that
1834 represents a virtual Ethernet interface connected to a virtual
1835 machine. These key-value pairs should not be present for other types
1836 of interfaces. Keys whose names end in <code>-uuid</code> have
1837 values that uniquely identify the entity in question. For a Citrix
1838 XenServer hypervisor, these values are UUIDs in RFC 4122 format.
1839 Other hypervisors may use other formats.
1842 <column name="external_ids" key="attached-mac">
1843 The MAC address programmed into the ``virtual hardware'' for this
1844 interface, in the form
1845 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>.
1846 For Citrix XenServer, this is the value of the <code>MAC</code> field
1847 in the VIF record for this interface.
1850 <column name="external_ids" key="iface-id">
1851 A system-unique identifier for the interface. On XenServer, this will
1852 commonly be the same as <ref column="external_ids" key="xs-vif-uuid"/>.
1855 <column name="external_ids" key="iface-status"
1856 type='{"type": "string",
1857 "enum": ["set", ["active", "inactive"]]}'>
1859 Hypervisors may sometimes have more than one interface associated
1860 with a given <ref column="external_ids" key="iface-id"/>, only one of
1861 which is actually in use at a given time. For example, in some
1862 circumstances XenServer has both a ``tap'' and a ``vif'' interface
1863 for a single <ref column="external_ids" key="iface-id"/>, but only
1864 uses one of them at a time. A hypervisor that behaves this way must
1865 mark the currently in use interface <code>active</code> and the
1866 others <code>inactive</code>. A hypervisor that never has more than
1867 one interface for a given <ref column="external_ids" key="iface-id"/>
1868 may mark that interface <code>active</code> or omit <ref
1869 column="external_ids" key="iface-status"/> entirely.
1873 During VM migration, a given <ref column="external_ids"
1874 key="iface-id"/> might transiently be marked <code>active</code> on
1875 two different hypervisors. That is, <code>active</code> means that
1876 this <ref column="external_ids" key="iface-id"/> is the active
1877 instance within a single hypervisor, not in a broader scope.
1881 <column name="external_ids" key="xs-vif-uuid">
1882 The virtual interface associated with this interface.
1885 <column name="external_ids" key="xs-network-uuid">
1886 The virtual network to which this interface is attached.
1889 <column name="external_ids" key="vm-id">
1890 The VM to which this interface belongs. On XenServer, this will be the
1891 same as <ref column="external_ids" key="xs-vm-uuid"/>.
1894 <column name="external_ids" key="xs-vm-uuid">
1895 The VM to which this interface belongs.
1899 <group title="VLAN Splinters">
1901 The ``VLAN splinters'' feature increases Open vSwitch compatibility
1902 with buggy network drivers in old versions of Linux that do not
1903 properly support VLANs when VLAN devices are not used, at some cost
1904 in memory and performance.
1908 When VLAN splinters are enabled on a particular interface, Open vSwitch
1909 creates a VLAN device for each in-use VLAN. For sending traffic tagged
1910 with a VLAN on the interface, it substitutes the VLAN device. Traffic
1911 received on the VLAN device is treated as if it had been received on
1912 the interface on the particular VLAN.
1916 VLAN splinters consider a VLAN to be in use if:
1921 The VLAN is the <ref table="Port" column="tag"/> value in any <ref
1922 table="Port"/> record.
1926 The VLAN is listed within the <ref table="Port" column="trunks"/>
1927 column of the <ref table="Port"/> record of an interface on which
1928 VLAN splinters are enabled.
1930 An empty <ref table="Port" column="trunks"/> does not influence the
1931 in-use VLANs: creating 4,096 VLAN devices is impractical because it
1932 will exceed the current 1,024 port per datapath limit.
1936 An OpenFlow flow within any bridge matches the VLAN.
1941 The same set of in-use VLANs applies to every interface on which VLAN
1942 splinters are enabled. That is, the set is not chosen separately for
1943 each interface but selected once as the union of all in-use VLANs based
1948 It does not make sense to enable VLAN splinters on an interface for an
1949 access port, or on an interface that is not a physical port.
1953 VLAN splinters are deprecated. When broken device drivers are no
1954 longer in widespread use, we will delete this feature.
1957 <column name="other_config" key="enable-vlan-splinters"
1958 type='{"type": "boolean"}'>
1960 Set to <code>true</code> to enable VLAN splinters on this interface.
1961 Defaults to <code>false</code>.
1965 VLAN splinters increase kernel and userspace memory overhead, so do
1966 not use them unless they are needed.
1970 VLAN splinters do not support 802.1p priority tags. Received
1971 priorities will appear to be 0, regardless of their actual values,
1972 and priorities on transmitted packets will also be cleared to 0.
1977 <group title="Common Columns">
1978 The overall purpose of these columns is described under <code>Common
1979 Columns</code> at the beginning of this document.
1981 <column name="other_config"/>
1982 <column name="external_ids"/>
1986 <table name="Flow_Table" title="OpenFlow table configuration">
1987 <p>Configuration for a particular OpenFlow table.</p>
1989 <column name="name">
1990 The table's name. Set this column to change the name that controllers
1991 will receive when they request table statistics, e.g. <code>ovs-ofctl
1992 dump-tables</code>. The name does not affect switch behavior.
