]>
Commit | Line | Data |
---|---|---|
ce887677 | 1 | <?xml version="1.0" encoding="utf-8"?> |
57ba0a77 | 2 | <database name="ovs-vswitchd.conf.db" title="Open vSwitch Configuration Database"> |
c5f341ab BP |
3 | <p> |
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 | |
89365653 | 7 | record. Records in other tables are significant only when they |
c5f341ab BP |
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 | |
3fd8d445 | 12 | ``root set'' tables. |
c5f341ab | 13 | </p> |
89365653 | 14 | |
3fd8d445 BP |
15 | <h2>Common Columns</h2> |
16 | ||
17 | <p> | |
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 | |
21 | later. | |
22 | </p> | |
23 | ||
24 | <dl> | |
25 | <dt><code>other_config</code>: map of string-string pairs</dt> | |
26 | <dd> | |
27 | <p> | |
28 | Key-value pairs for configuring rarely used features. Supported keys, | |
29 | along with the forms taken by their values, are documented individually | |
30 | for each table. | |
31 | </p> | |
32 | <p> | |
33 | A few tables do not have <code>other_config</code> columns because no | |
34 | key-value pairs have yet been defined for them. | |
35 | </p> | |
36 | </dd> | |
37 | ||
38 | <dt><code>external_ids</code>: map of string-string pairs</dt> | |
39 | <dd> | |
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 | |
46 | table. | |
47 | </dd> | |
48 | </dl> | |
49 | ||
89365653 | 50 | <table name="Open_vSwitch" title="Open vSwitch configuration."> |
c5f341ab BP |
51 | Configuration for an Open vSwitch daemon. There must be exactly |
52 | one record in the <ref table="Open_vSwitch"/> table. | |
89365653 BP |
53 | |
54 | <group title="Configuration"> | |
55 | <column name="bridges"> | |
56 | Set of bridges managed by the daemon. | |
57 | </column> | |
58 | ||
89365653 BP |
59 | <column name="ssl"> |
60 | SSL used globally by the daemon. | |
61 | </column> | |
f5e7ed5d | 62 | |
3fd8d445 BP |
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"/>. | |
3fe80505 BP |
68 | </column> |
69 | ||
3fd8d445 BP |
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>. | |
f5e7ed5d | 73 | </column> |
40358701 | 74 | |
9e6ec60a | 75 | <column name="external_ids" key="hostname"> |
207e85db BS |
76 | The hostname for the host running Open vSwitch. This is a fully |
77 | qualified domain name since version 2.6.2. | |
9e6ec60a RB |
78 | </column> |
79 | ||
34d4f74d RW |
80 | <column name="external_ids" key="rundir"> |
81 | In Open vSwitch 2.8 and later, the run directory of the running Open | |
82 | vSwitch daemon. This directory is used for runtime state such as | |
83 | control and management sockets. The value of <ref | |
84 | column="other_config" key="vhost-sock-dir"/> is relative to this | |
85 | directory. | |
86 | </column> | |
87 | ||
12eb035b AW |
88 | <column name="other_config" key="stats-update-interval" |
89 | type='{"type": "integer", "minInteger": 5000}'> | |
90 | <p> | |
91 | Interval for updating statistics to the database, in milliseconds. | |
92 | This option will affect the update of the <code>statistics</code> | |
93 | column in the following tables: <code>Port</code>, <code>Interface | |
94 | </code>, <code>Mirror</code>. | |
95 | </p> | |
96 | <p> | |
97 | Default value is 5000 ms. | |
98 | </p> | |
99 | <p> | |
100 | Getting statistics more frequently can be achieved via OpenFlow. | |
101 | </p> | |
102 | </column> | |
103 | ||
40358701 GS |
104 | <column name="other_config" key="flow-restore-wait" |
105 | type='{"type": "boolean"}'> | |
106 | <p> | |
107 | When <code>ovs-vswitchd</code> starts up, it has an empty flow table | |
108 | and therefore it handles all arriving packets in its default fashion | |
109 | according to its configuration, by dropping them or sending them to | |
110 | an OpenFlow controller or switching them as a standalone switch. | |
111 | This behavior is ordinarily desirable. However, if | |
112 | <code>ovs-vswitchd</code> is restarting as part of a ``hot-upgrade,'' | |
113 | then this leads to a relatively long period during which packets are | |
114 | mishandled. | |
115 | </p> | |
116 | <p> | |
117 | This option allows for improvement. When <code>ovs-vswitchd</code> | |
118 | starts with this value set as <code>true</code>, it will neither | |
119 | flush or expire previously set datapath flows nor will it send and | |
120 | receive any packets to or from the datapath. When this value is | |
121 | later set to <code>false</code>, <code>ovs-vswitchd</code> will | |
122 | start receiving packets from the datapath and re-setup the flows. | |
123 | </p> | |
124 | <p> | |
125 | Thus, with this option, the procedure for a hot-upgrade of | |
126 | <code>ovs-vswitchd</code> becomes roughly the following: | |
127 | </p> | |
128 | <ol> | |
129 | <li> | |
130 | Stop <code>ovs-vswitchd</code>. | |
131 | </li> | |
132 | <li> | |
133 | Set <ref column="other_config" key="flow-restore-wait"/> | |
134 | to <code>true</code>. | |
135 | </li> | |
136 | <li> | |
137 | Start <code>ovs-vswitchd</code>. | |
138 | </li> | |
139 | <li> | |
140 | Use <code>ovs-ofctl</code> (or some other program, such as an | |
141 | OpenFlow controller) to restore the OpenFlow flow table | |
142 | to the desired state. | |
143 | </li> | |
144 | <li> | |
145 | Set <ref column="other_config" key="flow-restore-wait"/> | |
146 | to <code>false</code> (or remove it entirely from the database). | |
147 | </li> | |
148 | </ol> | |
407b1473 GS |
149 | <p> |
150 | The <code>ovs-ctl</code>'s ``restart'' and ``force-reload-kmod'' | |
151 | functions use the above config option during hot upgrades. | |
152 | </p> | |
40358701 | 153 | </column> |
380f49c4 | 154 | |
e79a6c83 | 155 | <column name="other_config" key="flow-limit" |
380f49c4 EJ |
156 | type='{"type": "integer", "minInteger": 0}'> |
157 | <p> | |
e79a6c83 EJ |
158 | The maximum |
159 | number of flows allowed in the datapath flow table. Internally OVS | |
160 | will choose a flow limit which will likely be lower than this number, | |
38b366b1 JS |
161 | based on real time network conditions. Tweaking this value is |
162 | discouraged unless you know exactly what you're doing. | |
380f49c4 EJ |
163 | </p> |
164 | <p> | |
e79a6c83 | 165 | The default is 200000. |
380f49c4 EJ |
166 | </p> |
167 | </column> | |
7155fa52 | 168 | |
06745dc3 JS |
169 | <column name="other_config" key="max-idle" |
170 | type='{"type": "integer", "minInteger": 500}'> | |
171 | <p> | |
172 | The maximum time (in ms) that idle flows will remain cached in the | |
173 | datapath. Internally OVS will check the validity and activity for | |
174 | datapath flows regularly and may expire flows quicker than this | |
175 | number, based on real time network conditions. Tweaking this | |
176 | value is discouraged unless you know exactly what you're doing. | |
177 | </p> | |
178 | <p> | |
179 | The default is 10000. | |
180 | </p> | |
181 | </column> | |
182 | ||
bab69409 AC |
183 | <column name="other_config" key="dpdk-init" |
184 | type='{"type": "boolean"}'> | |
185 | <p> | |
186 | Set this value to <code>true</code> to enable runtime support for | |
187 | DPDK ports. The vswitch must have compile-time support for DPDK as | |
188 | well. | |
189 | </p> | |
190 | <p> | |
191 | The default value is <code>false</code>. Changing this value requires | |
192 | restarting the daemon | |
193 | </p> | |
194 | <p> | |
195 | If this value is <code>false</code> at startup, any dpdk ports which | |
196 | are configured in the bridge will fail due to memory errors. | |
197 | </p> | |
198 | </column> | |
199 | ||
200 | <column name="other_config" key="dpdk-lcore-mask" | |
201 | type='{"type": "integer", "minInteger": 1}'> | |
202 | <p> | |
203 | Specifies the CPU cores where dpdk lcore threads should be spawned. | |
204 | The DPDK lcore threads are used for DPDK library tasks, such as | |
205 | library internal message processing, logging, etc. Value should be in | |
206 | the form of a hex string (so '0x123') similar to the 'taskset' mask | |
207 | input. | |
208 | </p> | |
209 | <p> | |
210 | The lowest order bit corresponds to the first CPU core. A set bit | |
211 | means the corresponding core is available and an lcore thread will be | |
212 | created and pinned to it. If the input does not cover all cores, | |
213 | those uncovered cores are considered not set. | |
214 | </p> | |
215 | <p> | |
216 | For performance reasons, it is best to set this to a single core on | |
217 | the system, rather than allow lcore threads to float. | |
218 | </p> | |
219 | <p> | |
220 | If not specified, the value will be determined by choosing the lowest | |
221 | CPU core from initial cpu affinity list. Otherwise, the value will be | |
222 | passed directly to the DPDK library. | |
223 | </p> | |
224 | </column> | |
225 | ||
f2eee189 AW |
226 | <column name="other_config" key="pmd-cpu-mask"> |
227 | <p> | |
228 | Specifies CPU mask for setting the cpu affinity of PMD (Poll | |
229 | Mode Driver) threads. Value should be in the form of hex string, | |
230 | similar to the dpdk EAL '-c COREMASK' option input or the 'taskset' | |
231 | mask input. | |
232 | </p> | |
233 | <p> | |
234 | The lowest order bit corresponds to the first CPU core. A set bit | |
ee8627fa AW |
235 | means the corresponding core is available and a pmd thread will be |
236 | created and pinned to it. If the input does not cover all cores, | |
237 | those uncovered cores are considered not set. | |
f2eee189 AW |
238 | </p> |
239 | <p> | |
240 | If not specified, one pmd thread will be created for each numa node | |
241 | and pinned to any available core on the numa node by default. | |
242 | </p> | |
243 | </column> | |
244 | ||
bab69409 AC |
245 | <column name="other_config" key="dpdk-alloc-mem" |
246 | type='{"type": "integer", "minInteger": 0}'> | |
247 | <p> | |
248 | Specifies the amount of memory to preallocate from the hugepage pool, | |
249 | regardless of socket. It is recommended that dpdk-socket-mem is used | |
250 | instead. | |
251 | </p> | |
bab69409 AC |
252 | </column> |
253 | ||
254 | <column name="other_config" key="dpdk-socket-mem" | |
255 | type='{"type": "string"}'> | |
256 | <p> | |
257 | Specifies the amount of memory to preallocate from the hugepage pool, | |
258 | on a per-socket basis. | |
259 | </p> | |
260 | <p> | |
261 | The specifier is a comma-separated string, in ascending order of CPU | |
cd6c5bc8 KT |
262 | socket. E.g. On a four socket system 1024,0,2048 would set socket 0 |
263 | to preallocate 1024MB, socket 1 to preallocate 0MB, socket 2 to | |
264 | preallocate 2048MB and socket 3 (no value given) to preallocate 0MB. | |
bab69409 AC |
265 | </p> |
266 | <p> | |
b838f879 | 267 | If dpdk-socket-mem and dpdk-alloc-mem are not specified, dpdk-socket-mem |
268 | will be used and the default value is 1024,0. If dpdk-socket-mem and | |
269 | dpdk-alloc-mem are specified at same time, dpdk-socket-mem will be | |
270 | used as default. Changing this value requires restarting the daemon. | |
bab69409 AC |
271 | </p> |
272 | </column> | |
273 | ||
274 | <column name="other_config" key="dpdk-hugepage-dir" | |
275 | type='{"type": "string"}'> | |
276 | <p> | |
277 | Specifies the path to the hugetlbfs mount point. | |
278 | </p> | |
279 | <p> | |
280 | If not specified, this will be guessed by the DPDK library (default | |
281 | is /dev/hugepages). Changing this value requires restarting the | |
282 | daemon. | |
283 | </p> | |
284 | </column> | |
285 | ||
eac84432 AC |
286 | <column name="other_config" key="dpdk-extra" |
287 | type='{"type": "string"}'> | |
288 | <p> | |
289 | Specifies additional eal command line arguments for DPDK. | |
290 | </p> | |
291 | <p> | |
292 | The default is empty. Changing this value requires restarting the | |
293 | daemon | |
294 | </p> | |
295 | </column> | |
296 | ||
bab69409 AC |
297 | <column name="other_config" key="vhost-sock-dir" |
298 | type='{"type": "string"}'> | |
299 | <p> | |
34d4f74d RW |
300 | Specifies a relative path from <ref column="external_ids" |
301 | key="rundir"/> to the vhost-user unix domain socket files. If this | |
302 | value is unset, the sockets are put directly in <ref | |
303 | column="external_ids" key="rundir"/>. | |
bab69409 AC |
304 | </p> |
305 | <p> | |
34d4f74d | 306 | Changing this value requires restarting the daemon. |
bab69409 AC |
307 | </p> |
308 | </column> | |
309 | ||
448a4b2f AW |
310 | <column name="other_config" key="n-handler-threads" |
311 | type='{"type": "integer", "minInteger": 1}'> | |
312 | <p> | |
313 | Specifies the number of threads for software datapaths to use for | |
e79a6c83 EJ |
314 | handling new flows. The default the number of online CPU cores minus |
315 | the number of revalidators. | |
316 | </p> | |
317 | <p> | |
318 | This configuration is per datapath. If you have more than one | |
319 | software datapath (e.g. some <code>system</code> bridges and some | |
320 | <code>netdev</code> bridges), then the total number of threads is | |
321 | <code>n-handler-threads</code> times the number of software | |
322 | datapaths. | |
323 | </p> | |
324 | </column> | |
325 | ||
326 | <column name="other_config" key="n-revalidator-threads" | |
327 | type='{"type": "integer", "minInteger": 1}'> | |
328 | <p> | |
329 | Specifies the number of threads for software datapaths to use for | |
330 | revalidating flows in the datapath. Typically, there is a direct | |
331 | correlation between the number of revalidator threads, and the number | |
332 | of flows allowed in the datapath. The default is the number of cpu | |
333 | cores divided by four plus one. If <code>n-handler-threads</code> is | |
334 | set, the default changes to the number of cpu cores minus the number | |
335 | of handler threads. | |
448a4b2f AW |
336 | </p> |
337 | <p> | |
338 | This configuration is per datapath. If you have more than one | |
339 | software datapath (e.g. some <code>system</code> bridges and some | |
340 | <code>netdev</code> bridges), then the total number of threads is | |
341 | <code>n-handler-threads</code> times the number of software | |
342 | datapaths. | |
343 | </p> | |
344 | </column> | |
4c30b246 CL |
345 | |
346 | <column name="other_config" key="emc-insert-inv-prob" | |
347 | type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'> | |
348 | <p> | |
349 | Specifies the inverse probability (1/emc-insert-inv-prob) of a flow | |
350 | being inserted into the Exact Match Cache (EMC). On average one in | |
351 | every <code>emc-insert-inv-prob</code> packets that generate a unique | |
352 | flow will cause an insertion into the EMC. | |
353 | ||
354 | A value of 1 will result in an insertion for every flow (1/1 = 100%) | |
355 | whereas a value of zero will result in no insertions and essentially | |
356 | disable the EMC. | |
357 | </p> | |
358 | <p> | |
359 | Defaults to 100 ie. there is (1/100 =) 1% chance of EMC insertion. | |
360 | </p> | |
361 | </column> | |
f0fb825a EG |
362 | |
363 | <column name="other_config" key="vlan-limit" | |
364 | type='{"type": "integer", "minInteger": 0}'> | |
365 | <p> | |
366 | Limits the number of VLAN headers that can be matched to the | |
367 | specified number. Further VLAN headers will be treated as payload, | |
368 | e.g. a packet with more 802.1q headers will match Ethernet type | |
369 | 0x8100. | |
370 | </p> | |
371 | <p> | |
372 | Value <code>0</code> means unlimited. The actual number of supported | |
373 | VLAN headers is the smallest of <code>vlan-limit</code>, the number | |
374 | of VLANs supported by Open vSwitch userspace (currently 2), and the | |
375 | number supported by the datapath. | |
376 | </p> | |
377 | ||
378 | <p> | |
379 | If this value is absent, the default is currently 1. This maintains | |
380 | backward compatibility with controllers that were designed for use | |
381 | with Open vSwitch versions earlier than 2.8, which only supported one | |
382 | VLAN. | |
383 | </p> | |
384 | </column> | |
89365653 BP |
385 | </group> |
386 | ||
387 | <group title="Status"> | |
388 | <column name="next_cfg"> | |
389 | Sequence number for client to increment. When a client modifies | |
390 | any part of the database configuration and wishes to wait for | |
391 | Open vSwitch to finish applying the changes, it may increment | |
392 | this sequence number. | |
393 | </column> | |
394 | ||
395 | <column name="cur_cfg"> | |
396 | Sequence number that Open vSwitch sets to the current value of | |
2e57b537 | 397 | <ref column="next_cfg"/> after it finishes applying a set of |
89365653 BP |
398 | configuration changes. |
399 | </column> | |
c1c9c9c4 | 400 | |
3fd8d445 | 401 | <group title="Statistics"> |
018f1525 | 402 | <p> |
3fd8d445 BP |
403 | The <code>statistics</code> column contains key-value pairs that |
404 | report statistics about a system running an Open vSwitch. These are | |
405 | updated periodically (currently, every 5 seconds). Key-value pairs | |
406 | that cannot be determined or that do not apply to a platform are | |
407 | omitted. | |
ce887677 BP |
408 | </p> |
409 | ||
f9e5e5b3 BP |
410 | <column name="other_config" key="enable-statistics" |
411 | type='{"type": "boolean"}'> | |
3fd8d445 BP |
412 | Statistics are disabled by default to avoid overhead in the common |
413 | case when statistics gathering is not useful. Set this value to | |
414 | <code>true</code> to enable populating the <ref column="statistics"/> | |
415 | column or to <code>false</code> to explicitly disable it. | |
416 | </column> | |
3fe80505 | 417 | |
f9e5e5b3 BP |
418 | <column name="statistics" key="cpu" |
419 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
420 | <p> |
421 | Number of CPU processors, threads, or cores currently online and | |
422 | available to the operating system on which Open vSwitch is running, | |
423 | as an integer. This may be less than the number installed, if some | |
424 | are not online or if they are not available to the operating | |
425 | system. | |
426 | </p> | |
427 | <p> | |
428 | Open vSwitch userspace processes are not multithreaded, but the | |
429 | Linux kernel-based datapath is. | |
430 | </p> | |
431 | </column> | |
ce887677 | 432 | |
3fd8d445 BP |
433 | <column name="statistics" key="load_average"> |
434 | A comma-separated list of three floating-point numbers, | |
435 | representing the system load average over the last 1, 5, and 15 | |
436 | minutes, respectively. | |
437 | </column> | |
ce887677 | 438 | |
3fd8d445 BP |
439 | <column name="statistics" key="memory"> |
440 | <p> | |
441 | A comma-separated list of integers, each of which represents a | |
442 | quantity of memory in kilobytes that describes the operating | |
443 | system on which Open vSwitch is running. In respective order, | |
444 | these values are: | |
445 | </p> | |
ce887677 | 446 | |
3fd8d445 BP |
447 | <ol> |
448 | <li>Total amount of RAM allocated to the OS.</li> | |
449 | <li>RAM allocated to the OS that is in use.</li> | |
450 | <li>RAM that can be flushed out to disk or otherwise discarded | |
451 | if that space is needed for another purpose. This number is | |
452 | necessarily less than or equal to the previous value.</li> | |
453 | <li>Total disk space allocated for swap.</li> | |
454 | <li>Swap space currently in use.</li> | |
455 | </ol> | |
ce887677 | 456 | |
3fd8d445 BP |
457 | <p> |
458 | On Linux, all five values can be determined and are included. On | |
459 | other operating systems, only the first two values can be | |
460 | determined, so the list will only have two values. | |
461 | </p> | |
462 | </column> | |
ce887677 | 463 | |
3fd8d445 BP |
464 | <column name="statistics" key="process_NAME"> |
465 | <p> | |
466 | One such key-value pair, with <code>NAME</code> replaced by | |
467 | a process name, will exist for each running Open vSwitch | |
468 | daemon process, with <var>name</var> replaced by the | |
469 | daemon's name (e.g. <code>process_ovs-vswitchd</code>). The | |
470 | value is a comma-separated list of integers. The integers | |
471 | represent the following, with memory measured in kilobytes | |
472 | and durations in milliseconds: | |
473 | </p> | |
ce887677 | 474 | |
3fd8d445 BP |
475 | <ol> |
476 | <li>The process's virtual memory size.</li> | |
477 | <li>The process's resident set size.</li> | |
478 | <li>The amount of user and system CPU time consumed by the | |
479 | process.</li> | |
480 | <li>The number of times that the process has crashed and been | |
481 | automatically restarted by the monitor.</li> | |
482 | <li>The duration since the process was started.</li> | |
483 | <li>The duration for which the process has been running.</li> | |
484 | </ol> | |
ce887677 | 485 | |
3fd8d445 BP |
486 | <p> |
487 | The interpretation of some of these values depends on whether the | |
488 | process was started with the <option>--monitor</option>. If it | |
489 | was not, then the crash count will always be 0 and the two | |
490 | durations will always be the same. If <option>--monitor</option> | |
491 | was given, then the crash count may be positive; if it is, the | |
492 | latter duration is the amount of time since the most recent crash | |
493 | and restart. | |
494 | </p> | |
ce887677 | 495 | |
3fd8d445 BP |
496 | <p> |
497 | There will be one key-value pair for each file in Open vSwitch's | |
498 | ``run directory'' (usually <code>/var/run/openvswitch</code>) | |
499 | whose name ends in <code>.pid</code>, whose contents are a | |
500 | process ID, and which is locked by a running process. The | |
501 | <var>name</var> is taken from the pidfile's name. | |
502 | </p> | |
ce887677 | 503 | |
3fd8d445 BP |
504 | <p> |
505 | Currently Open vSwitch is only able to obtain all of the above | |
