1 <database title="Open vSwitch Configuration Database">
2 <p>A database with this schema holds the configuration for one Open
3 vSwitch daemon. The root of the configuration for the daemon is
4 the <ref table="Open_vSwitch"/> table, which must have exactly one
5 record. Records in other tables are significant only when they
6 can be reached directly or indirectly from the
7 <ref table="Open_vSwitch"/> table.</p>
9 <table name="Open_vSwitch" title="Open vSwitch configuration.">
10 Configuration for an Open vSwitch daemon. There must be exactly one record
11 in the <ref table="Open_vSwitch"/> table.
13 <group title="Configuration">
14 <column name="bridges">
15 Set of bridges managed by the daemon.
18 <column name="controller">
19 Default OpenFlow <ref table="Controller"/> set used by bridges. May be
20 overridden on a per-bridge basis by the <ref table="Bridge"
21 column="controller"/> column in <ref table="Bridge"/>.
24 <column name="managers">
25 Remote database clients to which the Open vSwitch's database server
26 should connect or to which it should listen.
30 SSL used globally by the daemon.
33 <column name="external_ids">
34 Key-value pairs that identify this Open vSwitch's role in
35 external systems. The currently defined key-value pairs are:
37 <dt><code>system-uuid</code></dt>
38 <dd>A universally unique identifier for the Open vSwitch's
39 physical host. The form of the identifier depends on the
40 type of the host. On a Citrix XenServer, this is the host
41 UUID displayed by, e.g., <code>xe host-list</code>.</dd>
46 <group title="Status">
47 <column name="next_cfg">
48 Sequence number for client to increment. When a client modifies
49 any part of the database configuration and wishes to wait for
50 Open vSwitch to finish applying the changes, it may increment
54 <column name="cur_cfg">
55 Sequence number that Open vSwitch sets to the current value of
56 <ref column="next_cfg"/> after it finishes applying a set of
57 configuration changes.
60 <column name="capabilities">
61 Describes functionality supported by the hardware and software platform
62 on which this Open vSwitch is based. Clients should not modify this
63 column. See the <ref table="Capability"/> description for defined
64 capability categories and the meaning of associated
65 <ref table="Capability"/> records.
68 <column name="statistics">
70 Key-value pairs that report statistics about a running Open_vSwitch
71 daemon. The current implementation updates these counters
72 periodically. In the future, we plan to, instead, update them only
73 when they are queried (e.g. using an OVSDB <code>select</code>
74 operation) and perhaps at other times, but not on any regular
77 The currently defined key-value pairs are listed below. Some Open
78 vSwitch implementations may not support some statistics, in which
79 case those key-value pairs are omitted.</p>
81 <dt><code>load-average</code></dt>
83 System load average multiplied by 100 and rounded to the nearest
92 Configuration for a bridge within an
93 <ref table="Open_vSwitch"/>.
96 A <ref table="Bridge"/> record represents an Ethernet switch with one or
97 more ``ports,'' which are the <ref table="Port"/> records pointed to by
98 the <ref table="Bridge"/>'s <ref column="ports"/> column.
101 <group title="Core Features">
103 Bridge identifier. Should be alphanumeric and no more than about 8
104 bytes long. Must be unique among the names of ports, interfaces, and
108 <column name="ports">
109 Ports included in the bridge.
112 <column name="mirrors">
113 Port mirroring configuration.
116 <column name="netflow">
117 NetFlow configuration.
120 <column name="sflow">
124 <column name="flood_vlans">
125 VLAN IDs of VLANs on which MAC address learning should be disabled, so
126 that packets are flooded instead of being sent to specific ports that
127 are believed to contain packets' destination MACs. This should
128 ordinarily be used to disable MAC learning on VLANs used for mirroring
129 (RSPAN VLANs). It may also be useful for debugging.
133 <group title="OpenFlow Configuration">
134 <column name="controller">
135 OpenFlow controller set. If unset, defaults to the set of
136 controllers specified by <ref column="controller"
137 table="Open_vSwitch"/> in the <ref table="Open_vSwitch"/>
138 table. If the default is also unset, then no OpenFlow
139 controllers will be used.
142 <column name="datapath_id">
143 Reports the OpenFlow datapath ID in use. Exactly 16 hex
144 digits. (Setting this column will have no useful effect. Set
145 <ref column="other_config"/>:<code>other-config</code>
150 <group title="Other Features">
151 <column name="datapath_type">
152 Name of datapath provider. The kernel datapath has
153 type <code>system</code>. The userspace datapath has
154 type <code>netdev</code>.