1995 <column name="flow_limit">
1996 If set, limits the number of flows that may be added to the table. Open
1997 vSwitch may limit the number of flows in a table for other reasons,
1998 e.g. due to hardware limitations or for resource availability or
1999 performance reasons.
2002 <column name="overflow_policy">
2004 Controls the switch's behavior when an OpenFlow flow table modification
2005 request would add flows in excess of <ref column="flow_limit"/>. The
2006 supported values are:
2010 <dt><code>refuse</code></dt>
2012 Refuse to add the flow or flows. This is also the default policy
2013 when <ref column="overflow_policy"/> is unset.
2016 <dt><code>evict</code></dt>
2018 Delete the flow that will expire soonest. See <ref column="groups"/>
2024 <column name="groups">
2026 When <ref column="overflow_policy"/> is <code>evict</code>, this
2027 controls how flows are chosen for eviction when the flow table would
2028 otherwise exceed <ref column="flow_limit"/> flows. Its value is a set
2029 of NXM fields or sub-fields, each of which takes one of the forms
2030 <code><var>field</var>[]</code> or
2031 <code><var>field</var>[<var>start</var>..<var>end</var>]</code>,
2032 e.g. <code>NXM_OF_IN_PORT[]</code>. Please see
2033 <code>nicira-ext.h</code> for a complete list of NXM field names.
2037 When a flow must be evicted due to overflow, the flow to evict is
2038 chosen through an approximation of the following algorithm:
2043 Divide the flows in the table into groups based on the values of the
2044 specified fields or subfields, so that all of the flows in a given
2045 group have the same values for those fields. If a flow does not
2046 specify a given field, that field's value is treated as 0.
2050 Consider the flows in the largest group, that is, the group that
2051 contains the greatest number of flows. If two or more groups all
2052 have the same largest number of flows, consider the flows in all of
2057 Among the flows under consideration, choose the flow that expires
2058 soonest for eviction.
2063 The eviction process only considers flows that have an idle timeout or
2064 a hard timeout. That is, eviction never deletes permanent flows.
2065 (Permanent flows do count against <ref column="flow_limit"/>.
2069 Open vSwitch ignores any invalid or unknown field specifications.
2073 When <ref column="overflow_policy"/> is not <code>evict</code>, this
2074 column has no effect.
2079 <table name="QoS" title="Quality of Service configuration">
2080 <p>Quality of Service (QoS) configuration for each Port that
2083 <column name="type">
2084 <p>The type of QoS to implement. The currently defined types are
2087 <dt><code>linux-htb</code></dt>
2089 Linux ``hierarchy token bucket'' classifier. See tc-htb(8) (also at
2090 <code>http://linux.die.net/man/8/tc-htb</code>) and the HTB manual
2091 (<code>http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm</code>)
2092 for information on how this classifier works and how to configure it.
2096 <dt><code>linux-hfsc</code></dt>
2098 Linux "Hierarchical Fair Service Curve" classifier.
2099 See <code>http://linux-ip.net/articles/hfsc.en/</code> for
2100 information on how this classifier works.
2105 <column name="queues">
2106 <p>A map from queue numbers to <ref table="Queue"/> records. The
2107 supported range of queue numbers depend on <ref column="type"/>. The
2108 queue numbers are the same as the <code>queue_id</code> used in
2109 OpenFlow in <code>struct ofp_action_enqueue</code> and other
2113 Queue 0 is the ``default queue.'' It is used by OpenFlow output
2114 actions when no specific queue has been set. When no configuration for
2115 queue 0 is present, it is automatically configured as if a <ref
2116 table="Queue"/> record with empty <ref table="Queue" column="dscp"/>
2117 and <ref table="Queue" column="other_config"/> columns had been
2119 (Before version 1.6, Open vSwitch would leave queue 0 unconfigured in
2120 this case. With some queuing disciplines, this dropped all packets
2121 destined for the default queue.)
2125 <group title="Configuration for linux-htb and linux-hfsc">
2127 The <code>linux-htb</code> and <code>linux-hfsc</code> classes support
2128 the following key-value pair:
2131 <column name="other_config" key="max-rate" type='{"type": "integer"}'>
2132 Maximum rate shared by all queued traffic, in bit/s. Optional. If not
2133 specified, for physical interfaces, the default is the link rate. For
2134 other interfaces or if the link rate cannot be determined, the default
2135 is currently 100 Mbps.
2139 <group title="Common Columns">
2140 The overall purpose of these columns is described under <code>Common
2141 Columns</code> at the beginning of this document.
2143 <column name="other_config"/>
2144 <column name="external_ids"/>
2148 <table name="Queue" title="QoS output queue.">
2149 <p>A configuration for a port output queue, used in configuring Quality of
2150 Service (QoS) features. May be referenced by <ref column="queues"
2151 table="QoS"/> column in <ref table="QoS"/> table.</p>
2153 <column name="dscp">
2154 If set, Open vSwitch will mark all traffic egressing this
2155 <ref table="Queue"/> with the given DSCP bits. Traffic egressing the
2156 default <ref table="Queue"/> is only marked if it was explicitly selected
2157 as the <ref table="Queue"/> at the time the packet was output. If unset,
2158 the DSCP bits of traffic egressing this <ref table="Queue"/> will remain
2162 <group title="Configuration for linux-htb QoS">
2164 <ref table="QoS"/> <ref table="QoS" column="type"/>
2165 <code>linux-htb</code> may use <code>queue_id</code>s less than 61440.