506 | detail on Linux systems. On other systems, the same key-value | |
507 | pairs will be present but the values will always be the empty | |
508 | string. | |
509 | </p> | |
510 | </column> | |
ce887677 | 511 | |
3fd8d445 BP |
512 | <column name="statistics" key="file_systems"> |
513 | <p> | |
514 | A space-separated list of information on local, writable file | |
515 | systems. Each item in the list describes one file system and | |
516 | consists in turn of a comma-separated list of the following: | |
517 | </p> | |
ce887677 | 518 | |
3fd8d445 BP |
519 | <ol> |
520 | <li>Mount point, e.g. <code>/</code> or <code>/var/log</code>. | |
521 | Any spaces or commas in the mount point are replaced by | |
522 | underscores.</li> | |
523 | <li>Total size, in kilobytes, as an integer.</li> | |
524 | <li>Amount of storage in use, in kilobytes, as an integer.</li> | |
525 | </ol> | |
ce887677 | 526 | |
3fd8d445 BP |
527 | <p> |
528 | This key-value pair is omitted if there are no local, writable | |
529 | file systems or if Open vSwitch cannot obtain the needed | |
530 | information. | |
531 | </p> | |
532 | </column> | |
533 | </group> | |
89365653 | 534 | </group> |
94db5407 | 535 | |
538c6dfa BP |
536 | <group title="Version Reporting"> |
537 | <p> | |
6b4186af EJ |
538 | These columns report the types and versions of the hardware and |
539 | software running Open vSwitch. We recommend in general that software | |
540 | should test whether specific features are supported instead of relying | |
541 | on version number checks. These values are primarily intended for | |
542 | reporting to human administrators. | |
538c6dfa BP |
543 | </p> |
544 | ||
545 | <column name="ovs_version"> | |
d4da3acc | 546 | The Open vSwitch version number, e.g. <code>1.1.0</code>. |
538c6dfa BP |
547 | </column> |
548 | ||
8159b984 | 549 | <column name="db_version"> |
6b4186af | 550 | <p> |
51738fe1 BP |
551 | The database schema version number, e.g. <code>1.2.3</code>. See |
552 | ovsdb-tool(1) for an explanation of the numbering scheme. | |
8159b984 BP |
553 | </p> |
554 | ||
555 | <p> | |
556 | The schema version is part of the database schema, so it can also be | |
557 | retrieved by fetching the schema using the Open vSwitch database | |
558 | protocol. | |
559 | </p> | |
560 | </column> | |
561 | ||
538c6dfa BP |
562 | <column name="system_type"> |
563 | <p> | |
6b4186af EJ |
564 | An identifier for the type of system on top of which Open vSwitch |
565 | runs, e.g. <code>XenServer</code> or <code>KVM</code>. | |
566 | </p> | |
567 | <p> | |
568 | System integrators are responsible for choosing and setting an | |
569 | appropriate value for this column. | |
570 | </p> | |
538c6dfa BP |
571 | </column> |
572 | ||
573 | <column name="system_version"> | |
574 | <p> | |
6b4186af | 575 | The version of the system identified by <ref column="system_type"/>, |
404c1692 | 576 | e.g. <code>5.6.100-39265p</code> on XenServer 5.6.100 build 39265. |
6b4186af EJ |
577 | </p> |
578 | <p> | |
579 | System integrators are responsible for choosing and setting an | |
580 | appropriate value for this column. | |
581 | </p> | |
538c6dfa | 582 | </column> |
6b4186af | 583 | |
538c6dfa BP |
584 | </group> |
585 | ||
842733c3 MG |
586 | <group title="Capabilities"> |
587 | <p> | |
588 | These columns report capabilities of the Open vSwitch instance. | |
589 | </p> | |
590 | <column name="datapath_types"> | |
591 | <p> | |
592 | This column reports the different dpifs registered with the system. | |
593 | These are the values that this instance supports in the <ref | |
594 | column="datapath_type" table="Bridge"/> column of the <ref | |
595 | table="Bridge"/> table. | |
596 | </p> | |
597 | </column> | |
598 | <column name="iface_types"> | |
599 | <p> | |
600 | This column reports the different netdevs registered with the system. | |
601 | These are the values that this instance supports in the <ref | |
602 | column="type" table="Interface"/> column of the <ref | |
603 | table="Interface"/> table. | |
604 | </p> | |
605 | </column> | |
606 | </group> | |
607 | ||
94db5407 BP |
608 | <group title="Database Configuration"> |
609 | <p> | |
610 | These columns primarily configure the Open vSwitch database | |
611 | (<code>ovsdb-server</code>), not the Open vSwitch switch | |
612 | (<code>ovs-vswitchd</code>). The OVSDB database also uses the <ref | |
613 | column="ssl"/> settings. | |
614 | </p> | |
615 | ||
616 | <p> | |
617 | The Open vSwitch switch does read the database configuration to | |
618 | determine remote IP addresses to which in-band control should apply. | |
619 | </p> | |
620 | ||
621 | <column name="manager_options"> | |
622 | Database clients to which the Open vSwitch database server should | |
623 | connect or to which it should listen, along with options for how these | |
624 | connection should be configured. See the <ref table="Manager"/> table | |
625 | for more information. | |
626 | </column> | |
94db5407 | 627 | </group> |
3fd8d445 BP |
628 | |
629 | <group title="Common Columns"> | |
630 | The overall purpose of these columns is described under <code>Common | |
631 | Columns</code> at the beginning of this document. | |
632 | ||
633 | <column name="other_config"/> | |
634 | <column name="external_ids"/> | |
635 | </group> | |
89365653 BP |
636 | </table> |
637 | ||
638 | <table name="Bridge"> | |
639 | <p> | |
640 | Configuration for a bridge within an | |
641 | <ref table="Open_vSwitch"/>. | |
642 | </p> | |
643 | <p> | |
644 | A <ref table="Bridge"/> record represents an Ethernet switch with one or | |
645 | more ``ports,'' which are the <ref table="Port"/> records pointed to by | |
646 | the <ref table="Bridge"/>'s <ref column="ports"/> column. | |
647 | </p> | |
648 | ||
649 | <group title="Core Features"> | |
650 | <column name="name"> | |
3cbe33df | 651 | <p> |
c885f934 BP |
652 | Bridge identifier. Must be unique among the names of ports, |
653 | interfaces, and bridges on a host. | |
3cbe33df BP |
654 | </p> |
655 | ||
656 | <p> | |
c885f934 BP |
657 | The name must be alphanumeric and must not contain forward or |
658 | backward slashes. The name of a bridge is also the name of an <ref | |
659 | table="Interface"/> (and a <ref table="Port"/>) within the bridge, so | |
660 | the restrictions on the <ref table="Interface" column="name"/> column | |
661 | in the <ref table="Interface"/> table, particularly on length, also | |
662 | apply to bridge names. Refer to the documentation for <ref | |
663 | table="Interface"/> names for details. | |
3cbe33df | 664 | </p> |
89365653 BP |
665 | </column> |
666 | ||
667 | <column name="ports"> | |
668 | Ports included in the bridge. | |
669 | </column> | |
670 | ||
671 | <column name="mirrors"> | |
672 | Port mirroring configuration. | |
673 | </column> | |
674 | ||
675 | <column name="netflow"> | |
676 | NetFlow configuration. | |
677 | </column> | |
678 | ||
679 | <column name="sflow"> | |
29089a54 RL |
680 | sFlow(R) configuration. |
681 | </column> | |
682 | ||
683 | <column name="ipfix"> | |
684 | IPFIX configuration. | |
89365653 BP |
685 | </column> |
686 | ||
687 | <column name="flood_vlans"> | |
6c2d2a9f BP |
688 | <p> |
689 | VLAN IDs of VLANs on which MAC address learning should be disabled, | |
690 | so that packets are flooded instead of being sent to specific ports | |
691 | that are believed to contain packets' destination MACs. This should | |
692 | ordinarily be used to disable MAC learning on VLANs used for | |
693 | mirroring (RSPAN VLANs). It may also be useful for debugging. | |
694 | </p> | |
695 | <p> | |
696 | SLB bonding (see the <ref table="Port" column="bond_mode"/> column in | |
697 | the <ref table="Port"/> table) is incompatible with | |
698 | <code>flood_vlans</code>. Consider using another bonding mode or | |
699 | a different type of mirror instead. | |
700 | </p> | |
89365653 | 701 | </column> |
99eef98b DF |
702 | |
703 | <column name="auto_attach"> | |
039a8ccd | 704 | Auto Attach configuration. |
99eef98b | 705 | </column> |
89365653 BP |
706 | </group> |
707 | ||
708 | <group title="OpenFlow Configuration"> | |
709 | <column name="controller"> | |
88f69f88 BP |
710 | <p> |
711 | OpenFlow controller set. If unset, then no OpenFlow controllers | |
712 | will be used. | |
713 | </p> | |
714 | ||
715 | <p> | |
716 | If there are primary controllers, removing all of them clears the | |
717 | flow table. If there are no primary controllers, adding one also | |
718 | clears the flow table. Other changes to the set of controllers, such | |
719 | as adding or removing a service controller, adding another primary | |
720 | controller to supplement an existing primary controller, or removing | |
721 | only one of two primary controllers, have no effect on the flow | |
722 | table. | |
723 | </p> | |
89365653 BP |
724 | </column> |
725 | ||
254750ce BP |
726 | <column name="flow_tables"> |
727 | Configuration for OpenFlow tables. Each pair maps from an OpenFlow | |
728 | table ID to configuration for that table. | |
729 | </column> | |
730 | ||
31681a5d JP |
731 | <column name="fail_mode"> |
732 | <p>When a controller is configured, it is, ordinarily, responsible | |
3fd8d445 BP |
733 | for setting up all flows on the switch. Thus, if the connection to |
734 | the controller fails, no new network connections can be set up. | |
735 | If the connection to the controller stays down long enough, | |
736 | no packets can pass through the switch at all. This setting | |
737 | determines the switch's response to such a situation. It may be set | |
738 | to one of the following: | |
739 | <dl> | |
740 | <dt><code>standalone</code></dt> | |
741 | <dd>If no message is received from the controller for three | |
742 | times the inactivity probe interval | |
743 | (see <ref column="inactivity_probe"/>), then Open vSwitch | |
744 | will take over responsibility for setting up flows. In | |
745 | this mode, Open vSwitch causes the bridge to act like an | |
746 | ordinary MAC-learning switch. Open vSwitch will continue | |
747 | to retry connecting to the controller in the background | |
748 | and, when the connection succeeds, it will discontinue its | |
749 | standalone behavior.</dd> | |
750 | <dt><code>secure</code></dt> | |
751 | <dd>Open vSwitch will not set up flows on its own when the | |
752 | controller connection fails or when no controllers are | |
753 | defined. The bridge will continue to retry connecting to | |
754 | any defined controllers forever.</dd> | |
755 | </dl> | |
31681a5d | 756 | </p> |
aed2db18 BP |
757 | <p> |
758 | The default is <code>standalone</code> if the value is unset, but | |
759 | future versions of Open vSwitch may change the default. | |
760 | </p> | |
7dea6ace BP |
761 | <p> |
762 | The <code>standalone</code> mode can create forwarding loops on a | |
763 | bridge that has more than one uplink port unless STP is enabled. To | |
764 | avoid loops on such a bridge, configure <code>secure</code> mode or | |
765 | enable STP (see <ref column="stp_enable"/>). | |
766 | </p> | |
299a244b | 767 | <p>When more than one controller is configured, |
3fd8d445 BP |
768 | <ref column="fail_mode"/> is considered only when none of the |
769 | configured controllers can be contacted.</p> | |
88f69f88 BP |
770 | <p> |
771 | Changing <ref column="fail_mode"/> when no primary controllers are | |
772 | configured clears the flow table. | |
773 | </p> | |
31681a5d JP |
774 | </column> |
775 | ||
89365653 | 776 | <column name="datapath_id"> |
8de67146 BP |
777 | Reports the OpenFlow datapath ID in use. Exactly 16 hex digits. |
778 | (Setting this column has no useful effect. Set <ref | |
779 | column="other-config" key="datapath-id"/> instead.) | |
89365653 | 780 | </column> |
3fd8d445 | 781 | |
b5cbbcf6 AZ |
782 | <column name="datapath_version"> |
783 | <p> | |
784 | Reports the version number of the Open vSwitch datapath in use. | |
785 | This allows management software to detect and report discrepancies | |
786 | between Open vSwitch userspace and datapath versions. (The <ref | |
787 | column="ovs_version" table="Open_vSwitch"/> column in the <ref | |
788 | table="Open_vSwitch"/> reports the Open vSwitch userspace version.) | |
789 | The version reported depends on the datapath in use: | |
790 | </p> | |
791 | ||
792 | <ul> | |
793 | <li> | |
794 | When the kernel module included in the Open vSwitch source tree is | |
795 | used, this column reports the Open vSwitch version from which the | |
796 | module was taken. | |
797 | </li> | |
798 | ||
799 | <li> | |
800 | When the kernel module that is part of the upstream Linux kernel is | |
801 | used, this column reports <code><unknown></code>. | |
802 | </li> | |
803 | ||
804 | <li> | |
805 | When the datapath is built into the <code>ovs-vswitchd</code> | |
806 | binary, this column reports <code><built-in></code>. A | |
807 | built-in datapath is by definition the same version as the rest of | |
808 | the Open VSwitch userspace. | |
809 | </li> | |
810 | ||
811 | <li> | |
812 | Other datapaths (such as the Hyper-V kernel datapath) currently | |
813 | report <code><unknown></code>. | |
814 | </li> | |
815 | </ul> | |
816 | ||
817 | <p> | |
818 | A version discrepancy between <code>ovs-vswitchd</code> and the | |
819 | datapath in use is not normally cause for alarm. The Open vSwitch | |
820 | kernel datapaths for Linux and Hyper-V, in particular, are designed | |
821 | for maximum inter-version compatibility: any userspace version works | |
822 | with with any kernel version. Some reasons do exist to insist on | |
823 | particular user/kernel pairings. First, newer kernel versions add | |
824 | new features, that can only be used by new-enough userspace, e.g. | |
825 | VXLAN tunneling requires certain minimal userspace and kernel | |
826 | versions. Second, as an extension to the first reason, some newer | |
827 | kernel versions add new features for enhancing performance that only | |
828 | new-enough userspace versions can take advantage of. | |
829 | </p> | |
830 | </column> | |
831 | ||
3fd8d445 BP |
832 | <column name="other_config" key="datapath-id"> |
833 | Exactly 16 hex digits to set the OpenFlow datapath ID to a specific | |
834 | value. May not be all-zero. | |
835 | </column> | |
836 | ||
8b6ff729 BP |
837 | <column name="other_config" key="dp-desc"> |
838 | Human readable description of datapath. It it a maximum 256 | |
839 | byte-long free-form string to describe the datapath for | |
840 | debugging purposes, e.g. <code>switch3 in room 3120</code>. | |
841 | </column> | |
842 | ||
f9e5e5b3 BP |
843 | <column name="other_config" key="disable-in-band" |
844 | type='{"type": "boolean"}'> | |
3fd8d445 BP |
845 | If set to <code>true</code>, disable in-band control on the bridge |
846 | regardless of controller and manager settings. | |
847 | </column> | |
848 | ||
f9e5e5b3 BP |
849 | <column name="other_config" key="in-band-queue" |
850 | type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'> | |
3fd8d445 BP |
851 | A queue ID as a nonnegative integer. This sets the OpenFlow queue ID |
852 | that will be used by flows set up by in-band control on this bridge. | |
853 | If unset, or if the port used by an in-band control flow does not have | |
854 | QoS configured, or if the port does not have a queue with the specified | |
855 | ID, the default queue is used instead. | |
856 | </column> | |
7beaa082 SH |
857 | |
858 | <column name="protocols"> | |
039a8ccd BP |
859 | <p> |
860 | List of OpenFlow protocols that may be used when negotiating | |
8d348579 BP |
861 | a connection with a controller. OpenFlow 1.0, 1.1, 1.2, 1.3, and |
862 | 1.4 are enabled by default if this column is empty. | |
039a8ccd | 863 | </p> |
ecb229be | 864 | |
039a8ccd | 865 | <p> |
8d348579 | 866 | OpenFlow 1.5 and 1.6 are not enabled by default because their |
b79d45a1 BP |
867 | implementations are missing features. In addition, the OpenFlow 1.6 |
868 | specification is still under development and thus subject to change. | |
42dccab5 | 869 | </p> |
7beaa082 | 870 | </column> |
89365653 BP |
871 | </group> |
872 | ||
21f7563c | 873 | <group title="Spanning Tree Configuration"> |
01f13d4f BP |
874 | <p> |
875 | The IEEE 802.1D Spanning Tree Protocol (STP) is a network protocol | |
876 | that ensures loop-free topologies. It allows redundant links to | |
877 | be included in the network to provide automatic backup paths if | |
878 | the active links fails. | |
879 | </p> | |
9cc6bf75 | 880 | |
01f13d4f BP |
881 | <p> |
882 | These settings configure the slower-to-converge but still widely | |
883 | supported version of Spanning Tree Protocol, sometimes known as | |
884 | 802.1D-1998. Open vSwitch also supports the newer Rapid Spanning Tree | |
885 | Protocol (RSTP), documented later in the section titled <code>Rapid | |
886 | Spanning Tree Configuration</code>. | |
887 | </p> | |
21f7563c | 888 | |
01f13d4f BP |
889 | <group title="STP Configuration"> |
890 | <column name="stp_enable" type='{"type": "boolean"}'> | |
891 | <p> | |
892 | Enable spanning tree on the bridge. By default, STP is disabled | |
893 | on bridges. Bond, internal, and mirror ports are not supported | |
894 | and will not participate in the spanning tree. | |
895 | </p> | |
21f7563c | 896 | |
01f13d4f BP |
897 | <p> |
898 | STP and RSTP are mutually exclusive. If both are enabled, RSTP | |
899 | will be used. | |
900 | </p> | |
901 | </column> | |
21f7563c | 902 | |
01f13d4f BP |
903 | <column name="other_config" key="stp-system-id"> |
904 | The bridge's STP identifier (the lower 48 bits of the bridge-id) | |
905 | in the form | |
906 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>. | |
907 | By default, the identifier is the MAC address of the bridge. | |
908 | </column> | |
21f7563c | 909 | |
01f13d4f BP |
910 | <column name="other_config" key="stp-priority" |
911 | type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'> | |
912 | The bridge's relative priority value for determining the root | |
913 | bridge (the upper 16 bits of the bridge-id). A bridge with the | |
914 | lowest bridge-id is elected the root. By default, the priority | |
915 | is 0x8000. | |
916 | </column> | |
dc2b70ba | 917 | |
01f13d4f BP |
918 | <column name="other_config" key="stp-hello-time" |
919 | type='{"type": "integer", "minInteger": 1, "maxInteger": 10}'> | |
920 | The interval between transmissions of hello messages by | |
921 | designated ports, in seconds. By default the hello interval is | |
922 | 2 seconds. | |
923 | </column> | |
dc2b70ba | 924 | |
01f13d4f BP |
925 | <column name="other_config" key="stp-max-age" |
926 | type='{"type": "integer", "minInteger": 6, "maxInteger": 40}'> | |
927 | The maximum age of the information transmitted by the bridge | |
928 | when it is the root bridge, in seconds. By default, the maximum | |
929 | age is 20 seconds. | |
930 | </column> | |
931 | ||
932 | <column name="other_config" key="stp-forward-delay" | |
933 | type='{"type": "integer", "minInteger": 4, "maxInteger": 30}'> | |
934 | The delay to wait between transitioning root and designated | |
935 | ports to <code>forwarding</code>, in seconds. By default, the | |
936 | forwarding delay is 15 seconds. | |
937 | </column> | |
938 | ||
939 | <column name="other_config" key="mcast-snooping-aging-time" | |
940 | type='{"type": "integer", "minInteger": 1}'> | |
941 | <p> | |
942 | The maximum number of seconds to retain a multicast snooping entry for | |
943 | which no packets have been seen. The default is currently 300 | |
944 | seconds (5 minutes). The value, if specified, is forced into a | |
945 | reasonable range, currently 15 to 3600 seconds. | |
946 | </p> | |
947 | </column> | |
948 | ||
949 | <column name="other_config" key="mcast-snooping-table-size" | |
950 | type='{"type": "integer", "minInteger": 1}'> | |
951 | <p> | |
952 | The maximum number of multicast snooping addresses to learn. The | |
953 | default is currently 2048. The value, if specified, is forced into | |
954 | a reasonable range, currently 10 to 1,000,000. | |
955 | </p> | |
956 | </column> | |
957 | <column name="other_config" key="mcast-snooping-disable-flood-unregistered" | |
958 | type='{"type": "boolean"}'> | |
959 | <p> | |
960 | If set to <code>false</code>, unregistered multicast packets are forwarded | |
961 | to all ports. | |
962 | If set to <code>true</code>, unregistered multicast packets are forwarded | |
963 | to ports connected to multicast routers. | |
964 | </p> | |
965 | </column> | |
966 | </group> | |
967 | ||
968 | <group title="STP Status"> | |
dc2b70ba | 969 | <p> |
01f13d4f BP |
970 | These key-value pairs report the status of 802.1D-1998. They are |
971 | present only if STP is enabled (via the <ref column="stp_enable"/> | |
972 | column). | |
dc2b70ba | 973 | </p> |
01f13d4f BP |
974 | <column name="status" key="stp_bridge_id"> |
975 | The bridge ID used in spanning tree advertisements, in the form | |
976 | <var>xxxx</var>.<var>yyyyyyyyyyyy</var> where the <var>x</var>s are | |
977 | the STP priority, the <var>y</var>s are the STP system ID, and each | |
978 | <var>x</var> and <var>y</var> is a hex digit. | |
979 | </column> | |
980 | <column name="status" key="stp_designated_root"> | |
981 | The designated root for this spanning tree, in the same form as <ref | |
982 | column="status" key="stp_bridge_id"/>. If this bridge is the root, | |
983 | this will have the same value as <ref column="status" | |
984 | key="stp_bridge_id"/>, otherwise it will differ. | |
985 | </column> | |
986 | <column name="status" key="stp_root_path_cost"> | |