157 <column name="external_ids">
158 Key-value pairs that identify this bridge's role in external systems.
159 The currently defined key-value pairs are:
161 <dt><code>network-uuids</code></dt>
162 <dd>Semicolon-delimited set of universally unique identifier(s) for
163 the network with which this bridge is associated. The form of the
164 identifier(s) depends on the type of the host. On a Citrix
165 XenServer host, the network identifiers are RFC 4122 UUIDs as
166 displayed by, e.g., <code>xe network-list</code>.</dd>
170 <column name="other_config">
171 Key-value pairs for configuring rarely used bridge
172 features. The currently defined key-value pairs are:
174 <dt><code>datapath-id</code></dt>
176 digits to set the OpenFlow datapath ID to a specific
178 <dt><code>hwaddr</code></dt>
179 <dd>An Ethernet address in the form
180 <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>
181 to set the hardware address of the local port and influence the
188 <table name="Port" table="Port or bond configuration.">
189 <p>A port within a <ref table="Bridge"/>.</p>
190 <p>Most commonly, a port has exactly one ``interface,'' pointed to by its
191 <ref column="interfaces"/> column. Such a port logically
192 corresponds to a port on a physical Ethernet switch. A port
193 with more than one interface is a ``bonded port'' (see
194 <ref group="Bonding Configuration"/>).</p>
195 <p>Some properties that one might think as belonging to a port are actually
196 part of the port's <ref table="Interface"/> members.</p>
199 Port name. Should be alphanumeric and no more than about 8
200 bytes long. May be the same as the interface name, for
201 non-bonded ports. Must otherwise be unique among the names of
202 ports, interfaces, and bridges on a host.
205 <column name="interfaces">
206 The port's interfaces. If there is more than one, this is a
210 <group title="VLAN Configuration">
211 <p>A bridge port must be configured for VLANs in one of two
212 mutually exclusive ways:
214 <li>A ``trunk port'' has an empty value for <ref
215 column="tag"/>. Its <ref column="trunks"/> value may be
216 empty or non-empty.</li>
217 <li>An ``implicitly tagged VLAN port'' or ``access port''
218 has an nonempty value for <ref column="tag"/>. Its
219 <ref column="trunks"/> value must be empty.</li>
221 If <ref column="trunks"/> and <ref column="tag"/> are both
222 nonempty, the configuration is ill-formed.
227 If this is an access port (see above), the port's implicitly
228 tagged VLAN. Must be empty if this is a trunk port.
231 Frames arriving on trunk ports will be forwarded to this
232 port only if they are tagged with the given VLAN (or, if
233 <ref column="tag"/> is 0, then if they lack a VLAN header).
234 Frames arriving on other access ports will be forwarded to
235 this port only if they have the same <ref column="tag"/>
236 value. Frames forwarded to this port will not have an
240 When a frame with a 802.1Q header that indicates a nonzero
241 VLAN is received on an access port, it is discarded.
245 <column name="trunks">
247 If this is a trunk port (see above), the 802.1Q VLAN(s) that
248 this port trunks; if it is empty, then the port trunks all
249 VLANs. Must be empty if this is an access port.
252 Frames arriving on trunk ports are dropped if they are not
253 in one of the specified VLANs. For this purpose, packets
254 that have no VLAN header are treated as part of VLAN 0.
259 <group title="Bonding Configuration">
260 <p>A port that has more than one interface is a ``bonded port.''
261 Bonding allows for load balancing and fail-over. Open vSwitch
262 supports ``source load balancing'' (SLB) bonding, which
263 assigns flows to slaves based on source MAC address, with
264 periodic rebalancing as traffic patterns change. This form of
265 bonding does not require 802.3ad or other special support from
266 the upstream switch to which the slave devices are
269 <p>These columns apply only to bonded ports. Their values are
270 otherwise ignored.</p>
272 <column name="bond_updelay">
273 <p>For a bonded port, the number of milliseconds for which carrier must
274 stay up on an interface before the interface is considered to be up.
275 Specify <code>0</code> to enable the interface immediately.</p>
276 <p>This setting is honored only when at least one bonded interface is
277 already enabled. When no interfaces are enabled, then the first bond
278 interface to come up is enabled immediately.</p>
281 <column name="bond_downdelay">
282 For a bonded port, the number of milliseconds for which carrier must
283 stay down on an interface before the interface is considered to be
284 down. Specify <code>0</code> to disable the interface immediately.
287 <column name="bond_fake_iface">
288 For a bonded port, whether to create a fake internal interface with the
289 name of the port. Use only for compatibility with legacy software that
294 <group title="Other Features">
296 Quality of Service configuration for this port.
300 The MAC address to use for this port for the purpose of choosing the
301 bridge's MAC address. This column does not necessarily reflect the
302 port's actual MAC address, nor will setting it change the port's actual
306 <column name="fake_bridge">
307 Does this port represent a sub-bridge for its tagged VLAN within the
308 Bridge? See ovs-vsctl(8) for more information.
311 <column name="external_ids">
312 Key-value pairs that identify this port's role in external systems. No
313 key-value pairs native to <ref table="Port"/> are currently defined.
314 For fake bridges (see the <ref column="fake_bridge"/> column), external
315 IDs for the fake bridge are defined here by prefixing a
316 <ref table="Bridge"/> <ref table="Bridge" column="external_ids"/> key
317 with <code>fake-bridge-</code>,
318 e.g. <code>fake-bridge-network-uuids</code>.