2166 It has the following key-value pairs defined.
2169 <column name="other_config" key="min-rate"
2170 type='{"type": "integer", "minInteger": 1}'>
2171 Minimum guaranteed bandwidth, in bit/s.
2174 <column name="other_config" key="max-rate"
2175 type='{"type": "integer", "minInteger": 1}'>
2176 Maximum allowed bandwidth, in bit/s. Optional. If specified, the
2177 queue's rate will not be allowed to exceed the specified value, even
2178 if excess bandwidth is available. If unspecified, defaults to no
2182 <column name="other_config" key="burst"
2183 type='{"type": "integer", "minInteger": 1}'>
2184 Burst size, in bits. This is the maximum amount of ``credits'' that a
2185 queue can accumulate while it is idle. Optional. Details of the
2186 <code>linux-htb</code> implementation require a minimum burst size, so
2187 a too-small <code>burst</code> will be silently ignored.
2190 <column name="other_config" key="priority"
2191 type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'>
2192 A queue with a smaller <code>priority</code> will receive all the
2193 excess bandwidth that it can use before a queue with a larger value
2194 receives any. Specific priority values are unimportant; only relative
2195 ordering matters. Defaults to 0 if unspecified.
2199 <group title="Configuration for linux-hfsc QoS">
2201 <ref table="QoS"/> <ref table="QoS" column="type"/>
2202 <code>linux-hfsc</code> may use <code>queue_id</code>s less than 61440.
2203 It has the following key-value pairs defined.
2206 <column name="other_config" key="min-rate"
2207 type='{"type": "integer", "minInteger": 1}'>
2208 Minimum guaranteed bandwidth, in bit/s.
2211 <column name="other_config" key="max-rate"
2212 type='{"type": "integer", "minInteger": 1}'>
2213 Maximum allowed bandwidth, in bit/s. Optional. If specified, the
2214 queue's rate will not be allowed to exceed the specified value, even if
2215 excess bandwidth is available. If unspecified, defaults to no
2220 <group title="Common Columns">
2221 The overall purpose of these columns is described under <code>Common
2222 Columns</code> at the beginning of this document.
2224 <column name="other_config"/>
2225 <column name="external_ids"/>
2229 <table name="Mirror" title="Port mirroring.">
2230 <p>A port mirror within a <ref table="Bridge"/>.</p>
2231 <p>A port mirror configures a bridge to send selected frames to special
2232 ``mirrored'' ports, in addition to their normal destinations. Mirroring
2233 traffic may also be referred to as SPAN or RSPAN, depending on how
2234 the mirrored traffic is sent.</p>
2236 <column name="name">
2237 Arbitrary identifier for the <ref table="Mirror"/>.
2240 <group title="Selecting Packets for Mirroring">
2242 To be selected for mirroring, a given packet must enter or leave the
2243 bridge through a selected port and it must also be in one of the
2247 <column name="select_all">
2248 If true, every packet arriving or departing on any port is
2249 selected for mirroring.
2252 <column name="select_dst_port">
2253 Ports on which departing packets are selected for mirroring.
2256 <column name="select_src_port">
2257 Ports on which arriving packets are selected for mirroring.
2260 <column name="select_vlan">
2261 VLANs on which packets are selected for mirroring. An empty set
2262 selects packets on all VLANs.