987 | The path cost of reaching the designated bridge. A lower number is | |
988 | better. The value is 0 if this bridge is the root, otherwise it is | |
989 | higher. | |
990 | </column> | |
991 | </group> | |
992 | </group> | |
993 | ||
994 | <group title="Rapid Spanning Tree"> | |
995 | <p> | |
996 | Rapid Spanning Tree Protocol (RSTP), like STP, is a network protocol | |
997 | that ensures loop-free topologies. RSTP superseded STP with the | |
998 | publication of 802.1D-2004. Compared to STP, RSTP converges more | |
999 | quickly and recovers more quickly from failures. | |
1000 | </p> | |
1001 | ||
1002 | <group title="RSTP Configuration"> | |
1003 | <column name="rstp_enable" type='{"type": "boolean"}'> | |
1004 | <p> | |
1005 | Enable Rapid Spanning Tree on the bridge. By default, RSTP is disabled | |
1006 | on bridges. Bond, internal, and mirror ports are not supported | |
1007 | and will not participate in the spanning tree. | |
1008 | </p> | |
1009 | ||
1010 | <p> | |
1011 | STP and RSTP are mutually exclusive. If both are enabled, RSTP | |
1012 | will be used. | |
1013 | </p> | |
1014 | </column> | |
1015 | ||
1016 | <column name="other_config" key="rstp-address"> | |
1017 | The bridge's RSTP address (the lower 48 bits of the bridge-id) | |
1018 | in the form | |
1019 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>. | |
1020 | By default, the address is the MAC address of the bridge. | |
1021 | </column> | |
1022 | ||
1023 | <column name="other_config" key="rstp-priority" | |
1024 | type='{"type": "integer", "minInteger": 0, "maxInteger": 61440}'> | |
1025 | The bridge's relative priority value for determining the root | |
1026 | bridge (the upper 16 bits of the bridge-id). A bridge with the | |
1027 | lowest bridge-id is elected the root. By default, the priority | |
1028 | is 0x8000 (32768). This value needs to be a multiple of 4096, | |
1029 | otherwise it's rounded to the nearest inferior one. | |
1030 | </column> | |
1031 | ||
1032 | <column name="other_config" key="rstp-ageing-time" | |
1033 | type='{"type": "integer", "minInteger": 10, "maxInteger": 1000000}'> | |
1034 | The Ageing Time parameter for the Bridge. The default value | |
1035 | is 300 seconds. | |
1036 | </column> | |
1037 | ||
1038 | <column name="other_config" key="rstp-force-protocol-version" | |
1039 | type='{"type": "integer"}'> | |
1040 | The Force Protocol Version parameter for the Bridge. This | |
1041 | can take the value 0 (STP Compatibility mode) or 2 | |
1042 | (the default, normal operation). | |
1043 | </column> | |
1044 | ||
1045 | <column name="other_config" key="rstp-max-age" | |
1046 | type='{"type": "integer", "minInteger": 6, "maxInteger": 40}'> | |
1047 | The maximum age of the information transmitted by the Bridge | |
1048 | when it is the Root Bridge. The default value is 20. | |
1049 | </column> | |
1050 | ||
1051 | <column name="other_config" key="rstp-forward-delay" | |
1052 | type='{"type": "integer", "minInteger": 4, "maxInteger": 30}'> | |
1053 | The delay used by STP Bridges to transition Root and Designated | |
1054 | Ports to Forwarding. The default value is 15. | |
1055 | </column> | |
1056 | ||
1057 | <column name="other_config" key="rstp-transmit-hold-count" | |
1058 | type='{"type": "integer", "minInteger": 1, "maxInteger": 10}'> | |
1059 | The Transmit Hold Count used by the Port Transmit state machine | |
1060 | to limit transmission rate. The default value is 6. | |
1061 | </column> | |
1062 | </group> | |
1063 | ||
1064 | <group title="RSTP Status"> | |
dc2b70ba | 1065 | <p> |
01f13d4f BP |
1066 | These key-value pairs report the status of 802.1D-2004. They are |
1067 | present only if RSTP is enabled (via the <ref column="rstp_enable"/> | |
1068 | column). | |
dc2b70ba | 1069 | </p> |
01f13d4f BP |
1070 | <column name="rstp_status" key="rstp_bridge_id"> |
1071 | The bridge ID used in rapid spanning tree advertisements, in the form | |
1072 | <var>x</var>.<var>yyy</var>.<var>zzzzzzzzzzzz</var> where | |
1073 | <var>x</var> is the RSTP priority, the <var>y</var>s are a locally | |
1074 | assigned system ID extension, the <var>z</var>s are the STP system | |
1075 | ID, and each <var>x</var>, <var>y</var>, or <var>z</var> is a hex | |
1076 | digit. | |
1077 | </column> | |
1078 | <column name="rstp_status" key="rstp_root_id"> | |
1079 | The root of this spanning tree, in the same form as <ref | |
1080 | column="rstp_status" key="rstp_bridge_id"/>. If this bridge is the | |
1081 | root, this will have the same value as <ref column="rstp_status" | |
1082 | key="rstp_bridge_id"/>, otherwise it will differ. | |
1083 | </column> | |
1084 | <column name="rstp_status" key="rstp_root_path_cost" | |
1085 | type='{"type": "integer", "minInteger": 0}'> | |
1086 | The path cost of reaching the root. A lower number is better. The | |
1087 | value is 0 if this bridge is the root, otherwise it is higher. | |
1088 | </column> | |
1089 | <column name="rstp_status" key="rstp_designated_id"> | |
1090 | The RSTP designated ID, in the same form as <ref column="rstp_status" | |
1091 | key="rstp_bridge_id"/>. | |
1092 | </column> | |
1093 | <column name="rstp_status" key="rstp_designated_port_id"> | |
1094 | The RSTP designated port ID, as a 4-digit hex number. | |
1095 | </column> | |
1096 | <column name="rstp_status" key="rstp_bridge_port_id"> | |
1097 | The RSTP bridge port ID, as a 4-digit hex number. | |
1098 | </column> | |
1099 | </group> | |
dc2b70ba FL |
1100 | </group> |
1101 | ||
1102 | <group title="Multicast Snooping Configuration"> | |
1103 | Multicast snooping (RFC 4541) monitors the Internet Group Management | |
06994f87 TLSC |
1104 | Protocol (IGMP) and Multicast Listener Discovery traffic between hosts |
1105 | and multicast routers. The switch uses what IGMP and MLD snooping | |
1106 | learns to forward multicast traffic only to interfaces that are connected | |
1107 | to interested receivers. Currently it supports IGMPv1, IGMPv2, IGMPv3, | |
1108 | MLDv1 and MLDv2 protocols. | |
dc2b70ba FL |
1109 | |
1110 | <column name="mcast_snooping_enable"> | |
1111 | Enable multicast snooping on the bridge. For now, the default | |
1112 | is disabled. | |
1113 | </column> | |
21f7563c JP |
1114 | </group> |
1115 | ||
89365653 BP |
1116 | <group title="Other Features"> |
1117 | <column name="datapath_type"> | |
842733c3 MG |
1118 | Name of datapath provider. The kernel datapath has type |
1119 | <code>system</code>. The userspace datapath has type | |
1120 | <code>netdev</code>. A manager may refer to the <ref | |
1121 | table="Open_vSwitch" column="datapath_types"/> column of the <ref | |
1122 | table="Open_vSwitch"/> table for a list of the types accepted by this | |
1123 | Open vSwitch instance. | |
89365653 BP |
1124 | </column> |
1125 | ||
3fd8d445 BP |
1126 | <column name="external_ids" key="bridge-id"> |
1127 | A unique identifier of the bridge. On Citrix XenServer this will | |
1128 | commonly be the same as | |
1129 | <ref column="external_ids" key="xs-network-uuids"/>. | |
89365653 BP |
1130 | </column> |
1131 | ||
3fd8d445 BP |
1132 | <column name="external_ids" key="xs-network-uuids"> |
1133 | Semicolon-delimited set of universally unique identifier(s) for the | |
1134 | network with which this bridge is associated on a Citrix XenServer | |
1135 | host. The network identifiers are RFC 4122 UUIDs as displayed by, | |
1136 | e.g., <code>xe network-list</code>. | |
1137 | </column> | |
1138 | ||
1139 | <column name="other_config" key="hwaddr"> | |
1140 | An Ethernet address in the form | |
1141 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var> | |
1142 | to set the hardware address of the local port and influence the | |
1143 | datapath ID. | |
1144 | </column> | |
1145 | ||
f9e5e5b3 BP |
1146 | <column name="other_config" key="forward-bpdu" |
1147 | type='{"type": "boolean"}'> | |
da1e25d5 | 1148 | |
039a8ccd | 1149 | <p> |
da1e25d5 BP |
1150 | Controls forwarding of BPDUs and other network control frames when |
1151 | NORMAL action is invoked. When this option is <code>false</code> or | |
1152 | unset, frames with reserved Ethernet addresses (see table below) will | |
1153 | not be forwarded. When this option is <code>true</code>, such frames | |
1154 | will not be treated specially. | |
039a8ccd BP |
1155 | </p> |
1156 | ||
1157 | <p> | |
1158 | The above general rule has the following exceptions: | |
1159 | </p> | |
1160 | ||
1161 | <ul> | |
1162 | <li> | |
1163 | If STP is enabled on the bridge (see the <ref column="stp_enable" | |
1164 | table="Bridge"/> column in the <ref table="Bridge"/> table), the | |
1165 | bridge processes all received STP packets and never passes them to | |
1166 | OpenFlow or forwards them. This is true even if STP is disabled on | |
1167 | an individual port. | |
1168 | </li> | |
1169 | ||
1170 | <li> | |
1171 | If LLDP is enabled on an interface (see the <ref column="lldp" | |
1172 | table="Interface"/> column in the <ref table="Interface"/> table), | |
1173 | the interface processes received LLDP packets and never passes them | |
1174 | to OpenFlow or forwards them. | |
1175 | </li> | |
1176 | </ul> | |
1177 | ||
1178 | <p> | |
1179 | Set this option to <code>true</code> if the Open vSwitch bridge | |
1180 | connects different Ethernet networks and is not configured to | |
1181 | participate in STP. | |
1182 | </p> | |
1183 | ||
1184 | <p> | |
1185 | This option affects packets with the following destination MAC | |
1186 | addresses: | |
1187 | </p> | |
da1e25d5 | 1188 | |
05be4e2c EJ |
1189 | <dl> |
1190 | <dt><code>01:80:c2:00:00:00</code></dt> | |
1191 | <dd>IEEE 802.1D Spanning Tree Protocol (STP).</dd> | |
1192 | ||
1193 | <dt><code>01:80:c2:00:00:01</code></dt> | |
1194 | <dd>IEEE Pause frame.</dd> | |
1195 | ||
1196 | <dt><code>01:80:c2:00:00:0<var>x</var></code></dt> | |
1197 | <dd>Other reserved protocols.</dd> | |
1198 | ||
7d48a4cc BP |
1199 | <dt><code>00:e0:2b:00:00:00</code></dt> |
1200 | <dd>Extreme Discovery Protocol (EDP).</dd> | |
c93f9a78 | 1201 | |
7d48a4cc | 1202 | <dt> |
039a8ccd BP |
1203 | <code>00:e0:2b:00:00:04</code> and <code>00:e0:2b:00:00:06</code> |
1204 | </dt> | |
7d48a4cc | 1205 | <dd>Ethernet Automatic Protection Switching (EAPS).</dd> |
c93f9a78 | 1206 | |
05be4e2c EJ |
1207 | <dt><code>01:00:0c:cc:cc:cc</code></dt> |
1208 | <dd> | |
1209 | Cisco Discovery Protocol (CDP), VLAN Trunking Protocol (VTP), | |
1210 | Dynamic Trunking Protocol (DTP), Port Aggregation Protocol (PAgP), | |
1211 | and others. | |
1212 | </dd> | |
1213 | ||
1214 | <dt><code>01:00:0c:cc:cc:cd</code></dt> | |
1215 | <dd>Cisco Shared Spanning Tree Protocol PVSTP+.</dd> | |
1216 | ||
1217 | <dt><code>01:00:0c:cd:cd:cd</code></dt> | |
1218 | <dd>Cisco STP Uplink Fast.</dd> | |
1219 | ||
1220 | <dt><code>01:00:0c:00:00:00</code></dt> | |
1221 | <dd>Cisco Inter Switch Link.</dd> | |
7d48a4cc BP |
1222 | |
1223 | <dt><code>01:00:0c:cc:cc:c<var>x</var></code></dt> | |
1224 | <dd>Cisco CFM.</dd> | |
05be4e2c | 1225 | </dl> |
21f7563c | 1226 | </column> |
e764773c BP |
1227 | |
1228 | <column name="other_config" key="mac-aging-time" | |
1229 | type='{"type": "integer", "minInteger": 1}'> | |
1230 | <p> | |
1231 | The maximum number of seconds to retain a MAC learning entry for | |
1232 | which no packets have been seen. The default is currently 300 | |
1233 | seconds (5 minutes). The value, if specified, is forced into a | |
1234 | reasonable range, currently 15 to 3600 seconds. | |
1235 | </p> | |
1236 | ||
1237 | <p> | |
1238 | A short MAC aging time allows a network to more quickly detect that a | |
1239 | host is no longer connected to a switch port. However, it also makes | |
1240 | it more likely that packets will be flooded unnecessarily, when they | |
1241 | are addressed to a connected host that rarely transmits packets. To | |
1242 | reduce the incidence of unnecessary flooding, use a MAC aging time | |
1243 | longer than the maximum interval at which a host will ordinarily | |
1244 | transmit packets. | |
1245 | </p> | |
1246 | </column> | |
c4069512 BP |
1247 | |
1248 | <column name="other_config" key="mac-table-size" | |
1249 | type='{"type": "integer", "minInteger": 1}'> | |
1250 | <p> | |
1251 | The maximum number of MAC addresses to learn. The default is | |
1252 | currently 2048. The value, if specified, is forced into a reasonable | |
1253 | range, currently 10 to 1,000,000. | |
1254 | </p> | |
1255 | </column> | |
21f7563c JP |
1256 | </group> |
1257 | ||
3fd8d445 BP |
1258 | <group title="Common Columns"> |
1259 | The overall purpose of these columns is described under <code>Common | |
1260 | Columns</code> at the beginning of this document. | |
1261 | ||
1262 | <column name="other_config"/> | |
1263 | <column name="external_ids"/> | |
1264 | </group> | |
89365653 | 1265 | </table> |
039a8ccd BP |
1266 | |
1267 | <table name="Port" table="Port or bond configuration."> | |
89365653 BP |
1268 | <p>A port within a <ref table="Bridge"/>.</p> |
1269 | <p>Most commonly, a port has exactly one ``interface,'' pointed to by its | |
3fd8d445 BP |
1270 | <ref column="interfaces"/> column. Such a port logically |
1271 | corresponds to a port on a physical Ethernet switch. A port | |
1272 | with more than one interface is a ``bonded port'' (see | |
1273 | <ref group="Bonding Configuration"/>).</p> | |
89365653 | 1274 | <p>Some properties that one might think as belonging to a port are actually |
3fd8d445 | 1275 | part of the port's <ref table="Interface"/> members.</p> |
89365653 BP |
1276 | |
1277 | <column name="name"> | |
c885f934 BP |
1278 | Port name. For a non-bonded port, this should be the same as its |
1279 | interface's name. Port names must otherwise be unique among the names of | |
1280 | ports, interfaces, and bridges on a host. Because port and interfaces | |
1281 | names are usually the same, the restrictions on the <ref | |
1282 | table="Interface" column="name"/> column in the <ref table="Interface"/> | |
1283 | table, particularly on length, also apply to port names. Refer to the | |
1284 | documentation for <ref table="Interface"/> names for details. | |
89365653 BP |
1285 | </column> |
1286 | ||
1287 | <column name="interfaces"> | |
1288 | The port's interfaces. If there is more than one, this is a | |
1289 | bonded Port. | |
1290 | </column> | |
1291 | ||
1292 | <group title="VLAN Configuration"> | |
fed8962a EG |
1293 | <p> |
1294 | In short, a VLAN (short for ``virtual LAN'') is a way to partition a | |
1295 | single switch into multiple switches. VLANs can be confusing, so for | |
1296 | an introduction, please refer to the question ``What's a VLAN?'' in the | |
1297 | Open vSwitch FAQ. | |
1298 | </p> | |
1299 | ||
1300 | <p> | |
1301 | A VLAN is sometimes encoded into a packet using a 802.1Q or 802.1ad | |
1302 | VLAN header, but every packet is part of some VLAN whether or not it is | |
1303 | encoded in the packet. (A packet that appears to have no VLAN is part | |
1304 | of VLAN 0, by default.) As a result, it's useful to think of a VLAN as | |
1305 | a metadata property of a packet, separate from how the VLAN is encoded. | |
1306 | For a given port, this column determines how the encoding of a packet | |
1307 | that ingresses or egresses the port maps to the packet's VLAN. When a | |
1308 | packet enters the switch, its VLAN is determined based on its setting | |
1309 | in this column and its VLAN headers, if any, and then, conceptually, | |
1310 | the VLAN headers are then stripped off. Conversely, when a packet | |
1311 | exits the switch, its VLAN and the settings in this column determine | |
1312 | what VLAN headers, if any, are pushed onto the packet before it | |
1313 | egresses the port. | |
1314 | </p> | |
1315 | ||
1316 | <p> | |
1317 | The VLAN configuration in this column affects Open vSwitch only when it | |
1318 | is doing ``normal switching.'' It does not affect flows set up by an | |
1319 | OpenFlow controller, outside of the OpenFlow ``normal action.'' | |
1320 | </p> | |
1321 | ||
1322 | <p> | |
1323 | Bridge ports support the following types of VLAN configuration: | |
1324 | </p> | |
1325 | ||
ecac4ebf BP |
1326 | <dl> |
1327 | <dt>trunk</dt> | |
1328 | <dd> | |
1329 | <p> | |
1330 | A trunk port carries packets on one or more specified VLANs | |
1331 | specified in the <ref column="trunks"/> column (often, on every | |
1332 | VLAN). A packet that ingresses on a trunk port is in the VLAN | |
1333 | specified in its 802.1Q header, or VLAN 0 if the packet has no | |
1334 | 802.1Q header. A packet that egresses through a trunk port will | |
5e9ceccd | 1335 | have an 802.1Q header if it has a nonzero VLAN ID. |
ecac4ebf BP |
1336 | </p> |
1337 | ||
1338 | <p> | |
1339 | Any packet that ingresses on a trunk port tagged with a VLAN that | |
1340 | the port does not trunk is dropped. | |
1341 | </p> | |
1342 | </dd> | |
1343 | ||
1344 | <dt>access</dt> | |
1345 | <dd> | |
1346 | <p> | |
1347 | An access port carries packets on exactly one VLAN specified in the | |
5e9ceccd BP |
1348 | <ref column="tag"/> column. Packets egressing on an access port |
1349 | have no 802.1Q header. | |
ecac4ebf BP |
1350 | </p> |
1351 | ||
1352 | <p> | |
5e9ceccd BP |
1353 | Any packet with an 802.1Q header with a nonzero VLAN ID that |
1354 | ingresses on an access port is dropped, regardless of whether the | |
1355 | VLAN ID in the header is the access port's VLAN ID. | |
ecac4ebf BP |
1356 | </p> |
1357 | </dd> | |
1358 | ||
1359 | <dt>native-tagged</dt> | |
1360 | <dd> | |
1361 | A native-tagged port resembles a trunk port, with the exception that | |
1362 | a packet without an 802.1Q header that ingresses on a native-tagged | |
1363 | port is in the ``native VLAN'' (specified in the <ref column="tag"/> | |
1364 | column). | |
1365 | </dd> | |
1366 | ||
1367 | <dt>native-untagged</dt> | |
1368 | <dd> | |
1369 | A native-untagged port resembles a native-tagged port, with the | |
1370 | exception that a packet that egresses on a native-untagged port in | |
5e9ceccd | 1371 | the native VLAN will not have an 802.1Q header. |
ecac4ebf | 1372 | </dd> |
fed8962a EG |
1373 | |
1374 | <dt>dot1q-tunnel</dt> | |
1375 | <dd> | |
1376 | <p> | |
1377 | A dot1q-tunnel port is somewhat like an access port. Like an | |
1378 | access port, it carries packets on the single VLAN specified in the | |
1379 | <ref column="tag"/> column and this VLAN, called the service VLAN, | |
1380 | does not appear in an 802.1Q header for packets that ingress or | |
1381 | egress on the port. The main difference lies in the behavior when | |
1382 | packets that include a 802.1Q header ingress on the port. Whereas | |
1383 | an access port drops such packets, a dot1q-tunnel port treats these | |
1384 | as double-tagged with the outer service VLAN <ref column="tag"/> | |
1385 | and the inner customer VLAN taken from the 802.1Q header. | |
1386 | Correspondingly, to egress on the port, a packet outer VLAN (or | |
1387 | only VLAN) must be <ref column="tag"/>, which is removed before | |
1388 | egress, which exposes the inner (customer) VLAN if one is present. | |
1389 | </p> | |
1390 | ||
1391 | <p> | |
1392 | If <ref column="cvlans"/> is set, only allows packets in the | |
1393 | specified customer VLANs. | |
1394 | </p> | |
1395 | </dd> | |
ecac4ebf BP |
1396 | </dl> |
1397 | <p> | |
1398 | A packet will only egress through bridge ports that carry the VLAN of | |
1399 | the packet, as described by the rules above. | |
89365653 BP |
1400 | </p> |
1401 | ||
ecac4ebf | 1402 | <column name="vlan_mode"> |
7894d33b | 1403 | <p> |
ecac4ebf BP |
1404 | The VLAN mode of the port, as described above. When this column is |
1405 | empty, a default mode is selected as follows: | |
7894d33b | 1406 | </p> |
ecac4ebf BP |
1407 | <ul> |
1408 | <li> | |
1409 | If <ref column="tag"/> contains a value, the port is an access | |
1410 | port. The <ref column="trunks"/> column should be empty. | |
1411 | </li> | |
1412 | <li> | |
1413 | Otherwise, the port is a trunk port. The <ref column="trunks"/> | |
1414 | column value is honored if it is present. | |
1415 | </li> | |
1416 | </ul> | |
1417 | </column> | |
1418 | ||
1419 | <column name="tag"> | |
7894d33b | 1420 | <p> |
ecac4ebf BP |
1421 | For an access port, the port's implicitly tagged VLAN. For a |
1422 | native-tagged or native-untagged port, the port's native VLAN. Must | |
1423 | be empty if this is a trunk port. | |
7894d33b | 1424 | </p> |
89365653 BP |
1425 | </column> |
1426 | ||
1427 | <column name="trunks"> | |
7894d33b | 1428 | <p> |
ecac4ebf BP |
1429 | For a trunk, native-tagged, or native-untagged port, the 802.1Q VLAN |
1430 | or VLANs that this port trunks; if it is empty, then the port trunks | |
1431 | all VLANs. Must be empty if this is an access port. | |
7894d33b BP |
1432 | </p> |
1433 | <p> | |
ecac4ebf BP |
1434 | A native-tagged or native-untagged port always trunks its native |
1435 | VLAN, regardless of whether <ref column="trunks"/> includes that | |
1436 | VLAN. | |
7894d33b | 1437 | </p> |
89365653 | 1438 | </column> |
5e9ceccd | 1439 | |
fed8962a EG |
1440 | <column name="cvlans"> |
1441 | <p> | |
1442 | For a dot1q-tunnel port, the customer VLANs that this port includes. | |
1443 | If this is empty, the port includes all customer VLANs. | |
1444 | </p> | |
1445 | <p> | |
1446 | For other kinds of ports, this setting is ignored. | |
1447 | </p> | |
1448 | </column> | |
1449 | ||
1450 | <column name="other_config" key="qinq-ethtype" | |
1451 | type='{"type": "string", "enum": ["set", ["802.1ad", "802.1q"]]}'> | |
1452 | <p> | |
1453 | For a dot1q-tunnel port, this is the TPID for the service tag, that | |