321 <column name="other_config">
322 Key-value pairs for configuring rarely used port features. The
323 currently defined key-value pairs are:
325 <dt><code>hwaddr</code></dt>
326 <dd>An Ethernet address in the form
327 <code><var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var></code>.</dd>
328 <dt><code>bond-rebalance-interval</code></dt>
329 <dd>For a bonded port, the number of milliseconds between
330 successive attempts to rebalance the bond, that is, to
331 move source MACs and their flows from one interface on
332 the bond to another in an attempt to keep usage of each
333 interface roughly equal. The default is 10000 (10
334 seconds), and the minimum is 1000 (1 second).</dd>
340 <table name="Interface" title="One physical network device in a Port.">
341 An interface within a <ref table="Port"/>.
343 <group title="Core Features">
345 Interface name. Should be alphanumeric and no more than about 8 bytes
346 long. May be the same as the port name, for non-bonded ports. Must
347 otherwise be unique among the names of ports, interfaces, and bridges
352 <p>Ethernet address to set for this interface. If unset then the
353 default MAC address is used:</p>
355 <li>For the local interface, the default is the lowest-numbered MAC
356 address among the other bridge ports, either the value of the
357 <ref table="Port" column="mac"/> in its <ref table="Port"/> record,
358 if set, or its actual MAC (for bonded ports, the MAC of its slave
359 whose name is first in alphabetical order). Internal ports and
360 bridge ports that are used as port mirroring destinations (see the
361 <ref table="Mirror"/> table) are ignored.</li>
362 <li>For other internal interfaces, the default MAC is randomly
364 <li>External interfaces typically have a MAC address associated with
367 <p>Some interfaces may not have a software-controllable MAC
371 <column name="ofport">
372 <p>OpenFlow port number for this interface. Unlike most columns, this
373 column's value should be set only by Open vSwitch itself. Other
374 clients should set this column to an empty set (the default) when
375 creating an <ref table="Interface"/>.</p>
376 <p>Open vSwitch populates this column when the port number becomes
377 known. If the interface is successfully added,
378 <ref column="ofport"/> will be set to a number between 1 and 65535
379 (generally either in the range 1 to 65280, exclusive, or 65534, the
380 port number for the OpenFlow ``local port''). If the interface
381 cannot be added then Open vSwitch sets this column
386 <group title="System-Specific Details">
388 The interface type, one of:
390 <dt><code>system</code></dt>
391 <dd>An ordinary network device, e.g. <code>eth0</code> on Linux.
392 Sometimes referred to as ``external interfaces'' since they are
393 generally connected to hardware external to that on which the Open
394 vSwitch is running. The empty string is a synonym for
395 <code>system</code>.</dd>
396 <dt><code>internal</code></dt>
397 <dd>A simulated network device that sends and receives traffic. An
398 internal interface whose <ref column="name"/> is the same as its
399 bridge's <ref table="Open_vSwitch" column="name"/> is called the
400 ``local interface.'' It does not make sense to bond an internal
401 interface, so the terms ``port'' and ``interface'' are often used
402 imprecisely for internal interfaces.</dd>
403 <dt><code>tap</code></dt>
404 <dd>A TUN/TAP device managed by Open vSwitch.</dd>
405 <dt><code>gre</code></dt>
406 <dd>An Ethernet over RFC 1702 Generic Routing Encapsulation over IPv4
407 tunnel. Each tunnel must be uniquely identified by the
408 combination of <code>remote_ip</code>, <code>local_ip</code>, and
409 <code>in_key</code>. Note that if two ports are defined that are
410 the same except one has an optional identifier and the other does
411 not, the more specific one is matched first. <code>in_key</code>
412 is considered more specific than <code>local_ip</code> if a port
413 defines one and another port defines the other. The arguments
416 <dt><code>remote_ip</code></dt>
417 <dd>Required. The tunnel endpoint.</dd>
420 <dt><code>local_ip</code></dt>
421 <dd>Optional. The destination IP that received packets must
422 match. Default is to match all addresses.</dd>
425 <dt><code>in_key</code></dt>
426 <dd>Optional. The GRE key that received packets must contain.
427 It may either be a 32-bit number (no key and a key of 0 are
428 treated as equivalent) or the word <code>flow</code>. If
429 <code>flow</code> is specified then any key will be accepted
430 and the key will be placed in the <code>tun_id</code> field
431 for matching in the flow table. The ovs-ofctl manual page
432 contains additional information about matching fields in
433 OpenFlow flows. Default is no key.</dd>
436 <dt><code>out_key</code></dt>
437 <dd>Optional. The GRE key to be set on outgoing packets. It may
438 either be a 32-bit number or the word <code>flow</code>. If
439 <code>flow</code> is specified then the key may be set using
440 the <code>set_tunnel</code> Nicira OpenFlow vendor extension (0
441 is used in the absense of an action). The ovs-ofctl manual
442 page contains additional information about the Nicira OpenFlow
443 vendor extensions. Default is no key.</dd>
446 <dt><code>key</code></dt>
447 <dd>Optional. Shorthand to set <code>in_key</code> and
448 <code>out_key</code> at the same time.</dd>
451 <dt><code>tos</code></dt>
452 <dd>Optional. The value of the ToS bits to be set on the
453 encapsulating packet. It may also be the word
454 <code>inherit</code>, in which case the ToS will be copied from
455 the inner packet if it is IPv4 or IPv6 (otherwise it will be
456 0). Note that the ECN fields are always inherited. Default is
460 <dt><code>ttl</code></dt>
461 <dd>Optional. The TTL to be set on the encapsulating packet.
462 It may also be the word <code>inherit</code>, in which case the
463 TTL will be copied from the inner packet if it is IPv4 or IPv6
464 (otherwise it will be the system default, typically 64).