2266 <group title="Mirroring Destination Configuration">
2268 These columns are mutually exclusive. Exactly one of them must be
2272 <column name="output_port">
2273 <p>Output port for selected packets, if nonempty.</p>
2274 <p>Specifying a port for mirror output reserves that port exclusively
2275 for mirroring. No frames other than those selected for mirroring
2277 will be forwarded to the port, and any frames received on the port
2278 will be discarded.</p>
2280 The output port may be any kind of port supported by Open vSwitch.
2281 It may be, for example, a physical port (sometimes called SPAN) or a
2286 <column name="output_vlan">
2287 <p>Output VLAN for selected packets, if nonempty.</p>
2288 <p>The frames will be sent out all ports that trunk
2289 <ref column="output_vlan"/>, as well as any ports with implicit VLAN
2290 <ref column="output_vlan"/>. When a mirrored frame is sent out a
2291 trunk port, the frame's VLAN tag will be set to
2292 <ref column="output_vlan"/>, replacing any existing tag; when it is
2293 sent out an implicit VLAN port, the frame will not be tagged. This
2294 type of mirroring is sometimes called RSPAN.</p>
2296 The following destination MAC addresses will not be mirrored to a
2297 VLAN to avoid confusing switches that interpret the protocols that
2301 <dt><code>01:80:c2:00:00:00</code></dt>
2302 <dd>IEEE 802.1D Spanning Tree Protocol (STP).</dd>
2304 <dt><code>01:80:c2:00:00:01</code></dt>
2305 <dd>IEEE Pause frame.</dd>
2307 <dt><code>01:80:c2:00:00:0<var>x</var></code></dt>
2308 <dd>Other reserved protocols.</dd>
2310 <dt><code>01:00:0c:cc:cc:cc</code></dt>
2312 Cisco Discovery Protocol (CDP), VLAN Trunking Protocol (VTP),
2313 Dynamic Trunking Protocol (DTP), Port Aggregation Protocol (PAgP),
2317 <dt><code>01:00:0c:cc:cc:cd</code></dt>
2318 <dd>Cisco Shared Spanning Tree Protocol PVSTP+.</dd>
2320 <dt><code>01:00:0c:cd:cd:cd</code></dt>
2321 <dd>Cisco STP Uplink Fast.</dd>
2323 <dt><code>01:00:0c:00:00:00</code></dt>
2324 <dd>Cisco Inter Switch Link.</dd>
2326 <p><em>Please note:</em> Mirroring to a VLAN can disrupt a network that
2327 contains unmanaged switches. Consider an unmanaged physical switch
2328 with two ports: port 1, connected to an end host, and port 2,
2329 connected to an Open vSwitch configured to mirror received packets
2330 into VLAN 123 on port 2. Suppose that the end host sends a packet on
2331 port 1 that the physical switch forwards to port 2. The Open vSwitch
2332 forwards this packet to its destination and then reflects it back on
2333 port 2 in VLAN 123. This reflected packet causes the unmanaged
2334 physical switch to replace the MAC learning table entry, which
2335 correctly pointed to port 1, with one that incorrectly points to port
2336 2. Afterward, the physical switch will direct packets destined for
2337 the end host to the Open vSwitch on port 2, instead of to the end
2338 host on port 1, disrupting connectivity. If mirroring to a VLAN is
2339 desired in this scenario, then the physical switch must be replaced
2340 by one that learns Ethernet addresses on a per-VLAN basis. In
2341 addition, learning should be disabled on the VLAN containing mirrored
2342 traffic. If this is not done then intermediate switches will learn
2343 the MAC address of each end host from the mirrored traffic. If
2344 packets being sent to that end host are also mirrored, then they will
2345 be dropped since the switch will attempt to send them out the input
2346 port. Disabling learning for the VLAN will cause the switch to
2347 correctly send the packet out all ports configured for that VLAN. If
2348 Open vSwitch is being used as an intermediate switch, learning can be
2349 disabled by adding the mirrored VLAN to <ref column="flood_vlans"/>
2350 in the appropriate <ref table="Bridge"/> table or tables.</p>
2352 Mirroring to a GRE tunnel has fewer caveats than mirroring to a
2353 VLAN and should generally be preferred.
2358 <group title="Statistics: Mirror counters">
2360 Key-value pairs that report mirror statistics.
2362 <column name="statistics" key="tx_packets">
2363 Number of packets transmitted through this mirror.
2365 <column name="statistics" key="tx_bytes">
2366 Number of bytes transmitted through this mirror.
2370 <group title="Common Columns">
2371 The overall purpose of these columns is described under <code>Common
2372 Columns</code> at the beginning of this document.
2374 <column name="external_ids"/>
2378 <table name="Controller" title="OpenFlow controller configuration.">
2379 <p>An OpenFlow controller.</p>
2382 Open vSwitch supports two kinds of OpenFlow controllers:
2386 <dt>Primary controllers</dt>
2389 This is the kind of controller envisioned by the OpenFlow 1.0
2390 specification. Usually, a primary controller implements a network
2391 policy by taking charge of the switch's flow table.
2395 Open vSwitch initiates and maintains persistent connections to
2396 primary controllers, retrying the connection each time it fails or
2397 drops. The <ref table="Bridge" column="fail_mode"/> column in the
2398 <ref table="Bridge"/> table applies to primary controllers.
2402 Open vSwitch permits a bridge to have any number of primary
2403 controllers. When multiple controllers are configured, Open
2404 vSwitch connects to all of them simultaneously. Because
2405 OpenFlow 1.0 does not specify how multiple controllers
2406 coordinate in interacting with a single switch, more than
2407 one primary controller should be specified only if the
2408 controllers are themselves designed to coordinate with each
2409 other. (The Nicira-defined <code>NXT_ROLE</code> OpenFlow
2410 vendor extension may be useful for this.)
2413 <dt>Service controllers</dt>
2416 These kinds of OpenFlow controller connections are intended for
2417 occasional support and maintenance use, e.g. with
2418 <code>ovs-ofctl</code>. Usually a service controller connects only
2419 briefly to inspect or modify some of a switch's state.
2423 Open vSwitch listens for incoming connections from service
2424 controllers. The service controllers initiate and, if necessary,
2425 maintain the connections from their end. The <ref table="Bridge"
2426 column="fail_mode"/> column in the <ref table="Bridge"/> table does
2427 not apply to service controllers.