1454 | is, for the 802.1Q header that contains the service VLAN ID. Because | |
1455 | packets that actually ingress and egress a dot1q-tunnel port do not | |
1456 | include an 802.1Q header for the service VLAN, this does not affect | |
1457 | packets on the dot1q-tunnel port itself. Rather, it determines the | |
1458 | service VLAN for a packet that ingresses on a dot1q-tunnel port and | |
1459 | egresses on a trunk port. | |
1460 | </p> | |
1461 | <p> | |
1462 | The value <code>802.1ad</code> specifies TPID 0x88a8, which is also | |
1463 | the default if the setting is omitted. The value <code>802.1q</code> | |
1464 | specifies TPID 0x8100. | |
1465 | </p> | |
1466 | <p> | |
1467 | For other kinds of ports, this setting is ignored. | |
1468 | </p> | |
1469 | </column> | |
1470 | ||
5e9ceccd BP |
1471 | <column name="other_config" key="priority-tags" |
1472 | type='{"type": "boolean"}'> | |
1473 | <p> | |
1474 | An 802.1Q header contains two important pieces of information: a VLAN | |
1475 | ID and a priority. A frame with a zero VLAN ID, called a | |
1476 | ``priority-tagged'' frame, is supposed to be treated the same way as | |
1477 | a frame without an 802.1Q header at all (except for the priority). | |
1478 | </p> | |
1479 | ||
1480 | <p> | |
1481 | However, some network elements ignore any frame that has 802.1Q | |
1482 | header at all, even when the VLAN ID is zero. Therefore, by default | |
1483 | Open vSwitch does not output priority-tagged frames, instead omitting | |
1484 | the 802.1Q header entirely if the VLAN ID is zero. Set this key to | |
1485 | <code>true</code> to enable priority-tagged frames on a port. | |
1486 | </p> | |
1487 | ||
1488 | <p> | |
1489 | Regardless of this setting, Open vSwitch omits the 802.1Q header on | |
1490 | output if both the VLAN ID and priority would be zero. | |
1491 | </p> | |
1492 | ||
1493 | <p> | |
1494 | All frames output to native-tagged ports have a nonzero VLAN ID, so | |
1495 | this setting is not meaningful on native-tagged ports. | |
1496 | </p> | |
1497 | </column> | |
89365653 BP |
1498 | </group> |
1499 | ||
1500 | <group title="Bonding Configuration"> | |
be02e7c3 | 1501 | <p>A port that has more than one interface is a ``bonded port.'' Bonding |
d64e1870 BP |
1502 | allows for load balancing and fail-over.</p> |
1503 | ||
1504 | <p> | |
1505 | The following types of bonding will work with any kind of upstream | |
1506 | switch. On the upstream switch, do not configure the interfaces as a | |
1507 | bond: | |
1508 | </p> | |
9f5073d8 EJ |
1509 | |
1510 | <dl> | |
1511 | <dt><code>balance-slb</code></dt> | |
1512 | <dd> | |
1513 | Balances flows among slaves based on source MAC address and output | |
1514 | VLAN, with periodic rebalancing as traffic patterns change. | |
1515 | </dd> | |
1516 | ||
1517 | <dt><code>active-backup</code></dt> | |
1518 | <dd> | |
1519 | Assigns all flows to one slave, failing over to a backup slave when | |
629d868c BP |
1520 | the active slave is disabled. This is the only bonding mode in which |
1521 | interfaces may be plugged into different upstream switches. | |
9f5073d8 EJ |
1522 | </dd> |
1523 | </dl> | |
1524 | ||
1525 | <p> | |
fb0b29a3 | 1526 | The following modes require the upstream switch to support 802.3ad with |
9dd165e0 RK |
1527 | successful LACP negotiation. If LACP negotiation fails and |
1528 | other-config:lacp-fallback-ab is true, then <code>active-backup</code> | |
1529 | mode is used: | |
9f5073d8 EJ |
1530 | </p> |
1531 | ||
1532 | <dl> | |
1533 | <dt><code>balance-tcp</code></dt> | |
1534 | <dd> | |
1535 | Balances flows among slaves based on L2, L3, and L4 protocol | |
1536 | information such as destination MAC address, IP address, and TCP | |
1537 | port. | |
1538 | </dd> | |
fb0b29a3 EJ |
1539 | </dl> |
1540 | ||
89365653 | 1541 | <p>These columns apply only to bonded ports. Their values are |
3fd8d445 | 1542 | otherwise ignored.</p> |
89365653 | 1543 | |
27dcaa1a | 1544 | <column name="bond_mode"> |
9f5073d8 | 1545 | <p>The type of bonding used for a bonded port. Defaults to |
4df08875 | 1546 | <code>active-backup</code> if unset. |
9f5073d8 | 1547 | </p> |
be02e7c3 EJ |
1548 | </column> |
1549 | ||
96ada1a4 EJ |
1550 | <column name="other_config" key="bond-hash-basis" |
1551 | type='{"type": "integer"}'> | |
1552 | An integer hashed along with flows when choosing output slaves in load | |
1553 | balanced bonds. When changed, all flows will be assigned different | |
1554 | hash values possibly causing slave selection decisions to change. Does | |
1555 | not affect bonding modes which do not employ load balancing such as | |
1556 | <code>active-backup</code>. | |
1557 | </column> | |
1558 | ||
3fd8d445 BP |
1559 | <group title="Link Failure Detection"> |
1560 | <p> | |
1561 | An important part of link bonding is detecting that links are down so | |
1562 | that they may be disabled. These settings determine how Open vSwitch | |
1563 | detects link failure. | |
1564 | </p> | |
89365653 | 1565 | |
f9e5e5b3 BP |
1566 | <column name="other_config" key="bond-detect-mode" |
1567 | type='{"type": "string", "enum": ["set", ["carrier", "miimon"]]}'> | |
1568 | The means used to detect link failures. Defaults to | |
3fd8d445 BP |
1569 | <code>carrier</code> which uses each interface's carrier to detect |
1570 | failures. When set to <code>miimon</code>, will check for failures | |
1571 | by polling each interface's MII. | |
1572 | </column> | |
89365653 | 1573 | |
f9e5e5b3 BP |
1574 | <column name="other_config" key="bond-miimon-interval" |
1575 | type='{"type": "integer"}'> | |
3fd8d445 BP |
1576 | The interval, in milliseconds, between successive attempts to poll |
1577 | each interface's MII. Relevant only when <ref column="other_config" | |
1578 | key="bond-detect-mode"/> is <code>miimon</code>. | |
1579 | </column> | |
1580 | ||
1581 | <column name="bond_updelay"> | |
1582 | <p> | |
1c144051 | 1583 | The number of milliseconds for which the link must stay up on an |
3fd8d445 BP |
1584 | interface before the interface is considered to be up. Specify |
1585 | <code>0</code> to enable the interface immediately. | |
1586 | </p> | |
1587 | ||
1588 | <p> | |
1589 | This setting is honored only when at least one bonded interface is | |
1590 | already enabled. When no interfaces are enabled, then the first | |
1591 | bond interface to come up is enabled immediately. | |
1592 | </p> | |
1593 | </column> | |
1594 | ||
1595 | <column name="bond_downdelay"> | |
1c144051 | 1596 | The number of milliseconds for which the link must stay down on an |
3fd8d445 BP |
1597 | interface before the interface is considered to be down. Specify |
1598 | <code>0</code> to disable the interface immediately. | |
1599 | </column> | |
1600 | </group> | |
c25c91fd | 1601 | |
3fd8d445 BP |
1602 | <group title="LACP Configuration"> |
1603 | <p> | |
1604 | LACP, the Link Aggregation Control Protocol, is an IEEE standard that | |
1605 | allows switches to automatically detect that they are connected by | |
1606 | multiple links and aggregate across those links. These settings | |
1607 | control LACP behavior. | |
1608 | </p> | |
1609 | ||
1610 | <column name="lacp"> | |
1611 | Configures LACP on this port. LACP allows directly connected | |
76ea8efd AE |
1612 | switches to negotiate which links may be bonded. LACP may be enabled |
1613 | on non-bonded ports for the benefit of any switches they may be | |
c25c91fd EJ |
1614 | connected to. <code>active</code> ports are allowed to initiate LACP |
1615 | negotiations. <code>passive</code> ports are allowed to participate | |
1616 | in LACP negotiations initiated by a remote switch, but not allowed to | |
bdebeece EJ |
1617 | initiate such negotiations themselves. If LACP is enabled on a port |
1618 | whose partner switch does not support LACP, the bond will be | |
9dd165e0 RK |
1619 | disabled, unless other-config:lacp-fallback-ab is set to true. |
1620 | Defaults to <code>off</code> if unset. | |
3fd8d445 BP |
1621 | </column> |
1622 | ||
1623 | <column name="other_config" key="lacp-system-id"> | |
1624 | The LACP system ID of this <ref table="Port"/>. The system ID of a | |
1625 | LACP bond is used to identify itself to its partners. Must be a | |
a9bf011b EJ |
1626 | nonzero MAC address. Defaults to the bridge Ethernet address if |
1627 | unset. | |
3fd8d445 BP |
1628 | </column> |
1629 | ||
f9e5e5b3 BP |
1630 | <column name="other_config" key="lacp-system-priority" |
1631 | type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'> | |
3fd8d445 BP |
1632 | The LACP system priority of this <ref table="Port"/>. In LACP |
1633 | negotiations, link status decisions are made by the system with the | |
f9e5e5b3 | 1634 | numerically lower priority. |
3fd8d445 BP |
1635 | </column> |
1636 | ||
bf83f7c8 | 1637 | <column name="other_config" key="lacp-time" |
039a8ccd | 1638 | type='{"type": "string", "enum": ["set", ["fast", "slow"]]}'> |
3fd8d445 BP |
1639 | <p> |
1640 | The LACP timing which should be used on this <ref table="Port"/>. | |
bf83f7c8 EJ |
1641 | By default <code>slow</code> is used. When configured to be |
1642 | <code>fast</code> LACP heartbeats are requested at a rate of once | |
1643 | per second causing connectivity problems to be detected more | |
1644 | quickly. In <code>slow</code> mode, heartbeats are requested at a | |
1645 | rate of once every 30 seconds. | |
3fd8d445 BP |
1646 | </p> |
1647 | </column> | |
9dd165e0 RK |
1648 | |
1649 | <column name="other_config" key="lacp-fallback-ab" | |
039a8ccd | 1650 | type='{"type": "boolean"}'> |
9dd165e0 RK |
1651 | <p> |
1652 | Determines the behavior of openvswitch bond in LACP mode. If | |
1653 | the partner switch does not support LACP, setting this option | |
1654 | to <code>true</code> allows openvswitch to fallback to | |
1655 | active-backup. If the option is set to <code>false</code>, the | |
1656 | bond will be disabled. In both the cases, once the partner switch | |
1657 | is configured to LACP mode, the bond will use LACP. | |
1658 | </p> | |
1659 | </column> | |
3fd8d445 BP |
1660 | </group> |
1661 | ||
b62ee96f | 1662 | <group title="Rebalancing Configuration"> |
3fd8d445 BP |
1663 | <p> |
1664 | These settings control behavior when a bond is in | |
b62ee96f | 1665 | <code>balance-slb</code> or <code>balance-tcp</code> mode. |
3fd8d445 BP |
1666 | </p> |
1667 | ||
f9e5e5b3 | 1668 | <column name="other_config" key="bond-rebalance-interval" |
bc1b010c EJ |
1669 | type='{"type": "integer", "minInteger": 0, "maxInteger": 10000}'> |
1670 | For a load balanced bonded port, the number of milliseconds between | |
1671 | successive attempts to rebalance the bond, that is, to move flows | |
1672 | from one interface on the bond to another in an attempt to keep usage | |
1673 | of each interface roughly equal. If zero, load balancing is disabled | |
1c144051 | 1674 | on the bond (link failure still cause flows to move). If |
bc1b010c | 1675 | less than 1000ms, the rebalance interval will be 1000ms. |
3fd8d445 BP |
1676 | </column> |
1677 | </group> | |
1678 | ||
1679 | <column name="bond_fake_iface"> | |
1680 | For a bonded port, whether to create a fake internal interface with the | |
1681 | name of the port. Use only for compatibility with legacy software that | |
1682 | requires this. | |
1683 | </column> | |
89365653 BP |
1684 | </group> |
1685 | ||
01f13d4f BP |
1686 | <group title="Spanning Tree Protocol"> |
1687 | <p> | |
1688 | The configuration here is only meaningful, and the status is only | |
1689 | populated, when 802.1D-1998 Spanning Tree Protocol is enabled on the | |
1690 | port's <ref column="Bridge"/> with its <ref column="stp_enable"/> | |
1691 | column. | |
1692 | </p> | |
21f7563c | 1693 | |
01f13d4f BP |
1694 | <group title="STP Configuration"> |
1695 | <column name="other_config" key="stp-enable" | |
1696 | type='{"type": "boolean"}'> | |
1697 | When STP is enabled on a bridge, it is enabled by default on all of | |
1698 | the bridge's ports except bond, internal, and mirror ports (which do | |
1699 | not work with STP). If this column's value is <code>false</code>, | |
1700 | STP is disabled on the port. | |
1701 | </column> | |
21f7563c | 1702 | |
01f13d4f BP |
1703 | <column name="other_config" key="stp-port-num" |
1704 | type='{"type": "integer", "minInteger": 1, "maxInteger": 255}'> | |
1705 | The port number used for the lower 8 bits of the port-id. By | |
1706 | default, the numbers will be assigned automatically. If any | |
1707 | port's number is manually configured on a bridge, then they | |
1708 | must all be. | |
1709 | </column> | |
21f7563c | 1710 | |
01f13d4f BP |
1711 | <column name="other_config" key="stp-port-priority" |
1712 | type='{"type": "integer", "minInteger": 0, "maxInteger": 255}'> | |
1713 | The port's relative priority value for determining the root | |
1714 | port (the upper 8 bits of the port-id). A port with a lower | |
1715 | port-id will be chosen as the root port. By default, the | |
1716 | priority is 0x80. | |
1717 | </column> | |
1718 | ||
1719 | <column name="other_config" key="stp-path-cost" | |
1720 | type='{"type": "integer", "minInteger": 0, "maxInteger": 65535}'> | |
1721 | Spanning tree path cost for the port. A lower number indicates | |
1722 | a faster link. By default, the cost is based on the maximum | |
1723 | speed of the link. | |
1724 | </column> | |
1725 | </group> | |
1726 | ||
1727 | <group title="STP Status"> | |
1728 | <column name="status" key="stp_port_id"> | |
1729 | The port ID used in spanning tree advertisements for this port, as 4 | |
1730 | hex digits. Configuring the port ID is described in the | |
1731 | <code>stp-port-num</code> and <code>stp-port-priority</code> keys of | |
1732 | the <code>other_config</code> section earlier. | |
1733 | </column> | |
1734 | <column name="status" key="stp_state" | |
1735 | type='{"type": "string", "enum": ["set", | |
1736 | ["disabled", "listening", "learning", | |
1737 | "forwarding", "blocking"]]}'> | |
1738 | STP state of the port. | |
1739 | </column> | |
1740 | <column name="status" key="stp_sec_in_state" | |
1741 | type='{"type": "integer", "minInteger": 0}'> | |
1742 | The amount of time this port has been in the current STP state, in | |
1743 | seconds. | |
1744 | </column> | |
1745 | <column name="status" key="stp_role" | |
1746 | type='{"type": "string", "enum": ["set", | |
1747 | ["root", "designated", "alternate"]]}'> | |
1748 | STP role of the port. | |
1749 | </column> | |
1750 | </group> | |
21f7563c | 1751 | </group> |
d62d7cb1 | 1752 | |
01f13d4f BP |
1753 | <group title="Rapid Spanning Tree Protocol"> |
1754 | <p> | |
1755 | The configuration here is only meaningful, and the status and | |
1756 | statistics are only populated, when 802.1D-1998 Spanning Tree Protocol | |
1757 | is enabled on the port's <ref column="Bridge"/> with its <ref | |
1758 | column="stp_enable"/> column. | |
1759 | </p> | |
d62d7cb1 | 1760 | |
01f13d4f BP |
1761 | <group title="RSTP Configuration"> |
1762 | <column name="other_config" key="rstp-enable" | |
1763 | type='{"type": "boolean"}'> | |
1764 | When RSTP is enabled on a bridge, it is enabled by default on all of | |
1765 | the bridge's ports except bond, internal, and mirror ports (which do | |
1766 | not work with RSTP). If this column's value is <code>false</code>, | |
1767 | RSTP is disabled on the port. | |
1768 | </column> | |
d62d7cb1 | 1769 | |
01f13d4f BP |
1770 | <column name="other_config" key="rstp-port-priority" |
1771 | type='{"type": "integer", "minInteger": 0, "maxInteger": 240}'> | |
1772 | The port's relative priority value for determining the root port, in | |
1773 | multiples of 16. By default, the port priority is 0x80 (128). Any | |
1774 | value in the lower 4 bits is rounded off. The significant upper 4 | |
1775 | bits become the upper 4 bits of the port-id. A port with the lowest | |
1776 | port-id is elected as the root. | |
1777 | </column> | |
d62d7cb1 | 1778 | |
01f13d4f BP |
1779 | <column name="other_config" key="rstp-port-num" |
1780 | type='{"type": "integer", "minInteger": 1, "maxInteger": 4095}'> | |
1781 | The local RSTP port number, used as the lower 12 bits of the port-id. | |
1782 | By default the port numbers are assigned automatically, and typically | |
1783 | may not correspond to the OpenFlow port numbers. A port with the | |
1784 | lowest port-id is elected as the root. | |
1785 | </column> | |
d62d7cb1 | 1786 | |
01f13d4f BP |
1787 | <column name="other_config" key="rstp-port-path-cost" |
1788 | type='{"type": "integer"}'> | |
1789 | The port path cost. The Port's contribution, when it is | |
1790 | the Root Port, to the Root Path Cost for the Bridge. By default the | |
1791 | cost is automatically calculated from the port's speed. | |
1792 | </column> | |
d62d7cb1 | 1793 | |
01f13d4f BP |
1794 | <column name="other_config" key="rstp-port-admin-edge" |
1795 | type='{"type": "boolean"}'> | |
1796 | The admin edge port parameter for the Port. Default is | |
1797 | <code>false</code>. | |
1798 | </column> | |
d62d7cb1 | 1799 | |
01f13d4f BP |
1800 | <column name="other_config" key="rstp-port-auto-edge" |
1801 | type='{"type": "boolean"}'> | |
1802 | The auto edge port parameter for the Port. Default is | |
d62d7cb1 | 1803 | <code>true</code>. |
01f13d4f BP |
1804 | </column> |
1805 | ||
1806 | <column name="other_config" key="rstp-port-mcheck" | |
1807 | type='{"type": "boolean"}'> | |
1808 | <p> | |
1809 | The mcheck port parameter for the Port. Default is | |
1810 | <code>false</code>. May be set to force the Port Protocol | |
1811 | Migration state machine to transmit RST BPDUs for a | |
1812 | MigrateTime period, to test whether all STP Bridges on the | |
1813 | attached LAN have been removed and the Port can continue to | |
1814 | transmit RSTP BPDUs. Setting mcheck has no effect if the | |
1815 | Bridge is operating in STP Compatibility mode. | |
1816 | </p> | |
1817 | <p> | |
1818 | Changing the value from <code>true</code> to | |
1819 | <code>false</code> has no effect, but needs to be done if | |
1820 | this behavior is to be triggered again by subsequently | |
1821 | changing the value from <code>false</code> to | |
1822 | <code>true</code>. | |
1823 | </p> | |
1824 | </column> | |
1825 | </group> | |
1826 | ||
1827 | <group title="RSTP Status"> | |
1828 | <column name="rstp_status" key="rstp_port_id"> | |
1829 | The port ID used in spanning tree advertisements for this port, as 4 | |
1830 | hex digits. Configuring the port ID is described in the | |
1831 | <code>rstp-port-num</code> and <code>rstp-port-priority</code> keys | |
1832 | of the <code>other_config</code> section earlier. | |
1833 | </column> | |
1834 | <column name="rstp_status" key="rstp_port_role" | |
1835 | type='{"type": "string", "enum": ["set", | |
1836 | ["Root", "Designated", "Alternate", "Backup", "Disabled"]]}'> | |
1837 | RSTP role of the port. | |
1838 | </column> | |
1839 | <column name="rstp_status" key="rstp_port_state" | |
1840 | type='{"type": "string", "enum": ["set", | |
1841 | ["Disabled", "Learning", "Forwarding", "Discarding"]]}'> | |
1842 | RSTP state of the port. | |
1843 | </column> | |
1844 | <column name="rstp_status" key="rstp_designated_bridge_id"> | |
1845 | The port's RSTP designated bridge ID, in the same form as <ref | |
1846 | column="rstp_status" key="rstp_bridge_id"/> in the <ref | |
1847 | table="Bridge"/> table. | |
1848 | </column> | |
1849 | <column name="rstp_status" key="rstp_designated_port_id"> | |
1850 | The port's RSTP designated port ID, as 4 hex digits. | |
1851 | </column> | |
1852 | <column name="rstp_status" key="rstp_designated_path_cost" | |
1853 | type='{"type": "integer"}'> | |
1854 | The port's RSTP designated path cost. Lower is better. | |
1855 | </column> | |
1856 | </group> | |
1857 | ||
1858 | <group title="RSTP Statistics"> | |
1859 | <column name="rstp_statistics" key="rstp_tx_count"> | |
1860 | Number of RSTP BPDUs transmitted through this port. | |
1861 | </column> | |
1862 | <column name="rstp_statistics" key="rstp_rx_count"> | |
1863 | Number of valid RSTP BPDUs received by this port. | |
1864 | </column> | |
1865 | <column name="rstp_statistics" key="rstp_error_count"> | |
1866 | Number of invalid RSTP BPDUs received by this port. | |
1867 | </column> | |
1868 | <column name="rstp_statistics" key="rstp_uptime"> | |
1869 | The duration covered by the other RSTP statistics, in seconds. | |
1870 | </column> | |
1871 | </group> | |
d62d7cb1 JR |
1872 | </group> |
1873 | ||
dc2b70ba FL |
1874 | <group title="Multicast Snooping"> |
1875 | <column name="other_config" key="mcast-snooping-flood" | |
1876 | type='{"type": "boolean"}'> | |
1877 | <p> | |
8e04a33f FL |
1878 | If set to <code>true</code>, multicast packets (except Reports) are |
1879 | unconditionally forwarded to the specific port. | |
1880 | </p> | |
1881 | </column> | |
1882 | <column name="other_config" key="mcast-snooping-flood-reports" | |
1883 | type='{"type": "boolean"}'> | |
1884 | <p> | |
1885 | If set to <code>true</code>, multicast Reports are unconditionally | |
dc2b70ba FL |
1886 | forwarded to the specific port. |
1887 | </p> | |
1888 | </column> | |
1889 | </group> | |
21f7563c | 1890 | |
89365653 | 1891 | <group title="Other Features"> |
c1c9c9c4 BP |
1892 | <column name="qos"> |
1893 | Quality of Service configuration for this port. | |
1894 | </column> | |
299a244b | 1895 | |
89365653 BP |
1896 | <column name="mac"> |
1897 | The MAC address to use for this port for the purpose of choosing the | |
1898 | bridge's MAC address. This column does not necessarily reflect the | |