465 Default is the system default TTL.</dd>
468 <dt><code>csum</code></dt>
469 <dd>Optional. Compute GRE checksums for outgoing packets and
470 require checksums for incoming packets. Default is enabled,
471 set to <code>false</code> to disable.</dd>
474 <dt><code>pmtud</code></dt>
475 <dd>Optional. Enable tunnel path MTU discovery. If enabled
476 ``ICMP destination unreachable - fragmentation'' needed
477 messages will be generated for IPv4 packets with the DF bit set
478 and IPv6 packets above the minimum MTU if the packet size
479 exceeds the path MTU minus the size of the tunnel headers. It
480 also forces the encapsulating packet DF bit to be set (it is
481 always set if the inner packet implies path MTU discovery).
482 Note that this option causes behavior that is typically
483 reserved for routers and therefore is not entirely in
484 compliance with the IEEE 802.1D specification for bridges.
485 Default is enabled, set to <code>false</code> to disable.</dd>
488 <dt><code>patch</code></dt>
489 <dd>A pair of virtual devices that act as a patch cable. A
490 <code>peer</code> argument is required that indicates the name
491 of the other side of the patch. Since a patch must work in
492 pairs, a second patch interface must be declared with the
493 <code>name</code> and <code>peer</code> arguments reversed.</dd>
497 <column name="options">
498 Configuration options whose interpretation varies based on
499 <ref column="type"/>.
503 <group title="Ingress Policing">
504 <column name="ingress_policing_burst">
505 <p>Maximum burst size for data received on this interface, in kb. The
506 default burst size if set to <code>0</code> is 1000 kb. This value
507 has no effect if <ref column="ingress_policing_rate"/>
508 is <code>0</code>.</p>
509 <p>The burst size should be at least the size of the interface's
513 <column name="ingress_policing_rate">
514 <p>Maximum rate for data received on this interface, in kbps. Data
515 received faster than this rate is dropped. Set to <code>0</code> to
516 disable policing.</p>
517 <p>The meaning of ``ingress'' is from Open vSwitch's perspective. If
518 configured on a physical interface, then it limits the rate at which
519 traffic is allowed into the system from the outside. If configured
520 on a virtual interface that is connected to a virtual machine, then
521 it limits the rate at which the guest is able to transmit.</p>
525 <group title="Other Features">
526 <column name="external_ids">
527 <p>Key-value pairs that identify this interface's role in external
528 systems. All of the currently defined key-value pairs specifically
529 apply to an interface that represents a virtual Ethernet interface
530 connected to a virtual machine. These key-value pairs should not be
531 present for other types of interfaces. Keys whose names end
532 in <code>-uuid</code> have values that uniquely identify the entity
533 in question. For a Citrix XenServer hypervisor, these values are
534 UUIDs in RFC 4122 format. Other hypervisors may use other
536 <p>The currently defined key-value pairs are:</p>
538 <dt><code>vif-uuid</code></dt>
539 <dd>The virtual interface associated with this interface.</dd>
540 <dt><code>network-uuid</code></dt>
541 <dd>The virtual network to which this interface is attached.</dd>
542 <dt><code>vm-uuid</code></dt>
543 <dd>The VM to which this interface belongs.</dd>
544 <dt><code>vif-mac</code></dt>
545 <dd>The MAC address programmed into the "virtual hardware" for this
547 form <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>.
548 For Citrix XenServer, this is the value of the <code>MAC</code>
549 field in the VIF record for this interface.</dd>
553 <column name="statistics">
555 Key-value pairs that report interface statistics. The current
556 implementation updates these counters periodically. In the future,
557 we plan to, instead, update them when an interface is created, when
558 they are queried (e.g. using an OVSDB <code>select</code> operation),
559 and just before an interface is deleted due to virtual interface
560 hot-unplug or VM shutdown, and perhaps at other times, but not on any
561 regular periodic basis.</p>
563 The currently defined key-value pairs are listed below. These are
564 the same statistics reported by OpenFlow in its <code>struct
565 ofp_port_stats</code> structure. If an interface does not support a
566 given statistic, then that pair is omitted.</p>
569 Successful transmit and receive counters:
571 <dt><code>rx_packets</code></dt>
572 <dd>Number of received packets.</dd>
573 <dt><code>rx_bytes</code></dt>
574 <dd>Number of received bytes.</dd>
575 <dt><code>tx_packets</code></dt>
576 <dd>Number of transmitted packets.</dd>
577 <dt><code>tx_bytes</code></dt>
578 <dd>Number of transmitted bytes.</dd>
584 <dt><code>rx_dropped</code></dt>
585 <dd>Number of packets dropped by RX.</dd>
586 <dt><code>rx_frame_err</code></dt>
587 <dd>Number of frame alignment errors.</dd>
588 <dt><code>rx_over_err</code></dt>
589 <dd>Number of packets with RX overrun.</dd>
590 <dt><code>rx_crc_err</code></dt>
591 <dd>Number of CRC errors.</dd>
592 <dt><code>rx_errors</code></dt>
594 Total number of receive errors, greater than or equal
595 to the sum of the above.
602 <dt><code>tx_dropped</code></dt>
603 <dd>Number of packets dropped by TX.</dd>
604 <dt><code>collisions</code></dt>
605 <dd>Number of collisions.</dd>
606 <dt><code>tx_errors</code></dt>
608 Total number of transmit errors, greater
609 than or equal to the sum of the above.