2431 Open vSwitch supports configuring any number of service controllers.
2437 The <ref column="target"/> determines the type of controller.
2440 <group title="Core Features">
2441 <column name="target">
2442 <p>Connection method for controller.</p>
2444 The following connection methods are currently supported for primary
2448 <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2450 <p>The specified SSL <var>port</var> (default: 6633) on the host at
2451 the given <var>ip</var>, which must be expressed as an IP address
2452 (not a DNS name). The <ref table="Open_vSwitch" column="ssl"/>
2453 column in the <ref table="Open_vSwitch"/> table must point to a
2454 valid SSL configuration when this form is used.</p>
2455 <p>SSL support is an optional feature that is not always built as
2456 part of Open vSwitch.</p>
2458 <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2459 <dd>The specified TCP <var>port</var> (default: 6633) on the host at
2460 the given <var>ip</var>, which must be expressed as an IP address
2461 (not a DNS name).</dd>
2464 The following connection methods are currently supported for service
2468 <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2471 Listens for SSL connections on the specified TCP <var>port</var>
2472 (default: 6633). If <var>ip</var>, which must be expressed as an
2473 IP address (not a DNS name), is specified, then connections are
2474 restricted to the specified local IP address.
2477 The <ref table="Open_vSwitch" column="ssl"/> column in the <ref
2478 table="Open_vSwitch"/> table must point to a valid SSL
2479 configuration when this form is used.
2481 <p>SSL support is an optional feature that is not always built as
2482 part of Open vSwitch.</p>
2484 <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2486 Listens for connections on the specified TCP <var>port</var>
2487 (default: 6633). If <var>ip</var>, which must be expressed as an
2488 IP address (not a DNS name), is specified, then connections are
2489 restricted to the specified local IP address.
2492 <p>When multiple controllers are configured for a single bridge, the
2493 <ref column="target"/> values must be unique. Duplicate
2494 <ref column="target"/> values yield unspecified results.</p>
2497 <column name="connection_mode">
2498 <p>If it is specified, this setting must be one of the following
2499 strings that describes how Open vSwitch contacts this OpenFlow
2500 controller over the network:</p>
2503 <dt><code>in-band</code></dt>
2504 <dd>In this mode, this controller's OpenFlow traffic travels over the
2505 bridge associated with the controller. With this setting, Open
2506 vSwitch allows traffic to and from the controller regardless of the
2507 contents of the OpenFlow flow table. (Otherwise, Open vSwitch
2508 would never be able to connect to the controller, because it did
2509 not have a flow to enable it.) This is the most common connection
2510 mode because it is not necessary to maintain two independent
2512 <dt><code>out-of-band</code></dt>
2513 <dd>In this mode, OpenFlow traffic uses a control network separate
2514 from the bridge associated with this controller, that is, the
2515 bridge does not use any of its own network devices to communicate
2516 with the controller. The control network must be configured
2517 separately, before or after <code>ovs-vswitchd</code> is started.
2521 <p>If not specified, the default is implementation-specific.</p>
2525 <group title="Controller Failure Detection and Handling">
2526 <column name="max_backoff">
2527 Maximum number of milliseconds to wait between connection attempts.
2528 Default is implementation-specific.
2531 <column name="inactivity_probe">
2532 Maximum number of milliseconds of idle time on connection to
2533 controller before sending an inactivity probe message. If Open
2534 vSwitch does not communicate with the controller for the specified
2535 number of seconds, it will send a probe. If a response is not
2536 received for the same additional amount of time, Open vSwitch
2537 assumes the connection has been broken and attempts to reconnect.
2538 Default is implementation-specific. A value of 0 disables
2543 <group title="Asynchronous Message Configuration">
2545 OpenFlow switches send certain messages to controllers spontanenously,
2546 that is, not in response to any request from the controller. These
2547 messages are called ``asynchronous messages.'' These columns allow
2548 asynchronous messages to be limited or disabled to ensure the best use
2549 of network resources.
2552 <column name="enable_async_messages">
2553 The OpenFlow protocol enables asynchronous messages at time of
2554 connection establishment, which means that a controller can receive
2555 asynchronous messages, potentially many of them, even if it turns them
2556 off immediately after connecting. Set this column to
2557 <code>false</code> to change Open vSwitch behavior to disable, by
2558 default, all asynchronous messages. The controller can use the
2559 <code>NXT_SET_ASYNC_CONFIG</code> Nicira extension to OpenFlow to turn
2560 on any messages that it does want to receive, if any.
2563 <column name="controller_rate_limit">
2565 The maximum rate at which the switch will forward packets to the
2566 OpenFlow controller, in packets per second. This feature prevents a
2567 single bridge from overwhelming the controller. If not specified,
2568 the default is implementation-specific.
2572 In addition, when a high rate triggers rate-limiting, Open vSwitch
2573 queues controller packets for each port and transmits them to the
2574 controller at the configured rate. The <ref
2575 column="controller_burst_limit"/> value limits the number of queued
2576 packets. Ports on a bridge share the packet queue fairly.
2580 Open vSwitch maintains two such packet rate-limiters per bridge: one
2581 for packets sent up to the controller because they do not correspond
2582 to any flow, and the other for packets sent up to the controller by
2583 request through flow actions. When both rate-limiters are filled with
2584 packets, the actual rate that packets are sent to the controller is
2585 up to twice the specified rate.
2589 <column name="controller_burst_limit">
2590 In conjunction with <ref column="controller_rate_limit"/>,
2591 the maximum number of unused packet credits that the bridge will
2592 allow to accumulate, in packets. If not specified, the default
2593 is implementation-specific.