1899 | port's actual MAC address, nor will setting it change the port's actual | |
1900 | MAC address. | |
1901 | </column> | |
1902 | ||
1903 | <column name="fake_bridge"> | |
1904 | Does this port represent a sub-bridge for its tagged VLAN within the | |
1905 | Bridge? See ovs-vsctl(8) for more information. | |
1906 | </column> | |
1907 | ||
ced947ae BK |
1908 | <column name="protected" type='{"type": "boolean"}'> |
1909 | The protected ports feature allows certain ports to be designated as | |
1910 | protected. Traffic between protected ports is blocked. Protected | |
1911 | ports can send traffic to unprotected ports. Unprotected ports can | |
1912 | send traffic to any port. | |
1913 | Default is false. | |
1914 | </column> | |
1915 | ||
3fd8d445 BP |
1916 | <column name="external_ids" key="fake-bridge-id-*"> |
1917 | External IDs for a fake bridge (see the <ref column="fake_bridge"/> | |
1918 | column) are defined by prefixing a <ref table="Bridge"/> <ref | |
1919 | table="Bridge" column="external_ids"/> key with | |
1920 | <code>fake-bridge-</code>, | |
1921 | e.g. <code>fake-bridge-xs-network-uuids</code>. | |
89365653 | 1922 | </column> |
54b21db7 TLSC |
1923 | |
1924 | <column name="other_config" key="transient" | |
1925 | type='{"type": "boolean"}'> | |
1926 | <p> | |
1927 | If set to <code>true</code>, the port will be removed when | |
1928 | <code>ovs-ctl start --delete-transient-ports</code> is used. | |
1929 | </p> | |
1930 | </column> | |
3fd8d445 | 1931 | </group> |
89365653 | 1932 | |
01f13d4f BP |
1933 | <column name="bond_active_slave"> |
1934 | For a bonded port, record the mac address of the current active slave. | |
1935 | </column> | |
21f7563c | 1936 | |
80740385 JP |
1937 | <group title="Port Statistics"> |
1938 | <p> | |
12eb035b AW |
1939 | Key-value pairs that report port statistics. The update period |
1940 | is controlled by <ref column="other_config" | |
1941 | key="stats-update-interval"/> in the <code>Open_vSwitch</code> table. | |
80740385 JP |
1942 | </p> |
1943 | <group title="Statistics: STP transmit and receive counters"> | |
1944 | <column name="statistics" key="stp_tx_count"> | |
1945 | Number of STP BPDUs sent on this port by the spanning | |
1946 | tree library. | |
1947 | </column> | |
1948 | <column name="statistics" key="stp_rx_count"> | |
1949 | Number of STP BPDUs received on this port and accepted by the | |
1950 | spanning tree library. | |
1951 | </column> | |
1952 | <column name="statistics" key="stp_error_count"> | |
1953 | Number of bad STP BPDUs received on this port. Bad BPDUs | |
1954 | include runt packets and those with an unexpected protocol ID. | |
1955 | </column> | |
1956 | </group> | |
1957 | </group> | |
1958 | ||
3fd8d445 BP |
1959 | <group title="Common Columns"> |
1960 | The overall purpose of these columns is described under <code>Common | |
1961 | Columns</code> at the beginning of this document. | |
1962 | ||
1963 | <column name="other_config"/> | |
1964 | <column name="external_ids"/> | |
89365653 BP |
1965 | </group> |
1966 | </table> | |
1967 | ||
1968 | <table name="Interface" title="One physical network device in a Port."> | |
1969 | An interface within a <ref table="Port"/>. | |
1970 | ||
1971 | <group title="Core Features"> | |
1972 | <column name="name"> | |
c885f934 BP |
1973 | <p> |
1974 | Interface name. Should be alphanumeric. For non-bonded port, this | |
1975 | should be the same as the port name. It must otherwise be unique | |
1976 | among the names of ports, interfaces, and bridges on a host. | |
1977 | </p> | |
1978 | ||
1979 | <p> | |
1980 | The maximum length of an interface name depends on the underlying | |
1981 | datapath: | |
1982 | </p> | |
1983 | ||
1984 | <ul> | |
1985 | <li> | |
1986 | The names of interfaces implemented as Linux and BSD network | |
1987 | devices, including interfaces with type <code>internal</code>, | |
59a0ef1d JG |
1988 | <code>tap</code>, or <code>system</code> plus the different types |
1989 | of tunnel ports, are limited to 15 bytes. Windows limits these | |
1990 | names to 255 bytes. | |
c885f934 BP |
1991 | </li> |
1992 | ||
1993 | <li> | |
59a0ef1d JG |
1994 | The names of patch ports are not used in the underlying datapath, |
1995 | so operating system restrictions do not apply. Thus, they may have | |
1996 | arbitrary length. | |
c885f934 BP |
1997 | </li> |
1998 | </ul> | |
1999 | ||
2000 | <p> | |
2001 | Regardless of other restrictions, OpenFlow only supports 15-byte | |
2002 | names, which means that <code>ovs-ofctl</code> and OpenFlow | |
2003 | controllers will show names truncated to 15 bytes. | |
2004 | </p> | |
89365653 BP |
2005 | </column> |
2006 | ||
ea401d9a NM |
2007 | <column name="ifindex"> |
2008 | A positive interface index as defined for SNMP MIB-II in RFCs 1213 and | |
2009 | 2863, if the interface has one, otherwise 0. The ifindex is useful for | |
2010 | seamless integration with protocols such as SNMP and sFlow. | |
2011 | </column> | |
2012 | ||
df867eda JP |
2013 | <column name="mac_in_use"> |
2014 | The MAC address in use by this interface. | |
2015 | </column> | |
2016 | ||
89365653 BP |
2017 | <column name="mac"> |
2018 | <p>Ethernet address to set for this interface. If unset then the | |
3fd8d445 | 2019 | default MAC address is used:</p> |
89365653 BP |
2020 | <ul> |
2021 | <li>For the local interface, the default is the lowest-numbered MAC | |
3fd8d445 BP |
2022 | address among the other bridge ports, either the value of the |
2023 | <ref table="Port" column="mac"/> in its <ref table="Port"/> record, | |
2024 | if set, or its actual MAC (for bonded ports, the MAC of its slave | |
2025 | whose name is first in alphabetical order). Internal ports and | |
2026 | bridge ports that are used as port mirroring destinations (see the | |
2027 | <ref table="Mirror"/> table) are ignored.</li> | |
2e57b537 | 2028 | <li>For other internal interfaces, the default MAC is randomly |
3fd8d445 | 2029 | generated.</li> |
89365653 | 2030 | <li>External interfaces typically have a MAC address associated with |
3fd8d445 | 2031 | their hardware.</li> |
89365653 BP |
2032 | </ul> |
2033 | <p>Some interfaces may not have a software-controllable MAC | |
2034 | address.</p> | |
2035 | </column> | |
2036 | ||
bbe6109d TG |
2037 | <column name="error"> |
2038 | If the configuration of the port failed, as indicated by -1 in <ref | |
2039 | column="ofport"/>, Open vSwitch sets this column to an error | |
2040 | description in human readable form. Otherwise, Open vSwitch clears | |
2041 | this column. | |
2042 | </column> | |
2043 | ||
484c8355 | 2044 | <group title="OpenFlow Port Number"> |
039a8ccd BP |
2045 | <p> |
2046 | When a client adds a new interface, Open vSwitch chooses an OpenFlow | |
2047 | port number for the new port. If the client that adds the port fills | |
2048 | in <ref column="ofport_request"/>, then Open vSwitch tries to use its | |
2049 | value as the OpenFlow port number. Otherwise, or if the requested | |
2050 | port number is already in use or cannot be used for another reason, | |
2051 | Open vSwitch automatically assigns a free port number. Regardless of | |
2052 | how the port number was obtained, Open vSwitch then reports in <ref | |
2053 | column="ofport"/> the port number actually assigned. | |
2054 | </p> | |
2055 | ||
2056 | <p> | |
2057 | Open vSwitch limits the port numbers that it automatically assigns to | |
2058 | the range 1 through 32,767, inclusive. Controllers therefore have | |
2059 | free use of ports 32,768 and up. | |
2060 | </p> | |
2061 | ||
2062 | <column name="ofport"> | |
2063 | <p> | |
2064 | OpenFlow port number for this interface. Open vSwitch sets this | |
2065 | column's value, so other clients should treat it as read-only. | |
2066 | </p> | |
2067 | <p> | |
2068 | The OpenFlow ``local'' port (<code>OFPP_LOCAL</code>) is 65,534. | |
2069 | The other valid port numbers are in the range 1 to 65,279, | |
2070 | inclusive. Value -1 indicates an error adding the interface. | |
2071 | </p> | |
2072 | </column> | |
2073 | ||
2074 | <column name="ofport_request" | |
2075 | type='{"type": "integer", "minInteger": 1, "maxInteger": 65279}'> | |
2076 | <p> | |
2077 | Requested OpenFlow port number for this interface. | |
2078 | </p> | |
2079 | ||
2080 | <p> | |
2081 | A client should ideally set this column's value in the same | |
2082 | database transaction that it uses to create the interface. Open | |
2083 | vSwitch version 2.1 and later will honor a later request for a | |
2084 | specific port number, althuogh it might confuse some controllers: | |
2085 | OpenFlow does not have a way to announce a port number change, so | |
2086 | Open vSwitch represents it over OpenFlow as a port deletion | |
2087 | followed immediately by a port addition. | |
2088 | </p> | |
2089 | ||
2090 | <p> | |
2091 | If <ref column="ofport_request"/> is set or changed to some other | |
2092 | port's automatically assigned port number, Open vSwitch chooses a | |
2093 | new port number for the latter port. | |
2094 | </p> | |
2095 | </column> | |
484c8355 | 2096 | </group> |
89365653 BP |
2097 | </group> |
2098 | ||
2099 | <group title="System-Specific Details"> | |
2100 | <column name="type"> | |
3fd8d445 | 2101 | <p> |
842733c3 MG |
2102 | The interface type. The types supported by a particular instance of |
2103 | Open vSwitch are listed in the <ref table="Open_vSwitch" | |
2104 | column="iface_types"/> column in the <ref table="Open_vSwitch"/> | |
2105 | table. The following types are defined: | |
3fd8d445 BP |
2106 | </p> |
2107 | ||
89365653 BP |
2108 | <dl> |
2109 | <dt><code>system</code></dt> | |
2110 | <dd>An ordinary network device, e.g. <code>eth0</code> on Linux. | |
3fd8d445 BP |
2111 | Sometimes referred to as ``external interfaces'' since they are |
2112 | generally connected to hardware external to that on which the Open | |
2113 | vSwitch is running. The empty string is a synonym for | |
2114 | <code>system</code>.</dd> | |
2115 | ||
89365653 | 2116 | <dt><code>internal</code></dt> |
2e57b537 | 2117 | <dd>A simulated network device that sends and receives traffic. An |
3fd8d445 BP |
2118 | internal interface whose <ref column="name"/> is the same as its |
2119 | bridge's <ref table="Open_vSwitch" column="name"/> is called the | |
2120 | ``local interface.'' It does not make sense to bond an internal | |
2121 | interface, so the terms ``port'' and ``interface'' are often used | |
2122 | imprecisely for internal interfaces.</dd> | |
2123 | ||
89365653 BP |
2124 | <dt><code>tap</code></dt> |
2125 | <dd>A TUN/TAP device managed by Open vSwitch.</dd> | |
3fd8d445 | 2126 | |
c1fc1411 JG |
2127 | <dt><code>geneve</code></dt> |
2128 | <dd> | |
59a0ef1d | 2129 | An Ethernet over Geneve (<code>http://tools.ietf.org/html/draft-ietf-nvo3-geneve</code>) |
80c4589a | 2130 | IPv4/IPv6 tunnel. |
c1fc1411 | 2131 | |
9558d2a5 JG |
2132 | A description of how to match and set Geneve options can be found |
2133 | in the <code>ovs-ofctl</code> manual page. | |
c1fc1411 JG |
2134 | </dd> |
2135 | ||
89365653 | 2136 | <dt><code>gre</code></dt> |
3fd8d445 | 2137 | <dd> |
80c4589a | 2138 | An Ethernet over RFC 2890 Generic Routing Encapsulation over IPv4/IPv6 |
79f827fa | 2139 | tunnel. |
e16a28b5 | 2140 | </dd> |
3fd8d445 | 2141 | |
79f827fa KM |
2142 | <dt><code>vxlan</code></dt> |
2143 | <dd> | |
039a8ccd BP |
2144 | <p> |
2145 | An Ethernet tunnel over the UDP-based VXLAN protocol described in | |
2146 | RFC 7348. | |
2147 | </p> | |
2148 | <p> | |
59a0ef1d JG |
2149 | Open vSwitch uses IANA-assigned UDP destination port 4789. The |
2150 | source port used for VXLAN traffic varies on a per-flow basis | |
2151 | and is in the ephemeral port range. | |
039a8ccd | 2152 | </p> |
79f827fa KM |
2153 | </dd> |
2154 | ||
a6ae068b LJ |
2155 | <dt><code>lisp</code></dt> |
2156 | <dd> | |
a6363cfd LJ |
2157 | <p> |
2158 | A layer 3 tunnel over the experimental, UDP-based Locator/ID | |
2159 | Separation Protocol (RFC 6830). | |
2160 | </p> | |
2161 | <p> | |
2162 | Only IPv4 and IPv6 packets are supported by the protocol, and | |
2163 | they are sent and received without an Ethernet header. Traffic | |
2164 | to/from LISP ports is expected to be configured explicitly, and | |
2165 | the ports are not intended to participate in learning based | |
2166 | switching. As such, they are always excluded from packet | |
2167 | flooding. | |
2168 | </p> | |
a6ae068b LJ |
2169 | </dd> |
2170 | ||
4237026e PS |
2171 | <dt><code>stt</code></dt> |
2172 | <dd> | |
039a8ccd BP |
2173 | The Stateless TCP Tunnel (STT) is particularly useful when tunnel |
2174 | endpoints are in end-systems, as it utilizes the capabilities of | |
2175 | standard network interface cards to improve performance. STT utilizes | |
2176 | a TCP-like header inside the IP header. It is stateless, i.e., there is | |
2177 | no TCP connection state of any kind associated with the tunnel. The | |
2178 | TCP-like header is used to leverage the capabilities of existing | |
2179 | network interface cards, but should not be interpreted as implying | |
2180 | any sort of connection state between endpoints. | |
2181 | Since the STT protocol does not engage in the usual TCP 3-way handshake, | |
2182 | so it will have difficulty traversing stateful firewalls. | |
2183 | The protocol is documented at | |
59a0ef1d | 2184 | https://tools.ietf.org/html/draft-davie-stt |
039a8ccd | 2185 | |
59a0ef1d | 2186 | All traffic uses a default destination port of 7471. |
4237026e PS |
2187 | </dd> |
2188 | ||
8aed4223 | 2189 | <dt><code>patch</code></dt> |
eca2df31 | 2190 | <dd> |
3fd8d445 | 2191 | A pair of virtual devices that act as a patch cable. |
eca2df31 | 2192 | </dd> |
3fd8d445 | 2193 | |
84b32864 | 2194 | <dt><code>null</code></dt> |
0faed346 | 2195 | <dd>An ignored interface. Deprecated and slated for removal in |
039a8ccd | 2196 | February 2013.</dd> |
89365653 BP |
2197 | </dl> |
2198 | </column> | |
3fd8d445 BP |
2199 | </group> |
2200 | ||
2201 | <group title="Tunnel Options"> | |
2202 | <p> | |
2203 | These options apply to interfaces with <ref column="type"/> of | |
2b02d770 PS |
2204 | <code>geneve</code>, <code>gre</code>, <code>vxlan</code>, |
2205 | <code>lisp</code> and <code>stt</code>. | |
3fd8d445 BP |
2206 | </p> |
2207 | ||
2208 | <p> | |
2209 | Each tunnel must be uniquely identified by the combination of <ref | |
2210 | column="type"/>, <ref column="options" key="remote_ip"/>, <ref | |
2211 | column="options" key="local_ip"/>, and <ref column="options" | |
2212 | key="in_key"/>. If two ports are defined that are the same except one | |
2213 | has an optional identifier and the other does not, the more specific | |
2214 | one is matched first. <ref column="options" key="in_key"/> is | |
2215 | considered more specific than <ref column="options" key="local_ip"/> if | |
2216 | a port defines one and another port defines the other. | |
2217 | </p> | |
2218 | ||
2219 | <column name="options" key="remote_ip"> | |
0ad90c84 JR |
2220 | <p>Required. The remote tunnel endpoint, one of:</p> |
2221 | ||
2222 | <ul> | |
2223 | <li> | |
80c4589a | 2224 | An IPv4 or IPv6 address (not a DNS name), e.g. <code>192.168.0.123</code>. |
0ad90c84 JR |
2225 | Only unicast endpoints are supported. |
2226 | </li> | |
2227 | <li> | |
2228 | The word <code>flow</code>. The tunnel accepts packets from any | |
2229 | remote tunnel endpoint. To process only packets from a specific | |
2230 | remote tunnel endpoint, the flow entries may match on the | |
80c4589a PS |
2231 | <code>tun_src</code> or <code>tun_ipv6_src</code>field. When |
2232 | sending packets to a <code>remote_ip=flow</code> tunnel, the flow | |
2233 | actions must explicitly set the <code>tun_dst</code> or | |
2234 | <code>tun_ipv6_dst</code> field to the IP address of the desired | |
2235 | remote tunnel endpoint, e.g. with a <code>set_field</code> action. | |
0ad90c84 JR |
2236 | </li> |
2237 | </ul> | |
2238 | ||
2239 | <p> | |
039a8ccd BP |
2240 | The remote tunnel endpoint for any packet received from a tunnel |
2241 | is available in the <code>tun_src</code> field for matching in the | |
2242 | flow table. | |
0ad90c84 | 2243 | </p> |
3fd8d445 BP |
2244 | </column> |
2245 | ||
2246 | <column name="options" key="local_ip"> | |
0ad90c84 JR |
2247 | <p> |
2248 | Optional. The tunnel destination IP that received packets must | |
2249 | match. Default is to match all addresses. If specified, may be one | |
2250 | of: | |
2251 | </p> | |
2252 | ||
2253 | <ul> | |
2254 | <li> | |
80c4589a | 2255 | An IPv4/IPv6 address (not a DNS name), e.g. <code>192.168.12.3</code>. |
0ad90c84 JR |
2256 | </li> |
2257 | <li> | |
2258 | The word <code>flow</code>. The tunnel accepts packets sent to any | |
2259 | of the local IP addresses of the system running OVS. To process | |
2260 | only packets sent to a specific IP address, the flow entries may | |
80c4589a PS |
2261 | match on the <code>tun_dst</code> or <code>tun_ipv6_dst</code> field. |
2262 | When sending packets to a <code>local_ip=flow</code> tunnel, the flow | |
2263 | actions may explicitly set the <code>tun_src</code> or <code>tun_ipv6_src</code> | |
2264 | field to the desired IP address, e.g. with a <code>set_field</code> action. | |
2265 | However, while routing the tunneled packet out, the local system may | |
2266 | override the specified address with the local IP address configured for the | |
0ad90c84 JR |
2267 | outgoing system interface. |
2268 | ||
2269 | <p> | |
2270 | This option is valid only for tunnels also configured with the | |
2271 | <code>remote_ip=flow</code> option. | |
2272 | </p> | |
2273 | </li> | |
2274 | </ul> | |
2275 | ||
2276 | <p> | |
2277 | The tunnel destination IP address for any packet received from a | |
80c4589a PS |
2278 | tunnel is available in the <code>tun_dst</code> or <code>tun_ipv6_dst</code> |
2279 | field for matching in the flow table. | |
0ad90c84 | 2280 | </p> |
3fd8d445 BP |
2281 | </column> |
2282 | ||
2283 | <column name="options" key="in_key"> | |
2284 | <p>Optional. The key that received packets must contain, one of:</p> | |
2285 | ||
2286 | <ul> | |
2287 | <li> | |
2288 | <code>0</code>. The tunnel receives packets with no key or with a | |
2289 | key of 0. This is equivalent to specifying no <ref column="options" | |
2290 | key="in_key"/> at all. | |
2291 | </li> | |
2292 | <li> | |
271e6bc7 | 2293 | A positive 24-bit (for Geneve, VXLAN, and LISP), 32-bit (for GRE) |
99e7b077 | 2294 | or 64-bit (for STT) number. The tunnel receives only |
4237026e | 2295 | packets with the specified key. |
3fd8d445 BP |
2296 | </li> |
2297 | <li> | |
2298 | The word <code>flow</code>. The tunnel accepts packets with any | |
2299 | key. The key will be placed in the <code>tun_id</code> field for | |
2300 | matching in the flow table. The <code>ovs-ofctl</code> manual page | |
2301 | contains additional information about matching fields in OpenFlow | |
2302 | flows. | |
2303 | </li> | |
2304 | </ul> | |
2305 | ||
2306 | <p> | |
2307 | </p> | |
2308 | </column> | |
2309 | ||
2310 | <column name="options" key="out_key"> | |
2311 | <p>Optional. The key to be set on outgoing packets, one of:</p> | |
2312 | ||
2313 | <ul> | |
2314 | <li> | |
2315 | <code>0</code>. Packets sent through the tunnel will have no key. | |
2316 | This is equivalent to specifying no <ref column="options" | |
2317 | key="out_key"/> at all. | |
2318 | </li> | |
2319 | <li> | |
271e6bc7 | 2320 | A positive 24-bit (for Geneve, VXLAN and LISP), 32-bit (for GRE) or |
99e7b077 | 2321 | 64-bit (for STT) number. Packets sent through the tunnel |
4237026e | 2322 | will have the specified key. |
3fd8d445 BP |
2323 | </li> |
2324 | <li> | |
2325 | The word <code>flow</code>. Packets sent through the tunnel will | |
2326 | have the key set using the <code>set_tunnel</code> Nicira OpenFlow | |
2327 | vendor extension (0 is used in the absence of an action). The | |
2328 | <code>ovs-ofctl</code> manual page contains additional information | |
2329 | about the Nicira OpenFlow vendor extensions. | |
2330 | </li> | |
2331 | </ul> | |
2332 | </column> | |
2333 | ||
2334 | <column name="options" key="key"> | |
2335 | Optional. Shorthand to set <code>in_key</code> and | |
2336 | <code>out_key</code> at the same time. | |
2337 | </column> | |
2338 | ||
2339 | <column name="options" key="tos"> | |
2340 | Optional. The value of the ToS bits to be set on the encapsulating | |
749ae950 PS |
2341 | packet. ToS is interpreted as DSCP and ECN bits, ECN part must be |
2342 | zero. It may also be the word <code>inherit</code>, in which case | |
3fd8d445 BP |
2343 | the ToS will be copied from the inner packet if it is IPv4 or IPv6 |
2344 | (otherwise it will be 0). The ECN fields are always inherited. | |
2345 | Default is 0. | |
2346 | </column> | |
2347 | ||
2348 | <column name="options" key="ttl"> | |
2349 | Optional. The TTL to be set on the encapsulating packet. It may also | |
2350 | be the word <code>inherit</code>, in which case the TTL will be copied | |
2351 | from the inner packet if it is IPv4 or IPv6 (otherwise it will be the | |
2352 | system default, typically 64). Default is the system default TTL. | |
2353 | </column> | |
9cc6bf75 | 2354 | |
f9e5e5b3 BP |
2355 | <column name="options" key="df_default" |
2356 | type='{"type": "boolean"}'> | |
9b9f4d60 EJ |
2357 | Optional. If enabled, the Don't Fragment bit will be set on tunnel |
2358 | outer headers to allow path MTU discovery. Default is enabled; set | |
2359 | to <code>false</code> to disable. | |
3fd8d445 BP |
2360 | </column> |
2361 | ||
bf4bbd0d PS |