618 <table name="QoS" title="Quality of Service configuration">
619 <p>Quality of Service (QoS) configuration for each Port that
623 <p>The type of QoS to implement. The <ref table="Open_vSwitch"
624 column="capabilities"/> column in the <ref table="Open_vSwitch"/> table
625 identifies the types that a switch actually supports. The currently
626 defined types are listed below:</p>
628 <dt><code>linux-htb</code></dt>
629 <dd>Linux ``hierarchy token bucket'' classifier.</dd>
633 <column name="queues">
634 <p>A map from queue numbers to <ref table="Queue"/> records. The
635 supported range of queue numbers depend on <ref column="type"/>. The
636 queue numbers are the same as the <code>queue_id</code> used in
637 OpenFlow in <code>struct ofp_action_enqueue</code> and other
638 structures. Queue 0 is used by OpenFlow output actions that do not
639 specify a specific queue.</p>
642 <column name="other_config">
643 <p>Key-value pairs for configuring QoS features that depend on
644 <ref column="type"/>.</p>
645 <p>The <code>linux-htb</code> class supports the following key-value
648 <dt><code>max-rate</code></dt>
649 <dd>Maximum rate shared by all queued traffic, in bit/s.
650 Optional. If not specified, for physical interfaces, the
651 default is the link rate. For other interfaces or if the
652 link rate cannot be determined, the default is currently 100
658 <table name="Queue" title="QoS output queue.">
659 <p>A configuration for a port output queue, used in configuring Quality of
660 Service (QoS) features. May be referenced by <ref column="queues"
661 table="QoS"/> column in <ref table="QoS"/> table.</p>
663 <column name="other_config">
664 <p>Key-value pairs for configuring the output queue. The supported
665 key-value pairs and their meanings depend on the <ref column="type"/>
666 of the <ref column="QoS"/> records that reference this row.</p>
667 <p>The key-value pairs defined for <ref table="QoS"/> <ref table="QoS"
668 column="type"/> of <code>min-rate</code> are:</p>
670 <dt><code>min-rate</code></dt>
671 <dd>Minimum guaranteed bandwidth, in bit/s. Required.</dd>
673 <p>The key-value pairs defined for <ref table="QoS"/> <ref table="QoS"
674 column="type"/> of <code>linux-htb</code> are:</p>
676 <dt><code>min-rate</code></dt>
677 <dd>Minimum guaranteed bandwidth, in bit/s. Required.</dd>
678 <dt><code>max-rate</code></dt>
679 <dd>Maximum allowed bandwidth, in bit/s. Optional. If specified, the
680 queue's rate will not be allowed to exceed the specified value, even
681 if excess bandwidth is available. If unspecified, defaults to no
683 <dt><code>burst</code></dt>
684 <dd>Burst size, in bits. This is the maximum amount of ``credits''
685 that a queue can accumulate while it is idle. Optional. Details of
686 the <code>linux-htb</code> implementation require a minimum burst
687 size, so a too-small <code>burst</code> will be silently
689 <dt><code>priority</code></dt>
690 <dd>A nonnegative 32-bit integer. Defaults to 0 if
691 unspecified. A queue with a smaller <code>priority</code>
692 will receive all the excess bandwidth that it can use before
693 a queue with a larger value receives any. Specific priority
694 values are unimportant; only relative ordering matters.</dd>
699 <table name="Mirror" title="Port mirroring (SPAN/RSPAN).">
700 <p>A port mirror within a <ref table="Bridge"/>.</p>
701 <p>A port mirror configures a bridge to send selected frames to special
702 ``mirrored'' ports, in addition to their normal destinations. Mirroring
703 traffic may also be referred to as SPAN or RSPAN, depending on the
704 mechanism used for delivery.</p>
707 Arbitrary identifier for the <ref table="Mirror"/>.
710 <group title="Selecting Packets for Mirroring">
711 <column name="select_all">
712 If true, every packet arriving or departing on any port is
713 selected for mirroring.
716 <column name="select_dst_port">
717 Ports on which departing packets are selected for mirroring.
720 <column name="select_src_port">
721 Ports on which arriving packets are selected for mirroring.
724 <column name="select_vlan">
725 VLANs on which packets are selected for mirroring. An empty set
726 selects packets on all VLANs.