2597 <group title="Additional In-Band Configuration">
2598 <p>These values are considered only in in-band control mode (see
2599 <ref column="connection_mode"/>).</p>
2601 <p>When multiple controllers are configured on a single bridge, there
2602 should be only one set of unique values in these columns. If different
2603 values are set for these columns in different controllers, the effect
2606 <column name="local_ip">
2607 The IP address to configure on the local port,
2608 e.g. <code>192.168.0.123</code>. If this value is unset, then
2609 <ref column="local_netmask"/> and <ref column="local_gateway"/> are
2613 <column name="local_netmask">
2614 The IP netmask to configure on the local port,
2615 e.g. <code>255.255.255.0</code>. If <ref column="local_ip"/> is set
2616 but this value is unset, then the default is chosen based on whether
2617 the IP address is class A, B, or C.
2620 <column name="local_gateway">
2621 The IP address of the gateway to configure on the local port, as a
2622 string, e.g. <code>192.168.0.1</code>. Leave this column unset if
2623 this network has no gateway.
2627 <group title="Controller Status">
2628 <column name="is_connected">
2629 <code>true</code> if currently connected to this controller,
2630 <code>false</code> otherwise.
2634 type='{"type": "string", "enum": ["set", ["other", "master", "slave"]]}'>
2635 <p>The level of authority this controller has on the associated
2636 bridge. Possible values are:</p>
2638 <dt><code>other</code></dt>
2639 <dd>Allows the controller access to all OpenFlow features.</dd>
2640 <dt><code>master</code></dt>
2641 <dd>Equivalent to <code>other</code>, except that there may be at
2642 most one master controller at a time. When a controller configures
2643 itself as <code>master</code>, any existing master is demoted to
2644 the <code>slave</code>role.</dd>
2645 <dt><code>slave</code></dt>
2646 <dd>Allows the controller read-only access to OpenFlow features.
2647 Attempts to modify the flow table will be rejected with an
2648 error. Slave controllers do not receive OFPT_PACKET_IN or
2649 OFPT_FLOW_REMOVED messages, but they do receive OFPT_PORT_STATUS
2654 <column name="status" key="last_error">
2655 A human-readable description of the last error on the connection
2656 to the controller; i.e. <code>strerror(errno)</code>. This key
2657 will exist only if an error has occurred.
2660 <column name="status" key="state"
2661 type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'>
2663 The state of the connection to the controller:
2666 <dt><code>VOID</code></dt>
2667 <dd>Connection is disabled.</dd>
2669 <dt><code>BACKOFF</code></dt>
2670 <dd>Attempting to reconnect at an increasing period.</dd>
2672 <dt><code>CONNECTING</code></dt>
2673 <dd>Attempting to connect.</dd>
2675 <dt><code>ACTIVE</code></dt>
2676 <dd>Connected, remote host responsive.</dd>
2678 <dt><code>IDLE</code></dt>
2679 <dd>Connection is idle. Waiting for response to keep-alive.</dd>
2682 These values may change in the future. They are provided only for
2687 <column name="status" key="sec_since_connect"
2688 type='{"type": "integer", "minInteger": 0}'>
2689 The amount of time since this controller last successfully connected to
2690 the switch (in seconds). Value is empty if controller has never
2691 successfully connected.
2694 <column name="status" key="sec_since_disconnect"
2695 type='{"type": "integer", "minInteger": 1}'>
2696 The amount of time since this controller last disconnected from
2697 the switch (in seconds). Value is empty if controller has never
2702 <group title="Connection Parameters">
2704 Additional configuration for a connection between the controller
2705 and the Open vSwitch.
2708 <column name="other_config" key="dscp"
2709 type='{"type": "integer"}'>
2710 The Differentiated Service Code Point (DSCP) is specified using 6 bits
2711 in the Type of Service (TOS) field in the IP header. DSCP provides a
2712 mechanism to classify the network traffic and provide Quality of
2713 Service (QoS) on IP networks.
2715 The DSCP value specified here is used when establishing the connection
2716 between the controller and the Open vSwitch. The connection must be
2717 reset for the new DSCP values to take effect. If no value is
2718 specified, a default value of 48 is chosen. Valid DSCP values must be
2719 in the range 0 to 63.
2724 <group title="Common Columns">
2725 The overall purpose of these columns is described under <code>Common
2726 Columns</code> at the beginning of this document.
2728 <column name="external_ids"/>
2729 <column name="other_config"/>
2733 <table name="Manager" title="OVSDB management connection.">
2735 Configuration for a database connection to an Open vSwitch database
2740 This table primarily configures the Open vSwitch database
2741 (<code>ovsdb-server</code>), not the Open vSwitch switch
2742 (<code>ovs-vswitchd</code>). The switch does read the table to determine
2743 what connections should be treated as in-band.
2747 The Open vSwitch database server can initiate and maintain active
2748 connections to remote clients. It can also listen for database
2752 <group title="Core Features">
2753 <column name="target">
2754 <p>Connection method for managers.</p>
2756 The following connection methods are currently supported:
2759 <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2762 The specified SSL <var>port</var> (default: 6632) on the host at
2763 the given <var>ip</var>, which must be expressed as an IP address
2764 (not a DNS name). The <ref table="Open_vSwitch" column="ssl"/>
2765 column in the <ref table="Open_vSwitch"/> table must point to a
2766 valid SSL configuration when this form is used.
2769 SSL support is an optional feature that is not always built as
2770 part of Open vSwitch.