2362 | <column name="options" key="egress_pkt_mark"> |
2363 | Optional. The pkt_mark to be set on the encapsulating packet. This | |
2364 | option sets packet mark for the tunnel endpoint for all tunnel packets | |
2365 | including tunnel monitoring. | |
2366 | </column> | |
2367 | ||
526df7d8 TG |
2368 | <group title="Tunnel Options: vxlan only"> |
2369 | ||
039a8ccd BP |
2370 | <column name="options" key="exts"> |
2371 | <p>Optional. Comma separated list of optional VXLAN extensions to | |
2372 | enable. The following extensions are supported:</p> | |
526df7d8 | 2373 | |
039a8ccd BP |
2374 | <ul> |
2375 | <li> | |
2376 | <code>gbp</code>: VXLAN-GBP allows to transport the group policy | |
2377 | context of a packet across the VXLAN tunnel to other network | |
96fee5e0 BP |
2378 | peers. See the description of <code>tun_gbp_id</code> and |
2379 | <code>tun_gbp_flags</code> in <code>ovs-fields</code>(7) for | |
2380 | additional information. | |
039a8ccd BP |
2381 | (<code>https://tools.ietf.org/html/draft-smith-vxlan-group-policy</code>) |
2382 | </li> | |
2383 | </ul> | |
2384 | </column> | |
526df7d8 | 2385 | |
039a8ccd | 2386 | </group> |
526df7d8 | 2387 | |
2b02d770 | 2388 | <group title="Tunnel Options: gre, geneve, and vxlan"> |
3fd8d445 | 2389 | <p> |
2b02d770 | 2390 | <code>gre</code>, <code>geneve</code>, and |
4752cc0c | 2391 | <code>vxlan</code> interfaces support these options. |
3fd8d445 BP |
2392 | </p> |
2393 | ||
f9e5e5b3 | 2394 | <column name="options" key="csum" type='{"type": "boolean"}'> |
3fd8d445 | 2395 | <p> |
4752cc0c JG |
2396 | Optional. Compute encapsulation header (either GRE or UDP) |
2397 | checksums on outgoing packets. Default is disabled, set to | |
2398 | <code>true</code> to enable. Checksums present on incoming | |
2399 | packets will be validated regardless of this setting. | |
039a8ccd | 2400 | </p> |
3fd8d445 | 2401 | |
4752cc0c JG |
2402 | <p> |
2403 | When using the upstream Linux kernel module, computation of | |
2404 | checksums for <code>geneve</code> and <code>vxlan</code> requires | |
2405 | Linux kernel version 4.0 or higher. <code>gre</code> supports | |
2406 | checksums for all versions of Open vSwitch that support GRE. | |
2407 | The out of tree kernel module distributed as part of OVS | |
2408 | can compute all tunnel checksums on any kernel version that it | |
2409 | is compatible with. | |
3fd8d445 BP |
2410 | </p> |
2411 | ||
3fd8d445 BP |
2412 | </column> |
2413 | </group> | |
2414 | </group> | |
2415 | ||
2416 | <group title="Patch Options"> | |
2417 | <p> | |
2418 | Only <code>patch</code> interfaces support these options. | |
2419 | </p> | |
89365653 | 2420 | |
3fd8d445 BP |
2421 | <column name="options" key="peer"> |
2422 | The <ref column="name"/> of the <ref table="Interface"/> for the other | |
2423 | side of the patch. The named <ref table="Interface"/>'s own | |
2424 | <code>peer</code> option must specify this <ref table="Interface"/>'s | |
2425 | name. That is, the two patch interfaces must have reversed <ref | |
2426 | column="name"/> and <code>peer</code> values. | |
89365653 | 2427 | </column> |
e210037e AE |
2428 | </group> |
2429 | ||
a14b8947 IM |
2430 | <group title="PMD (Poll Mode Driver) Options"> |
2431 | <p> | |
2432 | Only PMD netdevs support these options. | |
2433 | </p> | |
2434 | ||
81acebda | 2435 | <column name="options" key="n_rxq" |
a14b8947 IM |
2436 | type='{"type": "integer", "minInteger": 1}'> |
2437 | <p> | |
2438 | Specifies the maximum number of rx queues to be created for PMD | |
2439 | netdev. If not specified or specified to 0, one rx queue will | |
2440 | be created by default. | |
81acebda | 2441 | Not supported by DPDK vHost interfaces. |
a14b8947 IM |
2442 | </p> |
2443 | </column> | |
3eb67853 | 2444 | |
55e075e6 CL |
2445 | <column name="options" key="dpdk-devargs" |
2446 | type='{"type": "string"}'> | |
2447 | <p> | |
69876ed7 CL |
2448 | Specifies the PCI address associated with the port for physical |
2449 | devices, or the virtual driver to be used for the port when a virtual | |
2450 | PMD is intended to be used. For the latter, the argument string | |
2451 | typically takes the form of eth_<driver_name><x> where | |
2452 | 'driver_name' is a valid virtual DPDK PMD driver name and where 'x' | |
2453 | lies in the range of 0 to RTE_MAX_ETHPORTS-1. | |
2454 | Only supported by the dpdk port type. | |
55e075e6 CL |
2455 | </p> |
2456 | </column> | |
2457 | ||
3eb67853 IM |
2458 | <column name="other_config" key="pmd-rxq-affinity"> |
2459 | <p>Specifies mapping of RX queues of this interface to CPU cores.</p> | |
2460 | <p>Value should be set in the following form:</p> | |
2461 | <p> | |
2462 | <code>other_config:pmd-rxq-affinity=<rxq-affinity-list></code> | |
2463 | </p> | |
2464 | <p>where</p> | |
2465 | <p> | |
2466 | <ul> | |
2467 | <li> | |
2468 | <rxq-affinity-list> ::= NULL | <non-empty-list> | |
2469 | </li> | |
2470 | <li> | |
2471 | <non-empty-list> ::= <affinity-pair> | | |
2472 | <affinity-pair> , <non-empty-list> | |
2473 | </li> | |
2474 | <li> | |
2475 | <affinity-pair> ::= <queue-id> : <core-id> | |
2476 | </li> | |
2477 | </ul> | |
2478 | </p> | |
2479 | </column> | |
c1ff66ac CL |
2480 | |
2481 | <column name="options" key="vhost-server-path" | |
2482 | type='{"type": "string"}'> | |
2483 | <p> | |
2d24d165 CL |
2484 | The value specifies the path to the socket associated with a vHost |
2485 | User client mode device that has been or will be created by QEMU. | |
2486 | Only supported by dpdkvhostuserclient interfaces. | |
c1ff66ac CL |
2487 | </p> |
2488 | </column> | |
b685696b CL |
2489 | |
2490 | <column name="options" key="n_rxq_desc" | |
2491 | type='{"type": "integer", "minInteger": 1, "maxInteger": 4096}'> | |
2492 | <p> | |
2493 | Specifies the rx queue size (number rx descriptors) for dpdk ports. | |
2494 | The value must be a power of 2, less than 4096 and supported | |
2495 | by the hardware of the device being configured. | |
2496 | If not specified or an incorrect value is specified, 2048 rx | |
2497 | descriptors will be used by default. | |
2498 | </p> | |
2499 | </column> | |
2500 | ||
2501 | <column name="options" key="n_txq_desc" | |
2502 | type='{"type": "integer", "minInteger": 1, "maxInteger": 4096}'> | |
2503 | <p> | |
2504 | Specifies the tx queue size (number tx descriptors) for dpdk ports. | |
2505 | The value must be a power of 2, less than 4096 and supported | |
2506 | by the hardware of the device being configured. | |
2507 | If not specified or an incorrect value is specified, 2048 tx | |
2508 | descriptors will be used by default. | |
2509 | </p> | |
2510 | </column> | |
a14b8947 IM |
2511 | </group> |
2512 | ||
56abcf49 DDP |
2513 | <group title="MTU"> |
2514 | <p> | |
2515 | The MTU (maximum transmission unit) is the largest amount of data | |
2516 | that can fit into a single Ethernet frame. The standard Ethernet | |
2517 | MTU is 1500 bytes. Some physical media and many kinds of virtual | |
2518 | interfaces can be configured with higher MTUs. | |
2519 | </p> | |
2520 | ||
2521 | <p> | |
3a414a0a DDP |
2522 | A client may change an interface MTU by filling in |
2523 | <ref column="mtu_request"/>. Open vSwitch then reports in | |
2524 | <ref column="mtu"/> the currently configured value. | |
56abcf49 DDP |
2525 | </p> |
2526 | ||
2527 | <column name="mtu"> | |
3a414a0a DDP |
2528 | <p> |
2529 | The currently configured MTU for the interface. | |
2530 | </p> | |
2531 | ||
56abcf49 DDP |
2532 | <p> |
2533 | This column will be empty for an interface that does not | |
2534 | have an MTU as, for example, some kinds of tunnels do not. | |
2535 | </p> | |
2536 | ||
2537 | <p> | |
2538 | Open vSwitch sets this column's value, so other clients should treat | |
2539 | it as read-only. | |
2540 | </p> | |
2541 | </column> | |
2542 | ||
2543 | <column name="mtu_request" | |
2544 | type='{"type": "integer", "minInteger": 1}'> | |
2545 | <p> | |
2546 | Requested MTU (Maximum Transmission Unit) for the interface. A client | |
3a414a0a DDP |
2547 | can fill this column to change the MTU of an interface. |
2548 | </p> | |
2549 | ||
06ca9597 | 2550 | <p> |
2551 | RFC 791 requires every internet module to be able to forward a | |
2552 | datagram of 68 octets without further fragmentation. The maximum | |
2553 | size of an IP packet is 65535 bytes. | |
2554 | </p> | |
2555 | ||
3a414a0a DDP |
2556 | <p> |
2557 | If this is not set and if the interface has <code>internal</code> | |
2558 | type, Open vSwitch will change the MTU to match the minimum of the | |
2559 | other interfaces in the bridge. | |
56abcf49 DDP |
2560 | </p> |
2561 | </column> | |
2562 | ||
2563 | </group> | |
2564 | ||
e210037e AE |
2565 | <group title="Interface Status"> |
2566 | <p> | |
2567 | Status information about interfaces attached to bridges, updated every | |
2568 | 5 seconds. Not all interfaces have all of these properties; virtual | |
2569 | interfaces don't have a link speed, for example. Non-applicable | |
2570 | columns will have empty values. | |
2571 | </p> | |
2572 | <column name="admin_state"> | |
2573 | <p> | |
2574 | The administrative state of the physical network link. | |
2575 | </p> | |
2576 | </column> | |
2577 | ||
2578 | <column name="link_state"> | |
2579 | <p> | |
0b8024eb BP |
2580 | The observed state of the physical network link. This is ordinarily |
2581 | the link's carrier status. If the interface's <ref table="Port"/> is | |
2582 | a bond configured for miimon monitoring, it is instead the network | |
2583 | link's miimon status. | |
e210037e AE |
2584 | </p> |
2585 | </column> | |
2586 | ||
65c3058c EJ |
2587 | <column name="link_resets"> |
2588 | <p> | |
2589 | The number of times Open vSwitch has observed the | |
2590 | <ref column="link_state"/> of this <ref table="Interface"/> change. | |
2591 | </p> | |
2592 | </column> | |
2593 | ||
e210037e AE |
2594 | <column name="link_speed"> |
2595 | <p> | |
2596 | The negotiated speed of the physical network link. | |
2597 | Valid values are positive integers greater than 0. | |
2598 | </p> | |
2599 | </column> | |
2600 | ||
2601 | <column name="duplex"> | |
2602 | <p> | |
2603 | The duplex mode of the physical network link. | |
2604 | </p> | |
2605 | </column> | |
2606 | ||
3fd8d445 BP |
2607 | <column name="lacp_current"> |
2608 | Boolean value indicating LACP status for this interface. If true, this | |
2609 | interface has current LACP information about its LACP partner. This | |
2610 | information may be used to monitor the health of interfaces in a LACP | |
2611 | enabled port. This column will be empty if LACP is not enabled. | |
2612 | </column> | |
2613 | ||
573c1db9 | 2614 | <column name="status"> |
3fd8d445 BP |
2615 | Key-value pairs that report port status. Supported status values are |
2616 | <ref column="type"/>-dependent; some interfaces may not have a valid | |
2617 | <ref column="status" key="driver_name"/>, for example. | |
2618 | </column> | |
2619 | ||
2620 | <column name="status" key="driver_name"> | |
2621 | The name of the device driver controlling the network adapter. | |
2622 | </column> | |
2623 | ||
2624 | <column name="status" key="driver_version"> | |
2625 | The version string of the device driver controlling the network | |
2626 | adapter. | |
2627 | </column> | |
2628 | ||
2629 | <column name="status" key="firmware_version"> | |
2630 | The version string of the network adapter's firmware, if available. | |
2631 | </column> | |
2632 | ||
2633 | <column name="status" key="source_ip"> | |
80c4589a | 2634 | The source IP address used for an IPv4/IPv6 tunnel end-point, such as |
09538fdc | 2635 | <code>gre</code>. |
573c1db9 | 2636 | </column> |
3fd8d445 BP |
2637 | |
2638 | <column name="status" key="tunnel_egress_iface"> | |
271e6bc7 JG |
2639 | Egress interface for tunnels. Currently only relevant for tunnels |
2640 | on Linux systems, this column will show the name of the interface | |
09538fdc PS |
2641 | which is responsible for routing traffic destined for the configured |
2642 | <ref column="options" key="remote_ip"/>. This could be an internal | |
2643 | interface such as a bridge port. | |
3fd8d445 BP |
2644 | </column> |
2645 | ||
f9e5e5b3 BP |
2646 | <column name="status" key="tunnel_egress_iface_carrier" |
2647 | type='{"type": "string", "enum": ["set", ["down", "up"]]}'> | |
2648 | Whether carrier is detected on <ref column="status" | |
2649 | key="tunnel_egress_iface"/>. | |
3fd8d445 BP |
2650 | </column> |
2651 | </group> | |
2652 | ||
2653 | <group title="Statistics"> | |
2654 | <p> | |
2655 | Key-value pairs that report interface statistics. The current | |
12eb035b AW |
2656 | implementation updates these counters periodically. The update period |
2657 | is controlled by <ref column="other_config" | |
2658 | key="stats-update-interval"/> in the <code>Open_vSwitch</code> table. | |
2659 | Future implementations may update them when an interface is created, | |
2660 | when they are queried (e.g. using an OVSDB <code>select</code> | |
2661 | operation), and just before an interface is deleted due to virtual | |
2662 | interface hot-unplug or VM shutdown, and perhaps at other times, but | |
2663 | not on any regular periodic basis. | |
3fd8d445 BP |
2664 | </p> |
2665 | <p> | |
2666 | These are the same statistics reported by OpenFlow in its <code>struct | |
2667 | ofp_port_stats</code> structure. If an interface does not support a | |
2668 | given statistic, then that pair is omitted. | |
2669 | </p> | |
2670 | <group title="Statistics: Successful transmit and receive counters"> | |
2671 | <column name="statistics" key="rx_packets"> | |
2672 | Number of received packets. | |
2673 | </column> | |
2674 | <column name="statistics" key="rx_bytes"> | |
2675 | Number of received bytes. | |
2676 | </column> | |
2677 | <column name="statistics" key="tx_packets"> | |
2678 | Number of transmitted packets. | |
2679 | </column> | |
2680 | <column name="statistics" key="tx_bytes"> | |
2681 | Number of transmitted bytes. | |
2682 | </column> | |
2683 | </group> | |
2684 | <group title="Statistics: Receive errors"> | |
2685 | <column name="statistics" key="rx_dropped"> | |
2686 | Number of packets dropped by RX. | |
2687 | </column> | |
2688 | <column name="statistics" key="rx_frame_err"> | |
2689 | Number of frame alignment errors. | |
2690 | </column> | |
2691 | <column name="statistics" key="rx_over_err"> | |
2692 | Number of packets with RX overrun. | |
2693 | </column> | |
2694 | <column name="statistics" key="rx_crc_err"> | |
2695 | Number of CRC errors. | |
2696 | </column> | |
2697 | <column name="statistics" key="rx_errors"> | |
2698 | Total number of receive errors, greater than or equal to the sum of | |
2699 | the above. | |
2700 | </column> | |
9cc6bf75 | 2701 | </group> |
3fd8d445 BP |
2702 | <group title="Statistics: Transmit errors"> |
2703 | <column name="statistics" key="tx_dropped"> | |
2704 | Number of packets dropped by TX. | |
2705 | </column> | |
2706 | <column name="statistics" key="collisions"> | |
2707 | Number of collisions. | |
2708 | </column> | |
2709 | <column name="statistics" key="tx_errors"> | |
2710 | Total number of transmit errors, greater than or equal to the sum of | |
2711 | the above. | |
2712 | </column> | |
2713 | </group> | |
89365653 BP |
2714 | </group> |
2715 | ||
2716 | <group title="Ingress Policing"> | |
3f5d8c02 BP |
2717 | <p> |
2718 | These settings control ingress policing for packets received on this | |
2719 | interface. On a physical interface, this limits the rate at which | |
2720 | traffic is allowed into the system from the outside; on a virtual | |
2721 | interface (one connected to a virtual machine), this limits the rate at | |
2722 | which the VM is able to transmit. | |
2723 | </p> | |
2724 | <p> | |
2725 | Policing is a simple form of quality-of-service that simply drops | |
2726 | packets received in excess of the configured rate. Due to its | |
2727 | simplicity, policing is usually less accurate and less effective than | |
2728 | egress QoS (which is configured using the <ref table="QoS"/> and <ref | |
2729 | table="Queue"/> tables). | |
2730 | </p> | |
2731 | <p> | |
9509913a IS |
2732 | Policing is currently implemented on Linux and OVS with DPDK. Both |
2733 | implementations use a simple ``token bucket'' approach: | |
3f5d8c02 BP |
2734 | </p> |
2735 | <ul> | |
2736 | <li> | |
2737 | The size of the bucket corresponds to <ref | |
2738 | column="ingress_policing_burst"/>. Initially the bucket is full. | |
2739 | </li> | |
2740 | <li> | |
2741 | Whenever a packet is received, its size (converted to tokens) is | |
2742 | compared to the number of tokens currently in the bucket. If the | |
2743 | required number of tokens are available, they are removed and the | |
2744 | packet is forwarded. Otherwise, the packet is dropped. | |
2745 | </li> | |
2746 | <li> | |
2747 | Whenever it is not full, the bucket is refilled with tokens at the | |
2748 | rate specified by <ref column="ingress_policing_rate"/>. | |
2749 | </li> | |
2750 | </ul> | |
2751 | <p> | |
2752 | Policing interacts badly with some network protocols, and especially | |
2753 | with fragmented IP packets. Suppose that there is enough network | |
2754 | activity to keep the bucket nearly empty all the time. Then this token | |
2755 | bucket algorithm will forward a single packet every so often, with the | |
2756 | period depending on packet size and on the configured rate. All of the | |
2757 | fragments of an IP packets are normally transmitted back-to-back, as a | |
2758 | group. In such a situation, therefore, only one of these fragments | |
2759 | will be forwarded and the rest will be dropped. IP does not provide | |
2760 | any way for the intended recipient to ask for only the remaining | |
2761 | fragments. In such a case there are two likely possibilities for what | |
2762 | will happen next: either all of the fragments will eventually be | |
2763 | retransmitted (as TCP will do), in which case the same problem will | |
2764 | recur, or the sender will not realize that its packet has been dropped | |
2765 | and data will simply be lost (as some UDP-based protocols will do). | |
2766 | Either way, it is possible that no forward progress will ever occur. | |
2767 | </p> | |
2768 | <column name="ingress_policing_rate"> | |
2769 | <p> | |
2770 | Maximum rate for data received on this interface, in kbps. Data | |
2771 | received faster than this rate is dropped. Set to <code>0</code> | |
2772 | (the default) to disable policing. | |
2773 | </p> | |
2774 | </column> | |
2775 | ||
89365653 BP |
2776 | <column name="ingress_policing_burst"> |
2777 | <p>Maximum burst size for data received on this interface, in kb. The | |
79abacc8 | 2778 | default burst size if set to <code>0</code> is 8000 kbit. This value |
3fd8d445 BP |
2779 | has no effect if <ref column="ingress_policing_rate"/> |
2780 | is <code>0</code>.</p> | |
3f5d8c02 BP |
2781 | <p> |
2782 | Specifying a larger burst size lets the algorithm be more forgiving, | |
2783 | which is important for protocols like TCP that react severely to | |
2784 | dropped packets. The burst size should be at least the size of the | |
2785 | interface's MTU. Specifying a value that is numerically at least as | |
79abacc8 | 2786 | large as 80% of <ref column="ingress_policing_rate"/> helps TCP come |
3f5d8c02 BP |
2787 | closer to achieving the full rate. |
2788 | </p> | |
89365653 BP |
2789 | </column> |
2790 | </group> | |
2791 | ||
ccc09689 | 2792 | <group title="Bidirectional Forwarding Detection (BFD)"> |
e58855ec | 2793 | <p> |
039a8ccd BP |
2794 | BFD, defined in RFC 5880 and RFC 5881, allows point-to-point |
2795 | detection of connectivity failures by occasional transmission of | |
2796 | BFD control messages. Open vSwitch implements BFD to serve | |
2797 | as a more popular and standards compliant alternative to CFM. | |
e58855ec | 2798 | </p> |
ccc09689 | 2799 | |
e58855ec | 2800 | <p> |
039a8ccd BP |
2801 | BFD operates by regularly transmitting BFD control messages at a rate |
2802 | negotiated independently in each direction. Each endpoint specifies | |
2803 | the rate at which it expects to receive control messages, and the rate | |
2804 | at which it is willing to transmit them. Open vSwitch uses a detection | |
2805 | multiplier of three, meaning that an endpoint signals a connectivity | |
2806 | fault if three consecutive BFD control messages fail to arrive. In the | |
2807 | case of a unidirectional connectivity issue, the system not receiving | |
2808 | BFD control messages signals the problem to its peer in the messages it | |
2809 | transmits. | |
e58855ec | 2810 | </p> |
ccc09689 | 2811 | |
e58855ec | 2812 | <p> |
039a8ccd BP |
2813 | The Open vSwitch implementation of BFD aims to comply faithfully |
2814 | with RFC 5880 requirements. Open vSwitch does not implement the | |
2815 | optional Authentication or ``Echo Mode'' features. | |
e58855ec | 2816 | </p> |
ccc09689 | 2817 | |
e58855ec | 2818 | <group title="BFD Configuration"> |
039a8ccd BP |
2819 | <p> |
2820 | A controller sets up key-value pairs in the <ref column="bfd"/> | |
2821 | column to enable and configure BFD. | |
2822 | </p> | |
e58855ec | 2823 | |
039a8ccd | 2824 | <column name="bfd" key="enable" type='{"type": "boolean"}'> |
f7491dce AW |
2825 | True to enable BFD on this <ref table="Interface"/>. If not |
2826 | specified, BFD will not be enabled by default. | |
039a8ccd | 2827 | </column> |
e58855ec | 2828 | |
039a8ccd BP |
2829 | <column name="bfd" key="min_rx" |
2830 | type='{"type": "integer", "minInteger": 1}'> | |
e58855ec BP |