730 <group title="Mirroring Destination Configuration">
731 <column name="output_port">
732 <p>Output port for selected packets, if nonempty. Mutually exclusive
733 with <ref column="output_vlan"/>.</p>
734 <p>Specifying a port for mirror output reserves that port exclusively
735 for mirroring. No frames other than those selected for mirroring
736 will be forwarded to the port, and any frames received on the port
737 will be discarded.</p>
738 <p>This type of mirroring is sometimes called SPAN.</p>
741 <column name="output_vlan">
742 <p>Output VLAN for selected packets, if nonempty. Mutually exclusive
743 with <ref column="output_port"/>.</p>
744 <p>The frames will be sent out all ports that trunk
745 <ref column="output_vlan"/>, as well as any ports with implicit VLAN
746 <ref column="output_vlan"/>. When a mirrored frame is sent out a
747 trunk port, the frame's VLAN tag will be set to
748 <ref column="output_vlan"/>, replacing any existing tag; when it is
749 sent out an implicit VLAN port, the frame will not be tagged. This
750 type of mirroring is sometimes called RSPAN.</p>
751 <p><em>Please note:</em> Mirroring to a VLAN can disrupt a network that
752 contains unmanaged switches. Consider an unmanaged physical switch
753 with two ports: port 1, connected to an end host, and port 2,
754 connected to an Open vSwitch configured to mirror received packets
755 into VLAN 123 on port 2. Suppose that the end host sends a packet on
756 port 1 that the physical switch forwards to port 2. The Open vSwitch
757 forwards this packet to its destination and then reflects it back on
758 port 2 in VLAN 123. This reflected packet causes the unmanaged
759 physical switch to replace the MAC learning table entry, which
760 correctly pointed to port 1, with one that incorrectly points to port
761 2. Afterward, the physical switch will direct packets destined for
762 the end host to the Open vSwitch on port 2, instead of to the end
763 host on port 1, disrupting connectivity. If mirroring to a VLAN is
764 desired in this scenario, then the physical switch must be replaced
765 by one that learns Ethernet addresses on a per-VLAN basis. In
766 addition, learning should be disabled on the VLAN containing mirrored
767 traffic. If this is not done then intermediate switches will learn
768 the MAC address of each end host from the mirrored traffic. If
769 packets being sent to that end host are also mirrored, then they will
770 be dropped since the switch will attempt to send them out the input
771 port. Disabling learning for the VLAN will cause the switch to
772 correctly send the packet out all ports configured for that VLAN. If
773 Open vSwitch is being used as an intermediate switch, learning can be
774 disabled by adding the mirrored VLAN to <ref column="flood_vlans"/>
775 in the appropriate <ref table="Bridge"/> table or tables.</p>
780 <table name="Controller" title="OpenFlow controller configuration.">
781 <p>An OpenFlow controller.</p>
783 <p>Open vSwitch permits a bridge to have any number of OpenFlow
784 controllers. When multiple controllers are configured, Open vSwitch
785 connects to all of them simultaneously. OpenFlow 1.0 does not specify
786 how multiple controllers coordinate in interacting with a single switch,
787 so more than one controller should be specified only if the controllers
788 are themselves designed to coordinate with each other.</p>
790 <group title="Core Features">
791 <column name="target">
792 <p>Connection method for controller.
793 The following connection methods are currently
796 <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
798 <p>The specified SSL <var>port</var> (default: 6633) on the host at
799 the given <var>ip</var>, which must be expressed as an IP address
800 (not a DNS name). The <ref table="Open_vSwitch" column="ssl"/>
801 column in the <ref table="Open_vSwitch"/> must point to a valid
802 SSL configuration when this form is used.</p>
803 <p>SSL support is an optional feature that is not always built as
804 part of Open vSwitch.</p>
806 <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt>
807 <dd>The specified TCP <var>port</var> (default: 6633) on the host at
808 the given <var>ip</var>, which must be expressed as an IP address
809 (not a DNS name).</dd>
810 <dt><code>discover</code></dt>
812 <p>Enables controller discovery.</p>
813 <p>In controller discovery mode, Open vSwitch broadcasts a DHCP
814 request with vendor class identifier <code>OpenFlow</code> across
815 all of the bridge's network devices. It will accept any valid
816 DHCP reply that has the same vendor class identifier and includes
817 a vendor-specific option with code 1 whose contents are a string
818 specifying the location of the controller in the same format as
819 <ref column="target"/>.</p>
820 <p>The DHCP reply may also, optionally, include a vendor-specific
821 option with code 2 whose contents are a string specifying the URI
822 to the base of the OpenFlow PKI
823 (e.g. <code>http://192.168.0.1/openflow/pki</code>). This URI is
824 used only for bootstrapping the OpenFlow PKI at initial switch
825 setup; <code>ovs-vswitchd</code> does not use it at all.</p>
827 <dt><code>none</code></dt>
828 <dd>Disables the controller.</dd>
830 <p>When multiple controllers are configured for a single bridge, the
831 <ref column="target"/> values must be unique. Duplicate
832 <ref column="target"/> values yield unspecified results.</p>
835 <column name="connection_mode">
836 <p>If it is specified, this setting must be one of the following
837 strings that describes how Open vSwitch contacts this OpenFlow
838 controller over the network:</p>
841 <dt><code>in-band</code></dt>
842 <dd>In this mode, this controller's OpenFlow traffic travels over the
843 bridge associated with the controller. With this setting, Open
844 vSwitch allows traffic to and from the controller regardless of the
845 contents of the OpenFlow flow table. (Otherwise, Open vSwitch
846 would never be able to connect to the controller, because it did
847 not have a flow to enable it.) This is the most common connection
848 mode because it is not necessary to maintain two independent
850 <dt><code>out-of-band</code></dt>
851 <dd>In this mode, OpenFlow traffic uses a control network separate
852 from the bridge associated with this controller, that is, the
853 bridge does not use any of its own network devices to communicate
854 with the controller. The control network must be configured
855 separately, before or after <code>ovs-vswitchd</code> is started.