2774 <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
2776 The specified TCP <var>port</var> (default: 6632) on the host at
2777 the given <var>ip</var>, which must be expressed as an IP address
2780 <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2783 Listens for SSL connections on the specified TCP <var>port</var>
2784 (default: 6632). If <var>ip</var>, which must be expressed as an
2785 IP address (not a DNS name), is specified, then connections are
2786 restricted to the specified local IP address.
2789 The <ref table="Open_vSwitch" column="ssl"/> column in the <ref
2790 table="Open_vSwitch"/> table must point to a valid SSL
2791 configuration when this form is used.
2794 SSL support is an optional feature that is not always built as
2795 part of Open vSwitch.
2798 <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt>
2800 Listens for connections on the specified TCP <var>port</var>
2801 (default: 6632). If <var>ip</var>, which must be expressed as an
2802 IP address (not a DNS name), is specified, then connections are
2803 restricted to the specified local IP address.
2806 <p>When multiple managers are configured, the <ref column="target"/>
2807 values must be unique. Duplicate <ref column="target"/> values yield
2808 unspecified results.</p>
2811 <column name="connection_mode">
2813 If it is specified, this setting must be one of the following strings
2814 that describes how Open vSwitch contacts this OVSDB client over the
2819 <dt><code>in-band</code></dt>
2821 In this mode, this connection's traffic travels over a bridge
2822 managed by Open vSwitch. With this setting, Open vSwitch allows
2823 traffic to and from the client regardless of the contents of the
2824 OpenFlow flow table. (Otherwise, Open vSwitch would never be able
2825 to connect to the client, because it did not have a flow to enable
2826 it.) This is the most common connection mode because it is not
2827 necessary to maintain two independent networks.
2829 <dt><code>out-of-band</code></dt>
2831 In this mode, the client's traffic uses a control network separate
2832 from that managed by Open vSwitch, that is, Open vSwitch does not
2833 use any of its own network devices to communicate with the client.
2834 The control network must be configured separately, before or after
2835 <code>ovs-vswitchd</code> is started.
2840 If not specified, the default is implementation-specific.
2845 <group title="Client Failure Detection and Handling">
2846 <column name="max_backoff">
2847 Maximum number of milliseconds to wait between connection attempts.
2848 Default is implementation-specific.
2851 <column name="inactivity_probe">
2852 Maximum number of milliseconds of idle time on connection to the client
2853 before sending an inactivity probe message. If Open vSwitch does not
2854 communicate with the client for the specified number of seconds, it
2855 will send a probe. If a response is not received for the same
2856 additional amount of time, Open vSwitch assumes the connection has been
2857 broken and attempts to reconnect. Default is implementation-specific.
2858 A value of 0 disables inactivity probes.
2862 <group title="Status">
2863 <column name="is_connected">
2864 <code>true</code> if currently connected to this manager,
2865 <code>false</code> otherwise.
2868 <column name="status" key="last_error">
2869 A human-readable description of the last error on the connection
2870 to the manager; i.e. <code>strerror(errno)</code>. This key
2871 will exist only if an error has occurred.
2874 <column name="status" key="state"
2875 type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'>
2877 The state of the connection to the manager:
2880 <dt><code>VOID</code></dt>
2881 <dd>Connection is disabled.</dd>
2883 <dt><code>BACKOFF</code></dt>
2884 <dd>Attempting to reconnect at an increasing period.</dd>
2886 <dt><code>CONNECTING</code></dt>
2887 <dd>Attempting to connect.</dd>
2889 <dt><code>ACTIVE</code></dt>
2890 <dd>Connected, remote host responsive.</dd>
2892 <dt><code>IDLE</code></dt>
2893 <dd>Connection is idle. Waiting for response to keep-alive.</dd>
2896 These values may change in the future. They are provided only for
2901 <column name="status" key="sec_since_connect"
2902 type='{"type": "integer", "minInteger": 0}'>
2903 The amount of time since this manager last successfully connected
2904 to the database (in seconds). Value is empty if manager has never
2905 successfully connected.
2908 <column name="status" key="sec_since_disconnect"
2909 type='{"type": "integer", "minInteger": 0}'>
2910 The amount of time since this manager last disconnected from the
2911 database (in seconds). Value is empty if manager has never
2915 <column name="status" key="locks_held">
2916 Space-separated list of the names of OVSDB locks that the connection
2917 holds. Omitted if the connection does not hold any locks.
2920 <column name="status" key="locks_waiting">
2921 Space-separated list of the names of OVSDB locks that the connection is
2922 currently waiting to acquire. Omitted if the connection is not waiting
2926 <column name="status" key="locks_lost">
2927 Space-separated list of the names of OVSDB locks that the connection
2928 has had stolen by another OVSDB client. Omitted if no locks have been
2929 stolen from this connection.
2932 <column name="status" key="n_connections"
2933 type='{"type": "integer", "minInteger": 2}'>
2935 When <ref column="target"/> specifies a connection method that
2936 listens for inbound connections (e.g. <code>ptcp:</code> or
2937 <code>pssl:</code>) and more than one connection is actually active,
2938 the value is the number of active connections. Otherwise, this
2939 key-value pair is omitted.