2831 | The shortest interval, in milliseconds, at which this BFD session |
2832 | offers to receive BFD control messages. The remote endpoint may | |
2833 | choose to send messages at a slower rate. Defaults to | |
2834 | <code>1000</code>. | |
039a8ccd | 2835 | </column> |
e58855ec | 2836 | |
039a8ccd BP |
2837 | <column name="bfd" key="min_tx" |
2838 | type='{"type": "integer", "minInteger": 1}'> | |
e58855ec BP |
2839 | The shortest interval, in milliseconds, at which this BFD session is |
2840 | willing to transmit BFD control messages. Messages will actually be | |
2841 | transmitted at a slower rate if the remote endpoint is not willing to | |
2842 | receive as quickly as specified. Defaults to <code>100</code>. | |
039a8ccd BP |
2843 | </column> |
2844 | ||
2845 | <column name="bfd" key="decay_min_rx" type='{"type": "integer"}'> | |
2846 | An alternate receive interval, in milliseconds, that must be greater | |
2847 | than or equal to <ref column="bfd" key="min_rx"/>. The | |
2848 | implementation switches from <ref column="bfd" key="min_rx"/> to <ref | |
2849 | column="bfd" key="decay_min_rx"/> when there is no obvious incoming | |
2850 | data traffic at the interface, to reduce the CPU and bandwidth cost | |
2851 | of monitoring an idle interface. This feature may be disabled by | |
2852 | setting a value of 0. This feature is reset whenever <ref | |
2853 | column="bfd" key="decay_min_rx"/> or <ref column="bfd" key="min_rx"/> | |
2854 | changes. | |
2855 | </column> | |
2856 | ||
2857 | <column name="bfd" key="forwarding_if_rx" type='{"type": "boolean"}'> | |
34c88624 AW |
2858 | When <code>true</code>, traffic received on the |
2859 | <ref table="Interface"/> is used to indicate the capability of packet | |
2860 | I/O. BFD control packets are still transmitted and received. At | |
2861 | least one BFD control packet must be received every 100 * <ref | |
2862 | column="bfd" key="min_rx"/> amount of time. Otherwise, even if | |
2863 | traffic are received, the <ref column="bfd" key="forwarding"/> | |
2864 | will be <code>false</code>. | |
039a8ccd | 2865 | </column> |
e58855ec | 2866 | |
039a8ccd BP |
2867 | <column name="bfd" key="cpath_down" type='{"type": "boolean"}'> |
2868 | Set to true to notify the remote endpoint that traffic should not be | |
2869 | forwarded to this system for some reason other than a connectivty | |
2870 | failure on the interface being monitored. The typical underlying | |
2871 | reason is ``concatenated path down,'' that is, that connectivity | |
2872 | beyond the local system is down. Defaults to false. | |
2873 | </column> | |
e58855ec | 2874 | |
039a8ccd | 2875 | <column name="bfd" key="check_tnl_key" type='{"type": "boolean"}'> |
e58855ec BP |
2876 | Set to true to make BFD accept only control messages with a tunnel |
2877 | key of zero. By default, BFD accepts control messages with any | |
2878 | tunnel key. | |
039a8ccd BP |
2879 | </column> |
2880 | ||
2881 | <column name="bfd" key="bfd_local_src_mac"> | |
2882 | Set to an Ethernet address in the form | |
2883 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var> | |
2884 | to set the MAC used as source for transmitted BFD packets. The | |
2885 | default is the mac address of the BFD enabled interface. | |
2886 | </column> | |
2887 | ||
2888 | <column name="bfd" key="bfd_local_dst_mac"> | |
2889 | Set to an Ethernet address in the form | |
2890 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var> | |
2891 | to set the MAC used as destination for transmitted BFD packets. The | |
2892 | default is <code>00:23:20:00:00:01</code>. | |
2893 | </column> | |
2894 | ||
2895 | <column name="bfd" key="bfd_remote_dst_mac"> | |
2896 | Set to an Ethernet address in the form | |
2897 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var> | |
2898 | to set the MAC used for checking the destination of received BFD packets. | |
2899 | Packets with different destination MAC will not be considered as BFD packets. | |
2900 | If not specified the destination MAC address of received BFD packets | |
2901 | are not checked. | |
2902 | </column> | |
2903 | ||
2904 | <column name="bfd" key="bfd_src_ip"> | |
dfe37e6a | 2905 | Set to an IPv4 address to set the IP address used as source for |
1314739c | 2906 | transmitted BFD packets. The default is <code>169.254.1.1</code>. |
039a8ccd | 2907 | </column> |
dfe37e6a | 2908 | |
039a8ccd | 2909 | <column name="bfd" key="bfd_dst_ip"> |
dfe37e6a | 2910 | Set to an IPv4 address to set the IP address used as destination |
1314739c | 2911 | for transmitted BFD packets. The default is <code>169.254.1.0</code>. |
039a8ccd | 2912 | </column> |
2eb79142 JG |
2913 | |
2914 | <column name="bfd" key="oam"> | |
2915 | Some tunnel protocols (such as Geneve) include a bit in the header | |
2916 | to indicate that the encapsulated packet is an OAM frame. By setting | |
2917 | this to true, BFD packets will be marked as OAM if encapsulated in | |
2918 | one of these tunnels. | |
2919 | </column> | |
e58855ec | 2920 | </group> |
ccc09689 | 2921 | |
e58855ec | 2922 | <group title="BFD Status"> |
039a8ccd BP |
2923 | <p> |
2924 | The switch sets key-value pairs in the <ref column="bfd_status"/> | |
2925 | column to report the status of BFD on this interface. When BFD is | |
2926 | not enabled, with <ref column="bfd" key="enable"/>, the switch clears | |
2927 | all key-value pairs from <ref column="bfd_status"/>. | |
2928 | </p> | |
2929 | ||
2930 | <column name="bfd_status" key="state" | |
2931 | type='{"type": "string", | |
2932 | "enum": ["set", ["admin_down", "down", "init", "up"]]}'> | |
2933 | Reports the state of the BFD session. The BFD session is fully | |
2934 | healthy and negotiated if <code>UP</code>. | |
2935 | </column> | |
2936 | ||
2937 | <column name="bfd_status" key="forwarding" type='{"type": "boolean"}'> | |
2938 | Reports whether the BFD session believes this <ref | |
2939 | table="Interface"/> may be used to forward traffic. Typically this | |
2940 | means the local session is signaling <code>UP</code>, and the remote | |
2941 | system isn't signaling a problem such as concatenated path down. | |
2942 | </column> | |
2943 | ||
2944 | <column name="bfd_status" key="diagnostic"> | |
60a15922 AZ |
2945 | A diagnostic code specifying the local system's reason for the |
2946 | last change in session state. The error messages are defined in | |
2947 | section 4.1 of [RFC 5880]. | |
039a8ccd BP |
2948 | </column> |
2949 | ||
2950 | <column name="bfd_status" key="remote_state" | |
2951 | type='{"type": "string", | |
2952 | "enum": ["set", ["admin_down", "down", "init", "up"]]}'> | |
2953 | Reports the state of the remote endpoint's BFD session. | |
2954 | </column> | |
2955 | ||
2956 | <column name="bfd_status" key="remote_diagnostic"> | |
60a15922 AZ |
2957 | A diagnostic code specifying the remote system's reason for the |
2958 | last change in session state. The error messages are defined in | |
2959 | section 4.1 of [RFC 5880]. | |
039a8ccd | 2960 | </column> |
4905e2df AW |
2961 | |
2962 | <column name="bfd_status" key="flap_count" | |
039a8ccd | 2963 | type='{"type": "integer", "minInteger": 0}'> |
4905e2df AW |
2964 | Counts the number of <ref column="bfd_status" key="forwarding" /> |
2965 | flaps since start. A flap is considered as a change of the | |
2966 | <ref column="bfd_status" key="forwarding" /> value. | |
2967 | </column> | |
e58855ec | 2968 | </group> |
ccc09689 EJ |
2969 | </group> |
2970 | ||
93b8df38 EJ |
2971 | <group title="Connectivity Fault Management"> |
2972 | <p> | |
2973 | 802.1ag Connectivity Fault Management (CFM) allows a group of | |
2974 | Maintenance Points (MPs) called a Maintenance Association (MA) to | |
2975 | detect connectivity problems with each other. MPs within a MA should | |
2976 | have complete and exclusive interconnectivity. This is verified by | |
2977 | occasionally broadcasting Continuity Check Messages (CCMs) at a | |
2978 | configurable transmission interval. | |
2979 | </p> | |
2980 | ||
144216a3 EJ |
2981 | <p> |
2982 | According to the 802.1ag specification, each Maintenance Point should | |
2983 | be configured out-of-band with a list of Remote Maintenance Points it | |
2984 | should have connectivity to. Open vSwitch differs from the | |
2985 | specification in this area. It simply assumes the link is faulted if | |
2986 | no Remote Maintenance Points are reachable, and considers it not | |
2987 | faulted otherwise. | |
2988 | </p> | |
2989 | ||
b363bae4 | 2990 | <p> |
039a8ccd BP |
2991 | When operating over tunnels which have no <code>in_key</code>, or an |
2992 | <code>in_key</code> of <code>flow</code>. CFM will only accept CCMs | |
2993 | with a tunnel key of zero. | |
b363bae4 EJ |
2994 | </p> |
2995 | ||
93b8df38 | 2996 | <column name="cfm_mpid"> |
b1a6083a AW |
2997 | <p> |
2998 | A Maintenance Point ID (MPID) uniquely identifies each endpoint | |
2999 | within a Maintenance Association. The MPID is used to identify this | |
3000 | endpoint to other Maintenance Points in the MA. Each end of a link | |
3001 | being monitored should have a different MPID. Must be configured to | |
3002 | enable CFM on this <ref table="Interface"/>. | |
3003 | </p> | |
3004 | <p> | |
3005 | According to the 802.1ag specification, MPIDs can only range between | |
3006 | [1, 8191]. However, extended mode (see <ref column="other_config" | |
3007 | key="cfm_extended"/>) supports eight byte MPIDs. | |
3008 | </p> | |
93b8df38 | 3009 | </column> |
b31bcf60 | 3010 | |
76c4290d AW |
3011 | <column name="cfm_flap_count"> |
3012 | Counts the number of cfm fault flapps since boot. A flap is | |
3013 | considered to be a change of the <ref column="cfm_fault"/> value. | |
3014 | </column> | |
3015 | ||
93b8df38 | 3016 | <column name="cfm_fault"> |
144216a3 EJ |
3017 | <p> |
3018 | Indicates a connectivity fault triggered by an inability to receive | |
3019 | heartbeats from any remote endpoint. When a fault is triggered on | |
3020 | <ref table="Interface"/>s participating in bonds, they will be | |
3021 | disabled. | |
3022 | </p> | |
3023 | <p> | |
3024 | Faults can be triggered for several reasons. Most importantly they | |
3025 | are triggered when no CCMs are received for a period of 3.5 times the | |
3026 | transmission interval. Faults are also triggered when any CCMs | |
3027 | indicate that a Remote Maintenance Point is not receiving CCMs but | |
3028 | able to send them. Finally, a fault is triggered if a CCM is | |
3029 | received which indicates unexpected configuration. Notably, this | |
3030 | case arises when a CCM is received which advertises the local MPID. | |
3031 | </p> | |
93b8df38 | 3032 | </column> |
a5faa982 | 3033 | |
b9380396 EJ |
3034 | <column name="cfm_fault_status" key="recv"> |
3035 | Indicates a CFM fault was triggered due to a lack of CCMs received on | |
3036 | the <ref table="Interface"/>. | |
3037 | </column> | |
3038 | ||
3039 | <column name="cfm_fault_status" key="rdi"> | |
3040 | Indicates a CFM fault was triggered due to the reception of a CCM with | |
3041 | the RDI bit flagged. Endpoints set the RDI bit in their CCMs when they | |
3042 | are not receiving CCMs themselves. This typically indicates a | |
3043 | unidirectional connectivity failure. | |
3044 | </column> | |
3045 | ||
3046 | <column name="cfm_fault_status" key="maid"> | |
3047 | Indicates a CFM fault was triggered due to the reception of a CCM with | |
3048 | a MAID other than the one Open vSwitch uses. CFM broadcasts are tagged | |
3049 | with an identification number in addition to the MPID called the MAID. | |
3050 | Open vSwitch only supports receiving CCM broadcasts tagged with the | |
3051 | MAID it uses internally. | |
3052 | </column> | |
3053 | ||
3054 | <column name="cfm_fault_status" key="loopback"> | |
3055 | Indicates a CFM fault was triggered due to the reception of a CCM | |
3056 | advertising the same MPID configured in the <ref column="cfm_mpid"/> | |
3057 | column of this <ref table="Interface"/>. This may indicate a loop in | |
3058 | the network. | |
3059 | </column> | |
3060 | ||
3061 | <column name="cfm_fault_status" key="overflow"> | |
3062 | Indicates a CFM fault was triggered because the CFM module received | |
3063 | CCMs from more remote endpoints than it can keep track of. | |
3064 | </column> | |
3065 | ||
3066 | <column name="cfm_fault_status" key="override"> | |
3067 | Indicates a CFM fault was manually triggered by an administrator using | |
3068 | an <code>ovs-appctl</code> command. | |
3069 | </column> | |
3070 | ||
2b540ecb MM |
3071 | <column name="cfm_fault_status" key="interval"> |
3072 | Indicates a CFM fault was triggered due to the reception of a CCM | |
3073 | frame having an invalid interval. | |
3074 | </column> | |
3075 | ||
1c0333b6 EJ |
3076 | <column name="cfm_remote_opstate"> |
3077 | <p>When in extended mode, indicates the operational state of the | |
039a8ccd BP |
3078 | remote endpoint as either <code>up</code> or <code>down</code>. See |
3079 | <ref column="other_config" key="cfm_opstate"/>. | |
1c0333b6 EJ |
3080 | </p> |
3081 | </column> | |
3082 | ||
3967a833 MM |
3083 | <column name="cfm_health"> |
3084 | <p> | |
3085 | Indicates the health of the interface as a percentage of CCM frames | |
3086 | received over 21 <ref column="other_config" key="cfm_interval"/>s. | |
3087 | The health of an interface is undefined if it is communicating with | |
3088 | more than one <ref column="cfm_remote_mpids"/>. It reduces if | |
3089 | healthy heartbeats are not received at the expected rate, and | |
3090 | gradually improves as healthy heartbeats are received at the desired | |
3091 | rate. Every 21 <ref column="other_config" key="cfm_interval"/>s, the | |
3092 | health of the interface is refreshed. | |
3093 | </p> | |
3094 | <p> | |
3095 | As mentioned above, the faults can be triggered for several reasons. | |
3096 | The link health will deteriorate even if heartbeats are received but | |
3097 | they are reported to be unhealthy. An unhealthy heartbeat in this | |
3098 | context is a heartbeat for which either some fault is set or is out | |
3099 | of sequence. The interface health can be 100 only on receiving | |
3100 | healthy heartbeats at the desired rate. | |
3101 | </p> | |
3102 | </column> | |
3103 | ||
a5faa982 EJ |
3104 | <column name="cfm_remote_mpids"> |
3105 | When CFM is properly configured, Open vSwitch will occasionally | |
3106 | receive CCM broadcasts. These broadcasts contain the MPID of the | |
3107 | sending Maintenance Point. The list of MPIDs from which this | |
3108 | <ref table="Interface"/> is receiving broadcasts from is regularly | |
3109 | collected and written to this column. | |
3110 | </column> | |
3fd8d445 | 3111 | |
f9e5e5b3 BP |
3112 | <column name="other_config" key="cfm_interval" |
3113 | type='{"type": "integer"}'> | |
612ca9c5 BP |
3114 | <p> |
3115 | The interval, in milliseconds, between transmissions of CFM | |
3116 | heartbeats. Three missed heartbeat receptions indicate a | |
3117 | connectivity fault. | |
3118 | </p> | |
3119 | ||
3120 | <p> | |
3121 | In standard operation only intervals of 3, 10, 100, 1,000, 10,000, | |
3122 | 60,000, or 600,000 ms are supported. Other values will be rounded | |
3123 | down to the nearest value on the list. Extended mode (see <ref | |
3124 | column="other_config" key="cfm_extended"/>) supports any interval up | |
3125 | to 65,535 ms. In either mode, the default is 1000 ms. | |
3126 | </p> | |
3127 | ||
3128 | <p>We do not recommend using intervals less than 100 ms.</p> | |
3fd8d445 BP |
3129 | </column> |
3130 | ||
f9e5e5b3 BP |
3131 | <column name="other_config" key="cfm_extended" |
3132 | type='{"type": "boolean"}'> | |
3fd8d445 BP |
3133 | When <code>true</code>, the CFM module operates in extended mode. This |
3134 | causes it to use a nonstandard destination address to avoid conflicting | |
3135 | with compliant implementations which may be running concurrently on the | |
3136 | network. Furthermore, extended mode increases the accuracy of the | |
3137 | <code>cfm_interval</code> configuration parameter by breaking wire | |
b1a6083a AW |
3138 | compatibility with 802.1ag compliant implementations. And extended |
3139 | mode allows eight byte MPIDs. Defaults to <code>false</code>. | |
3fd8d445 | 3140 | </column> |
90967e95 EJ |
3141 | |
3142 | <column name="other_config" key="cfm_demand" type='{"type": "boolean"}'> | |
3143 | <p> | |
3144 | When <code>true</code>, and | |
3145 | <ref column="other_config" key="cfm_extended"/> is true, the CFM | |
3146 | module operates in demand mode. When in demand mode, traffic | |
3147 | received on the <ref table="Interface"/> is used to indicate | |
5767a79a AW |
3148 | liveness. CCMs are still transmitted and received. At least one |
3149 | CCM must be received every 100 * <ref column="other_config" | |
3150 | key="cfm_interval"/> amount of time. Otherwise, even if traffic | |
3151 | are received, the CFM module will raise the connectivity fault. | |
90967e95 EJ |
3152 | </p> |
3153 | ||
3154 | <p> | |
039a8ccd | 3155 | Demand mode has a couple of caveats: |
90967e95 EJ |
3156 | <ul> |
3157 | <li> | |
3158 | To ensure that ovs-vswitchd has enough time to pull statistics | |
03f209ba AW |
3159 | from the datapath, the fault detection interval is set to |
3160 | 3.5 * MAX(<ref column="other_config" key="cfm_interval"/>, 500) | |
3161 | ms. | |
90967e95 EJ |
3162 | </li> |
3163 | ||
3164 | <li> | |
3165 | To avoid ambiguity, demand mode disables itself when there are | |
3166 | multiple remote maintenance points. | |
3167 | </li> | |
3168 | ||
3169 | <li> | |
3170 | If the <ref table="Interface"/> is heavily congested, CCMs | |
3171 | containing the <ref column="other_config" key="cfm_opstate"/> | |
3172 | status may be dropped causing changes in the operational state to | |
3173 | be delayed. Similarly, if CCMs containing the RDI bit are not | |
3174 | received, unidirectional link failures may not be detected. | |
3175 | </li> | |
3176 | </ul> | |
3177 | </p> | |
3178 | </column> | |
3179 | ||
dae57238 BP |
3180 | <column name="other_config" key="cfm_opstate" |
3181 | type='{"type": "string", "enum": ["set", ["down", "up"]]}'> | |
86dc6501 EJ |
3182 | When <code>down</code>, the CFM module marks all CCMs it generates as |
3183 | operationally down without triggering a fault. This allows remote | |
3184 | maintenance points to choose not to forward traffic to the | |
3185 | <ref table="Interface"/> on which this CFM module is running. | |
3186 | Currently, in Open vSwitch, the opdown bit of CCMs affects | |
3187 | <ref table="Interface"/>s participating in bonds, and the bundle | |
3188 | OpenFlow action. This setting is ignored when CFM is not in extended | |
3189 | mode. Defaults to <code>up</code>. | |
3190 | </column> | |
75a4ead1 EJ |
3191 | |
3192 | <column name="other_config" key="cfm_ccm_vlan" | |
039a8ccd | 3193 | type='{"type": "integer", "minInteger": 1, "maxInteger": 4095}'> |
75a4ead1 | 3194 | When set, the CFM module will apply a VLAN tag to all CCMs it generates |
189cb9e4 EJ |
3195 | with the given value. May be the string <code>random</code> in which |
3196 | case each CCM will be tagged with a different randomly generated VLAN. | |
75a4ead1 EJ |
3197 | </column> |
3198 | ||
a7aa2d3c | 3199 | <column name="other_config" key="cfm_ccm_pcp" |
039a8ccd | 3200 | type='{"type": "integer", "minInteger": 1, "maxInteger": 7}'> |
a7aa2d3c | 3201 | When set, the CFM module will apply a VLAN tag to all CCMs it generates |
b363bae4 | 3202 | with the given PCP value, the VLAN ID of the tag is governed by the |
a7aa2d3c EJ |
3203 | value of <ref column="other_config" key="cfm_ccm_vlan"/>. If |
3204 | <ref column="other_config" key="cfm_ccm_vlan"/> is unset, a VLAN ID of | |
3205 | zero is used. | |
3206 | </column> | |
3207 | ||
93b8df38 EJ |
3208 | </group> |
3209 | ||
3fd8d445 | 3210 | <group title="Bonding Configuration"> |
f9e5e5b3 BP |
3211 | <column name="other_config" key="lacp-port-id" |
3212 | type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'> | |
3fd8d445 BP |
3213 | The LACP port ID of this <ref table="Interface"/>. Port IDs are |
3214 | used in LACP negotiations to identify individual ports | |
f9e5e5b3 | 3215 | participating in a bond. |
a8172aa3 EJ |
3216 | </column> |
3217 | ||
f9e5e5b3 BP |
3218 | <column name="other_config" key="lacp-port-priority" |
3219 | type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'> | |
3fd8d445 BP |
3220 | The LACP port priority of this <ref table="Interface"/>. In LACP |
3221 | negotiations <ref table="Interface"/>s with numerically lower | |
f9e5e5b3 | 3222 | priorities are preferred for aggregation. |
89365653 | 3223 | </column> |
018f1525 | 3224 | |
f9e5e5b3 BP |
3225 | <column name="other_config" key="lacp-aggregation-key" |
3226 | type='{"type": "integer", "minInteger": 1, "maxInteger": 65535}'> | |
3fd8d445 BP |
3227 | The LACP aggregation key of this <ref table="Interface"/>. <ref |
3228 | table="Interface"/>s with different aggregation keys may not be active | |
f9e5e5b3 | 3229 | within a given <ref table="Port"/> at the same time. |
a3acf0b0 | 3230 | </column> |
3fd8d445 | 3231 | </group> |
a3acf0b0 | 3232 | |
3fd8d445 BP |
3233 | <group title="Virtual Machine Identifiers"> |
3234 | <p> | |
3235 | These key-value pairs specifically apply to an interface that | |
3236 | represents a virtual Ethernet interface connected to a virtual | |
3237 | machine. These key-value pairs should not be present for other types | |
3238 | of interfaces. Keys whose names end in <code>-uuid</code> have | |
3239 | values that uniquely identify the entity in question. For a Citrix | |