859 <p>If not specified, the default is implementation-specific. If
860 <ref column="target"/> is <code>discover</code>, the connection mode
861 is always treated as <code>in-band</code> regardless of the actual
866 <group title="Controller Failure Detection and Handling">
867 <column name="max_backoff">
868 Maximum number of milliseconds to wait between connection attempts.
869 Default is implementation-specific.
872 <column name="inactivity_probe">
873 Maximum number of milliseconds of idle time on connection to
874 controller before sending an inactivity probe message. If Open
875 vSwitch does not communicate with the controller for the specified
876 number of seconds, it will send a probe. If a response is not
877 received for the same additional amount of time, Open vSwitch
878 assumes the connection has been broken and attempts to reconnect.
879 Default is implementation-specific.
882 <column name="fail_mode">
883 <p>When a controller is configured, it is, ordinarily, responsible
884 for setting up all flows on the switch. Thus, if the connection to
885 the controller fails, no new network connections can be set up.
886 If the connection to the controller stays down long enough,
887 no packets can pass through the switch at all. This setting
888 determines the switch's response to such a situation. It may be set
889 to one of the following:
891 <dt><code>standalone</code></dt>
892 <dd>If no message is received from the controller for three
893 times the inactivity probe interval
894 (see <ref column="inactivity_probe"/>), then Open vSwitch
895 will take over responsibility for setting up flows. In
896 this mode, Open vSwitch causes the bridge to act like an
897 ordinary MAC-learning switch. Open vSwitch will continue
898 to retry connecting to the controller in the background
899 and, when the connection succeeds, it will discontinue its
900 standalone behavior.</dd>
901 <dt><code>secure</code></dt>
902 <dd>Open vSwitch will not set up flows on its own when the
903 controller connection fails. It will continue retry
904 connecting to the controller forever.</dd>
907 <p>If this value is unset, the default is implementation-specific.</p>
908 <p>When more than one controller is configured,
909 <ref column="fail_mode"/> is considered only when none of the
910 configured controllers can be contacted. At that point, the bridge
911 enters secure mode if any of the controllers'
912 <ref column="fail_mode"/> is set to <code>secure</code>. Otherwise,
913 it enters standalone mode if at least one <ref column="fail_mode"/>
914 is set to <code>standalone</code>. If none of the
915 <ref column="fail_mode"/> values are set, the default is
916 implementation-defined.</p>
920 <group title="OpenFlow Rate Limiting">
921 <column name="controller_rate_limit">
922 <p>The maximum rate at which packets in unknown flows will be
923 forwarded to the OpenFlow controller, in packets per second. This
924 feature prevents a single bridge from overwhelming the controller.
925 If not specified, the default is implementation-specific.</p>
926 <p>In addition, when a high rate triggers rate-limiting, Open
927 vSwitch queues controller packets for each port and transmits
928 them to the controller at the configured rate. The number of
929 queued packets is limited by
930 the <ref column="controller_burst_limit"/> value. The packet
931 queue is shared fairly among the ports on a bridge.</p><p>Open
932 vSwitch maintains two such packet rate-limiters per bridge.
933 One of these applies to packets sent up to the controller
934 because they do not correspond to any flow. The other applies
935 to packets sent up to the controller by request through flow
936 actions. When both rate-limiters are filled with packets, the
937 actual rate that packets are sent to the controller is up to
938 twice the specified rate.</p>
941 <column name="controller_burst_limit">
942 In conjunction with <ref column="controller_rate_limit"/>,
943 the maximum number of unused packet credits that the bridge will
944 allow to accumulate, in packets. If not specified, the default
945 is implementation-specific.
949 <group title="Additional Discovery Configuration">
950 <p>These values are considered only when <ref column="target"/>
951 is <code>discover</code>.</p>
953 <column name="discover_accept_regex">
955 extended regular expression against which the discovered controller
956 location is validated. The regular expression is implicitly
957 anchored at the beginning of the controller location string, as
958 if it begins with <code>^</code>. If not specified, the default
959 is implementation-specific.
962 <column name="discover_update_resolv_conf">
963 Whether to update <code>/etc/resolv.conf</code> when the
964 controller is discovered. If not specified, the default
965 is implementation-specific. Open vSwitch will only modify
966 <code>/etc/resolv.conf</code> if the DHCP response that it receives
967 specifies one or more DNS servers.
971 <group title="Additional In-Band Configuration">
972 <p>These values are considered only in in-band control mode (see
973 <ref column="connection_mode"/>) and only when <ref column="target"/>
974 is not <code>discover</code>. (For controller discovery, the network
975 configuration obtained via DHCP is used instead.)</p>
977 <p>When multiple controllers are configured on a single bridge, there
978 should be only one set of unique values in these columns. If different
979 values are set for these columns in different controllers, the effect
982 <column name="local_ip">
983 The IP address to configure on the local port,
984 e.g. <code>192.168.0.123</code>. If this value is unset, then
985 <ref column="local_netmask"/> and <ref column="local_gateway"/> are
989 <column name="local_netmask">
990 The IP netmask to configure on the local port,
991 e.g. <code>255.255.255.0</code>. If <ref column="local_ip"/> is set
992 but this value is unset, then the default is chosen based on whether
993 the IP address is class A, B, or C.
996 <column name="local_gateway">
997 The IP address of the gateway to configure on the local port, as a
998 string, e.g. <code>192.168.0.1</code>. Leave this column unset if
999 this network has no gateway.