2942 When multiple connections are active, status columns and key-value
2943 pairs (other than this one) report the status of one arbitrarily
2949 <group title="Connection Parameters">
2951 Additional configuration for a connection between the manager
2952 and the Open vSwitch Database.
2955 <column name="other_config" key="dscp"
2956 type='{"type": "integer"}'>
2957 The Differentiated Service Code Point (DSCP) is specified using 6 bits
2958 in the Type of Service (TOS) field in the IP header. DSCP provides a
2959 mechanism to classify the network traffic and provide Quality of
2960 Service (QoS) on IP networks.
2962 The DSCP value specified here is used when establishing the connection
2963 between the manager and the Open vSwitch. The connection must be
2964 reset for the new DSCP values to take effect. If no value is
2965 specified, a default value of 48 is chosen. Valid DSCP values must be
2966 in the range 0 to 63.
2970 <group title="Common Columns">
2971 The overall purpose of these columns is described under <code>Common
2972 Columns</code> at the beginning of this document.
2974 <column name="external_ids"/>
2975 <column name="other_config"/>
2979 <table name="NetFlow">
2980 A NetFlow target. NetFlow is a protocol that exports a number of
2981 details about terminating IP flows, such as the principals involved
2984 <column name="targets">
2985 NetFlow targets in the form
2986 <code><var>ip</var>:<var>port</var></code>. The <var>ip</var>
2987 must be specified numerically, not as a DNS name.
2990 <column name="engine_id">
2991 Engine ID to use in NetFlow messages. Defaults to datapath index
2995 <column name="engine_type">
2996 Engine type to use in NetFlow messages. Defaults to datapath
2997 index if not specified.
3000 <column name="active_timeout">
3001 The interval at which NetFlow records are sent for flows that are
3002 still active, in seconds. A value of <code>0</code> requests the
3003 default timeout (currently 600 seconds); a value of <code>-1</code>
3004 disables active timeouts.
3007 <column name="add_id_to_interface">
3008 <p>If this column's value is <code>false</code>, the ingress and egress
3009 interface fields of NetFlow flow records are derived from OpenFlow port
3010 numbers. When it is <code>true</code>, the 7 most significant bits of
3011 these fields will be replaced by the least significant 7 bits of the
3012 engine id. This is useful because many NetFlow collectors do not
3013 expect multiple switches to be sending messages from the same host, so
3014 they do not store the engine information which could be used to
3015 disambiguate the traffic.</p>
3016 <p>When this option is enabled, a maximum of 508 ports are supported.</p>
3019 <group title="Common Columns">
3020 The overall purpose of these columns is described under <code>Common
3021 Columns</code> at the beginning of this document.
3023 <column name="external_ids"/>
3028 SSL configuration for an Open_vSwitch.
3030 <column name="private_key">
3031 Name of a PEM file containing the private key used as the switch's
3032 identity for SSL connections to the controller.
3035 <column name="certificate">
3036 Name of a PEM file containing a certificate, signed by the
3037 certificate authority (CA) used by the controller and manager,
3038 that certifies the switch's private key, identifying a trustworthy
3042 <column name="ca_cert">
3043 Name of a PEM file containing the CA certificate used to verify
3044 that the switch is connected to a trustworthy controller.
3047 <column name="bootstrap_ca_cert">
3048 If set to <code>true</code>, then Open vSwitch will attempt to
3049 obtain the CA certificate from the controller on its first SSL
3050 connection and save it to the named PEM file. If it is successful,
3051 it will immediately drop the connection and reconnect, and from then
3052 on all SSL connections must be authenticated by a certificate signed
3053 by the CA certificate thus obtained. <em>This option exposes the
3054 SSL connection to a man-in-the-middle attack obtaining the initial
3055 CA certificate.</em> It may still be useful for bootstrapping.
3058 <group title="Common Columns">
3059 The overall purpose of these columns is described under <code>Common
3060 Columns</code> at the beginning of this document.
3062 <column name="external_ids"/>
3066 <table name="sFlow">
3067 <p>An sFlow(R) target. sFlow is a protocol for remote monitoring
3070 <column name="agent">
3071 Name of the network device whose IP address should be reported as the
3072 ``agent address'' to collectors. If not specified, the agent device is
3073 figured from the first target address and the routing table. If the
3074 routing table does not contain a route to the target, the IP address
3075 defaults to the <ref table="Controller" column="local_ip"/> in the
3076 collector's <ref table="Controller"/>. If an agent IP address cannot be
3077 determined any of these ways, sFlow is disabled.
3080 <column name="header">
3081 Number of bytes of a sampled packet to send to the collector.
3082 If not specified, the default is 128 bytes.
3085 <column name="polling">
3086 Polling rate in seconds to send port statistics to the collector.
3087 If not specified, defaults to 30 seconds.
3090 <column name="sampling">
3091 Rate at which packets should be sampled and sent to the collector.
3092 If not specified, defaults to 400, which means one out of 400
3093 packets, on average, will be sent to the collector.
3096 <column name="targets">
3097 sFlow targets in the form
3098 <code><var>ip</var>:<var>port</var></code>.
3101 <group title="Common Columns">
3102 The overall purpose of these columns is described under <code>Common
3103 Columns</code> at the beginning of this document.
3105 <column name="external_ids"/>