1004 <table name="NetFlow">
1005 A NetFlow target. NetFlow is a protocol that exports a number of
1006 details about terminating IP flows, such as the principals involved
1009 <column name="targets">
1010 NetFlow targets in the form
1011 <code><var>ip</var>:<var>port</var></code>. The <var>ip</var>
1012 must be specified numerically, not as a DNS name.
1015 <column name="engine_id">
1016 Engine ID to use in NetFlow messages. Defaults to datapath index
1020 <column name="engine_type">
1021 Engine type to use in NetFlow messages. Defaults to datapath
1022 index if not specified.
1025 <column name="active_timeout">
1026 The interval at which NetFlow records are sent for flows that are
1027 still active, in seconds. A value of <code>0</code> requests the
1028 default timeout (currently 600 seconds); a value of <code>-1</code>
1029 disables active timeouts.
1032 <column name="add_id_to_interface">
1033 <p>If this column's value is <code>false</code>, the ingress and egress
1034 interface fields of NetFlow flow records are derived from OpenFlow port
1035 numbers. When it is <code>true</code>, the 7 most significant bits of
1036 these fields will be replaced by the least significant 7 bits of the
1037 engine id. This is useful because many NetFlow collectors do not
1038 expect multiple switches to be sending messages from the same host, so
1039 they do not store the engine information which could be used to
1040 disambiguate the traffic.</p>
1041 <p>When this option is enabled, a maximum of 508 ports are supported.</p>
1046 SSL configuration for an Open_vSwitch.
1048 <column name="private_key">
1049 Name of a PEM file containing the private key used as the switch's
1050 identity for SSL connections to the controller.
1053 <column name="certificate">
1054 Name of a PEM file containing a certificate, signed by the
1055 certificate authority (CA) used by the controller and manager,
1056 that certifies the switch's private key, identifying a trustworthy
1060 <column name="ca_cert">
1061 Name of a PEM file containing the CA certificate used to verify
1062 that the switch is connected to a trustworthy controller.
1065 <column name="bootstrap_ca_cert">
1066 If set to <code>true</code>, then Open vSwitch will attempt to
1067 obtain the CA certificate from the controller on its first SSL
1068 connection and save it to the named PEM file. If it is successful,
1069 it will immediately drop the connection and reconnect, and from then
1070 on all SSL connections must be authenticated by a certificate signed
1071 by the CA certificate thus obtained. <em>This option exposes the
1072 SSL connection to a man-in-the-middle attack obtaining the initial
1073 CA certificate.</em> It may still be useful for bootstrapping.
1077 <table name="sFlow">
1078 <p>An sFlow(R) target. sFlow is a protocol for remote monitoring
1081 <column name="agent">
1082 Name of the network device whose IP address should be reported as the
1083 ``agent address'' to collectors. If not specified, the IP address
1084 defaults to the <ref table="Controller" column="local_ip"/> in the
1085 collector's <ref table="Controller"/>. If an agent IP address cannot be
1086 determined either way, sFlow is disabled.
1089 <column name="header">
1090 Number of bytes of a sampled packet to send to the collector.
1091 If not specified, the default is 128 bytes.
1094 <column name="polling">
1095 Polling rate in seconds to send port statistics to the collector.
1096 If not specified, defaults to 30 seconds.
1099 <column name="sampling">
1100 Rate at which packets should be sampled and sent to the collector.
1101 If not specified, defaults to 400, which means one out of 400
1102 packets, on average, will be sent to the collector.
1105 <column name="targets">
1106 sFlow targets in the form
1107 <code><var>ip</var>:<var>port</var></code>.
1111 <table name="Capability">
1112 <p>Records in this table describe functionality supported by the hardware
1113 and software platform on which this Open vSwitch is based. Clients
1114 should not modify this table.</p>
1116 <p>A record in this table is meaningful only if it is referenced by the
1117 <ref table="Open_vSwitch" column="capabilities"/> column in the
1118 <ref table="Open_vSwitch"/> table. The key used to reference it, called
1119 the record's ``category,'' determines the meanings of the
1120 <ref column="details"/> column. The following general forms of
1121 categories are currently defined:</p>
1124 <dt><code>qos-<var>type</var></code></dt>
1125 <dd><var>type</var> is supported as the value for
1126 <ref column="type" table="QoS"/> in the <ref table="QoS"/> table.
1130 <column name="details">
1131 <p>Key-value pairs that describe capabilities. The meaning of the pairs
1132 depends on the category key that the <ref table="Open_vSwitch"
1133 column="capabilities"/> column in the <ref table="Open_vSwitch"/> table
1134 uses to reference this record, as described above.</p>
1136 <p>The presence of a record for category <code>qos-<var>type</var></code>
1137 indicates that the switch supports <var>type</var> as the value of
1138 the <ref table="QoS" column="type"/> column in the <ref table="QoS"/>
1139 table. The following key-value pairs are defined to further describe
1140 QoS capabilities:</p>
1143 <dt><code>n-queues</code></dt>
1144 <dd>Number of supported queues, as a positive integer. Keys in the
1145 <ref table="QoS" column="queues"/> column for <ref table="QoS"/>
1146 records whose <ref table="QoS" column="type"/> value
1147 equals <var>type</var> must range between 0 and this value minus one,