3240 | XenServer hypervisor, these values are UUIDs in RFC 4122 format. | |
3241 | Other hypervisors may use other formats. | |
3242 | </p> | |
3243 | ||
3244 | <column name="external_ids" key="attached-mac"> | |
3245 | The MAC address programmed into the ``virtual hardware'' for this | |
3246 | interface, in the form | |
3247 | <var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>:<var>xx</var>. | |
3248 | For Citrix XenServer, this is the value of the <code>MAC</code> field | |
3249 | in the VIF record for this interface. | |
3250 | </column> | |
3251 | ||
3252 | <column name="external_ids" key="iface-id"> | |
3253 | A system-unique identifier for the interface. On XenServer, this will | |
3254 | commonly be the same as <ref column="external_ids" key="xs-vif-uuid"/>. | |
3255 | </column> | |
3256 | ||
cf9deac5 BP |
3257 | <column name="external_ids" key="iface-status" |
3258 | type='{"type": "string", | |
3259 | "enum": ["set", ["active", "inactive"]]}'> | |
3260 | <p> | |
3261 | Hypervisors may sometimes have more than one interface associated | |
3262 | with a given <ref column="external_ids" key="iface-id"/>, only one of | |
3263 | which is actually in use at a given time. For example, in some | |
3264 | circumstances XenServer has both a ``tap'' and a ``vif'' interface | |
3265 | for a single <ref column="external_ids" key="iface-id"/>, but only | |
3266 | uses one of them at a time. A hypervisor that behaves this way must | |
3267 | mark the currently in use interface <code>active</code> and the | |
3268 | others <code>inactive</code>. A hypervisor that never has more than | |
3269 | one interface for a given <ref column="external_ids" key="iface-id"/> | |
3270 | may mark that interface <code>active</code> or omit <ref | |
3271 | column="external_ids" key="iface-status"/> entirely. | |
3272 | </p> | |
3273 | ||
3274 | <p> | |
3275 | During VM migration, a given <ref column="external_ids" | |
3276 | key="iface-id"/> might transiently be marked <code>active</code> on | |
3277 | two different hypervisors. That is, <code>active</code> means that | |
3278 | this <ref column="external_ids" key="iface-id"/> is the active | |
3279 | instance within a single hypervisor, not in a broader scope. | |
3634eb99 BP |
3280 | There is one exception: some hypervisors support ``migration'' from a |
3281 | given hypervisor to itself (most often for test purposes). During | |
3282 | such a ``migration,'' two instances of a single <ref | |
3283 | column="external_ids" key="iface-id"/> might both be briefly marked | |
3284 | <code>active</code> on a single hypervisor. | |
cf9deac5 BP |
3285 | </p> |
3286 | </column> | |
3287 | ||
3fd8d445 BP |
3288 | <column name="external_ids" key="xs-vif-uuid"> |
3289 | The virtual interface associated with this interface. | |
3290 | </column> | |
3291 | ||
3292 | <column name="external_ids" key="xs-network-uuid"> | |
3293 | The virtual network to which this interface is attached. | |
3294 | </column> | |
3295 | ||
c473936b GS |
3296 | <column name="external_ids" key="vm-id"> |
3297 | The VM to which this interface belongs. On XenServer, this will be the | |
3298 | same as <ref column="external_ids" key="xs-vm-uuid"/>. | |
3299 | </column> | |
3300 | ||
3fd8d445 BP |
3301 | <column name="external_ids" key="xs-vm-uuid"> |
3302 | The VM to which this interface belongs. | |
018f1525 | 3303 | </column> |
89365653 | 3304 | </group> |
3fd8d445 | 3305 | |
99eef98b DF |
3306 | <group title="Auto Attach Configuration"> |
3307 | <p> | |
039a8ccd | 3308 | Auto Attach configuration for a particular interface. |
99eef98b DF |
3309 | </p> |
3310 | ||
3311 | <column name="lldp" key="enable" type='{"type": "boolean"}'> | |
039a8ccd BP |
3312 | True to enable LLDP on this <ref table="Interface"/>. If not |
3313 | specified, LLDP will be disabled by default. | |
99eef98b DF |
3314 | </column> |
3315 | </group> | |
3316 | ||
9fd39370 SC |
3317 | <group title="Flow control Configuration"> |
3318 | <p> | |
3319 | Ethernet flow control defined in IEEE 802.1Qbb provides link level flow | |
3320 | control using MAC pause frames. Implemented only for interfaces with | |
3321 | type <code>dpdk</code>. | |
3322 | </p> | |
3323 | ||
3324 | <column name="options" key="rx-flow-ctrl" type='{"type": "boolean"}'> | |
3325 | Set to <code>true</code> to enable Rx flow control on physical ports. | |
3326 | By default, Rx flow control is disabled. | |
3327 | </column> | |
3328 | ||
3329 | <column name="options" key="tx-flow-ctrl" type='{"type": "boolean"}'> | |
3330 | Set to <code>true</code> to enable Tx flow control on physical ports. | |
3331 | By default, Tx flow control is disabled. | |
3332 | </column> | |
3333 | ||
3334 | <column name="options" key="flow-ctrl-autoneg" | |
3335 | type='{"type": "boolean"}'> | |
3336 | Set to <code>true</code> to enable flow control auto negotiation on | |
3337 | physical ports. By default, auto-neg is disabled. | |
3338 | </column> | |
3339 | </group> | |
3340 | ||
1a2bb118 SC |
3341 | <group title="Rx Checksum Offload Configuration"> |
3342 | <p> | |
3343 | The checksum validation on the incoming packets are performed on NIC | |
3344 | using Rx checksum offload feature. Implemented only for <code>dpdk | |
3345 | </code>physical interfaces. | |
3346 | </p> | |
3347 | ||
3348 | <column name="options" key="rx-checksum-offload" | |
3349 | type='{"type": "boolean"}'> | |
3350 | Set to <code>false</code> to disble Rx checksum offloading on <code> | |
3351 | dpdk</code>physical ports. By default, Rx checksum offload is enabled. | |
3352 | </column> | |
3353 | </group> | |
3354 | ||
3fd8d445 BP |
3355 | <group title="Common Columns"> |
3356 | The overall purpose of these columns is described under <code>Common | |
3357 | Columns</code> at the beginning of this document. | |
3358 | ||
3359 | <column name="other_config"/> | |
3360 | <column name="external_ids"/> | |
3361 | </group> | |
89365653 BP |
3362 | </table> |
3363 | ||
254750ce BP |
3364 | <table name="Flow_Table" title="OpenFlow table configuration"> |
3365 | <p>Configuration for a particular OpenFlow table.</p> | |
3366 | ||
3367 | <column name="name"> | |
3368 | The table's name. Set this column to change the name that controllers | |
3369 | will receive when they request table statistics, e.g. <code>ovs-ofctl | |
3370 | dump-tables</code>. The name does not affect switch behavior. | |
3371 | </column> | |
3372 | ||
82c22d34 | 3373 | <group title="Eviction Policy"> |
254750ce | 3374 | <p> |
82c22d34 BP |
3375 | Open vSwitch supports limiting the number of flows that may be |
3376 | installed in a flow table, via the <ref column="flow_limit"/> column. | |
3377 | When adding a flow would exceed this limit, by default Open vSwitch | |
3378 | reports an error, but there are two ways to configure Open vSwitch to | |
3379 | instead delete (``evict'') a flow to make room for the new one: | |
254750ce BP |
3380 | </p> |
3381 | ||
82c22d34 BP |
3382 | <ul> |
3383 | <li> | |
3384 | Set the <ref column="overflow_policy"/> column to <code>evict</code>. | |
3385 | </li> | |
254750ce | 3386 | |
82c22d34 BP |
3387 | <li> |
3388 | Send an OpenFlow 1.4+ ``table mod request'' to enable eviction for | |
3389 | the flow table (e.g. <code>ovs-ofctl -O OpenFlow14 mod-table br0 0 | |
3390 | evict</code> to enable eviction on flow table 0 of bridge | |
3391 | <code>br0</code>). | |
3392 | </li> | |
3393 | </ul> | |
254750ce BP |
3394 | |
3395 | <p> | |
3396 | When a flow must be evicted due to overflow, the flow to evict is | |
f70b94de BP |
3397 | chosen through an approximation of the following algorithm. This |
3398 | algorithm is used regardless of how eviction was enabled: | |
254750ce BP |
3399 | </p> |
3400 | ||
3401 | <ol> | |
3402 | <li> | |
3403 | Divide the flows in the table into groups based on the values of the | |
f70b94de BP |
3404 | fields or subfields specified in the <ref column="groups"/> column, |
3405 | so that all of the flows in a given group have the same values for | |
3406 | those fields. If a flow does not specify a given field, that field's | |
3407 | value is treated as 0. If <ref column="groups"/> is empty, then all | |
3408 | of the flows in the flow table are treated as a single group. | |
254750ce BP |
3409 | </li> |
3410 | ||
3411 | <li> | |
3412 | Consider the flows in the largest group, that is, the group that | |
3413 | contains the greatest number of flows. If two or more groups all | |
3414 | have the same largest number of flows, consider the flows in all of | |
3415 | those groups. | |
3416 | </li> | |
3417 | ||
f70b94de BP |
3418 | <li> |
3419 | If the flows under consideration have different importance values, | |
3420 | eliminate from consideration any flows except those with the lowest | |
3421 | importance. (``Importance,'' a 16-bit integer value attached to each | |
3422 | flow, was introduced in OpenFlow 1.4. Flows inserted with older | |
3423 | versions of OpenFlow always have an importance of 0.) | |
3424 | </li> | |
3425 | ||
254750ce BP |
3426 | <li> |
3427 | Among the flows under consideration, choose the flow that expires | |
3428 | soonest for eviction. | |
3429 | </li> | |
3430 | </ol> | |
3431 | ||
3432 | <p> | |
82c22d34 BP |
3433 | The eviction process only considers flows that have an idle timeout |
3434 | or a hard timeout. That is, eviction never deletes permanent flows. | |
7792bfe0 | 3435 | (Permanent flows do count against <ref column="flow_limit"/>.) |
254750ce BP |
3436 | </p> |
3437 | ||
82c22d34 BP |
3438 | <column name="flow_limit"> |
3439 | If set, limits the number of flows that may be added to the table. | |
3440 | Open vSwitch may limit the number of flows in a table for other | |
3441 | reasons, e.g. due to hardware limitations or for resource availability | |
3442 | or performance reasons. | |
3443 | </column> | |
254750ce | 3444 | |
82c22d34 BP |
3445 | <column name="overflow_policy"> |
3446 | <p> | |
3447 | Controls the switch's behavior when an OpenFlow flow table | |
3448 | modification request would add flows in excess of <ref | |
3449 | column="flow_limit"/>. The supported values are: | |
3450 | </p> | |
13751fd8 | 3451 | |
82c22d34 BP |
3452 | <dl> |
3453 | <dt><code>refuse</code></dt> | |
3454 | <dd> | |
3455 | Refuse to add the flow or flows. This is also the default policy | |
3456 | when <ref column="overflow_policy"/> is unset. | |
3457 | </dd> | |
f017d986 | 3458 | |
82c22d34 BP |
3459 | <dt><code>evict</code></dt> |
3460 | <dd> | |
3461 | Delete a flow chosen according to the algorithm described above. | |
3462 | </dd> | |
3463 | </dl> | |
3464 | </column> | |
f017d986 | 3465 | |
82c22d34 BP |
3466 | <column name="groups"> |
3467 | <p> | |
3468 | When <ref column="overflow_policy"/> is <code>evict</code>, this | |
3469 | controls how flows are chosen for eviction when the flow table would | |
3470 | otherwise exceed <ref column="flow_limit"/> flows. Its value is a | |
3471 | set of NXM fields or sub-fields, each of which takes one of the forms | |
3472 | <code><var>field</var>[]</code> or | |
3473 | <code><var>field</var>[<var>start</var>..<var>end</var>]</code>, | |
3474 | e.g. <code>NXM_OF_IN_PORT[]</code>. Please see | |
0da61033 | 3475 | <code>meta-flow.h</code> for a complete list of NXM field names. |
82c22d34 | 3476 | </p> |
f017d986 | 3477 | |
82c22d34 BP |
3478 | <p> |
3479 | Open vSwitch ignores any invalid or unknown field specifications. | |
3480 | </p> | |
f017d986 | 3481 | |
82c22d34 BP |
3482 | <p> |
3483 | When eviction is not enabled, via <ref column="overflow_policy"/> or | |
3484 | an OpenFlow 1.4+ ``table mod,'' this column has no effect. | |
3485 | </p> | |
3486 | </column> | |
3487 | </group> | |
13751fd8 | 3488 | |
82c22d34 BP |
3489 | <group title="Classifier Optimization"> |
3490 | <column name="prefixes"> | |
3491 | <p> | |
3492 | This string set specifies which fields should be used for | |
3493 | address prefix tracking. Prefix tracking allows the | |
3494 | classifier to skip rules with longer than necessary prefixes, | |
3495 | resulting in better wildcarding for datapath flows. | |
3496 | </p> | |
3497 | <p> | |
3498 | Prefix tracking may be beneficial when a flow table contains | |
3499 | matches on IP address fields with different prefix lengths. | |
3500 | For example, when a flow table contains IP address matches on | |
3501 | both full addresses and proper prefixes, the full address | |
3502 | matches will typically cause the datapath flow to un-wildcard | |
3503 | the whole address field (depending on flow entry priorities). | |
3504 | In this case each packet with a different address gets handed | |
3505 | to the userspace for flow processing and generates its own | |
3506 | datapath flow. With prefix tracking enabled for the address | |
3507 | field in question packets with addresses matching shorter | |
3508 | prefixes would generate datapath flows where the irrelevant | |
3509 | address bits are wildcarded, allowing the same datapath flow | |
3510 | to handle all the packets within the prefix in question. In | |
3511 | this case many userspace upcalls can be avoided and the | |
3512 | overall performance can be better. | |
3513 | </p> | |
3514 | <p> | |
3515 | This is a performance optimization only, so packets will | |
3516 | receive the same treatment with or without prefix tracking. | |
3517 | </p> | |
3518 | <p> | |
3519 | The supported fields are: <code>tun_id</code>, | |
3520 | <code>tun_src</code>, <code>tun_dst</code>, | |
80c4589a | 3521 | <code>tun_ipv6_src</code>, <code>tun_ipv6_dst</code>, |
82c22d34 BP |
3522 | <code>nw_src</code>, <code>nw_dst</code> (or aliases |
3523 | <code>ip_src</code> and <code>ip_dst</code>), | |
3524 | <code>ipv6_src</code>, and <code>ipv6_dst</code>. (Using this | |
3525 | feature for <code>tun_id</code> would only make sense if the | |
3526 | tunnel IDs have prefix structure similar to IP addresses.) | |
3527 | </p> | |
13751fd8 | 3528 | |
82c22d34 BP |
3529 | <p> |
3530 | By default, the <code>prefixes=ip_dst,ip_src</code> are used | |
3531 | on each flow table. This instructs the flow classifier to | |
3532 | track the IP destination and source addresses used by the | |
3533 | rules in this specific flow table. | |
3534 | </p> | |
3535 | ||
3536 | <p> | |
3537 | The keyword <code>none</code> is recognized as an explicit | |
3538 | override of the default values, causing no prefix fields to be | |
3539 | tracked. | |
3540 | </p> | |
3541 | ||
3542 | <p> | |
3543 | To set the prefix fields, the flow table record needs to | |
3544 | exist: | |
3545 | </p> | |
3546 | ||
3547 | <dl> | |
3548 | <dt><code>ovs-vsctl set Bridge br0 flow_tables:0=@N1 -- --id=@N1 create Flow_Table name=table0</code></dt> | |
3549 | <dd> | |
3550 | Creates a flow table record for the OpenFlow table number 0. | |
3551 | </dd> | |
3552 | ||
3553 | <dt><code>ovs-vsctl set Flow_Table table0 prefixes=ip_dst,ip_src</code></dt> | |
3554 | <dd> | |
3555 | Enables prefix tracking for IP source and destination | |
3556 | address fields. | |
3557 | </dd> | |
3558 | </dl> | |
3559 | ||
3560 | <p> | |
3561 | There is a maximum number of fields that can be enabled for any | |
3562 | one flow table. Currently this limit is 3. | |
3563 | </p> | |
3564 | </column> | |
3565 | </group> | |
e3fbd9df BP |
3566 | |
3567 | <group title="Common Columns"> | |
3568 | The overall purpose of these columns is described under <code>Common | |
3569 | Columns</code> at the beginning of this document. | |
3570 | ||
3571 | <column name="external_ids"/> | |
3572 | </group> | |
254750ce BP |
3573 | </table> |
3574 | ||
c1c9c9c4 BP |
3575 | <table name="QoS" title="Quality of Service configuration"> |
3576 | <p>Quality of Service (QoS) configuration for each Port that | |
3fd8d445 | 3577 | references it.</p> |
c1c9c9c4 BP |
3578 | |
3579 | <column name="type"> | |
b850dc6d BP |
3580 | <p>The type of QoS to implement. The currently defined types are |
3581 | listed below:</p> | |
c1c9c9c4 BP |
3582 | <dl> |
3583 | <dt><code>linux-htb</code></dt> | |
6784cb57 BP |
3584 | <dd> |
3585 | Linux ``hierarchy token bucket'' classifier. See tc-htb(8) (also at | |
3586 | <code>http://linux.die.net/man/8/tc-htb</code>) and the HTB manual | |
3587 | (<code>http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm</code>) | |
3588 | for information on how this classifier works and how to configure it. | |
3589 | </dd> | |
6cf888b8 | 3590 | |
a339aa81 EJ |
3591 | <dt><code>linux-hfsc</code></dt> |
3592 | <dd> | |
3593 | Linux "Hierarchical Fair Service Curve" classifier. | |
3594 | See <code>http://linux-ip.net/articles/hfsc.en/</code> for | |
3595 | information on how this classifier works. | |
3596 | </dd> | |
6cf888b8 | 3597 | |
677d9158 JV |
3598 | <dt><code>linux-sfq</code></dt> |
3599 | <dd> | |
3600 | Linux ``Stochastic Fairness Queueing'' classifier. See | |
3601 | <code>tc-sfq</code>(8) (also at | |
3602 | <code>http://linux.die.net/man/8/tc-sfq</code>) for information on | |
3603 | how this classifier works. | |
3604 | </dd> | |
6cf888b8 | 3605 | |
677d9158 JV |
3606 | <dt><code>linux-codel</code></dt> |
3607 | <dd> | |
3608 | Linux ``Controlled Delay'' classifier. See <code>tc-codel</code>(8) | |
3609 | (also at | |
3610 | <code>http://man7.org/linux/man-pages/man8/tc-codel.8.html</code>) | |
3611 | for information on how this classifier works. | |
3612 | </dd> | |
6cf888b8 | 3613 | |
677d9158 JV |
3614 | <dt><code>linux-fq_codel</code></dt> |
3615 | <dd> | |
3616 | Linux ``Fair Queuing with Controlled Delay'' classifier. See | |
3617 | <code>tc-fq_codel</code>(8) (also at | |
3618 | <code>http://man7.org/linux/man-pages/man8/tc-fq_codel.8.html</code>) | |
3619 | for information on how this classifier works. | |
3620 | </dd> | |
6cf888b8 BS |
3621 | |
3622 | <dt><code>linux-noop</code></dt> | |
3623 | <dd> | |
3624 | Linux ``No operation.'' By default, Open vSwitch manages quality of | |
3625 | service on all of its configured ports. This can be helpful, but | |
3626 | sometimes administrators prefer to use other software to manage QoS. | |
3627 | This <ref column="type"/> prevents Open vSwitch from changing the QoS | |
3628 | configuration for a port. | |
3629 | </dd> | |
3630 | ||
0bf765f7 IS |
3631 | <dt><code>egress-policer</code></dt> |
3632 | <dd> | |
6cf888b8 | 3633 | A DPDK egress policer algorithm using the DPDK |
0bf765f7 IS |
3634 | rte_meter library. The rte_meter library provides an implementation |
3635 | which allows the metering and policing of traffic. The implementation | |
3636 | in OVS essentially creates a single token bucket used to police | |
3637 | traffic. It should be noted that when the rte_meter is configured as | |
3638 | part of QoS there will be a performance overhead as the rte_meter | |
3639 | itself will consume CPU cycles in order to police traffic. These CPU | |
3640 | cycles ordinarily are used for packet proccessing. As such the drop | |
3641 | in performance will be noticed in terms of overall aggregate traffic | |
3642 | throughput. | |
3643 | </dd> | |
3644 | </dl> | |
c1c9c9c4 BP |
3645 | </column> |
3646 | ||
3647 | <column name="queues"> | |
3648 | <p>A map from queue numbers to <ref table="Queue"/> records. The | |
3fd8d445 BP |
3649 | supported range of queue numbers depend on <ref column="type"/>. The |
3650 | queue numbers are the same as the <code>queue_id</code> used in | |
3651 | OpenFlow in <code>struct ofp_action_enqueue</code> and other | |
2c999774 BP |
3652 | structures.</p> |
3653 | ||
3654 | <p> | |
3655 | Queue 0 is the ``default queue.'' It is used by OpenFlow output | |
8bddb894 BP |
3656 | actions when no specific queue has been set. When no configuration for |
3657 | queue 0 is present, it is automatically configured as if a <ref | |
3658 | table="Queue"/> record with empty <ref table="Queue" column="dscp"/> | |
3659 | and <ref table="Queue" column="other_config"/> columns had been | |
3660 | specified. | |
2c999774 BP |
3661 | (Before version 1.6, Open vSwitch would leave queue 0 unconfigured in |
3662 | this case. With some queuing disciplines, this dropped all packets | |
3663 | destined for the default queue.) | |
3664 | </p> | |
c1c9c9c4 BP |
3665 | </column> |
3666 | ||
3fd8d445 BP |
3667 | <group title="Configuration for linux-htb and linux-hfsc"> |
3668 | <p> | |
3669 | The <code>linux-htb</code> and <code>linux-hfsc</code> classes support | |
3670 | the following key-value pair: | |
3671 | </p> | |
9cc6bf75 | 3672 | |
f9e5e5b3 | 3673 | <column name="other_config" key="max-rate" type='{"type": "integer"}'> |
3fd8d445 BP |
3674 | Maximum rate shared by all queued traffic, in bit/s. Optional. If not |
3675 | specified, for physical interfaces, the default is the link rate. For | |
3676 | other interfaces or if the link rate cannot be determined, the default | |
3677 | is currently 100 Mbps. | |
3678 | </column> | |
3679 | </group> | |
13008eb3 | 3680 | |
0bf765f7 IS |
3681 | <group title="Configuration for egress-policer QoS"> |
3682 | <p> | |
3683 | <ref table="QoS"/> <ref table="QoS" column="type"/> | |
3684 | <code>egress-policer</code> provides egress policing for userspace | |
3685 | port types with DPDK. | |
3686 | ||
3687 | It has the following key-value pairs defined. | |
3688 | </p> | |
3689 | ||
3690 | <column name="other_config" key="cir" type='{"type": "integer"}'> | |
3691 | The Committed Information Rate (CIR) is measured in bytes of IP | |
3692 | packets per second, i.e. it includes the IP header, but not link | |
3693 | specific (e.g. Ethernet) headers. This represents the bytes per second | |
3694 | rate at which the token bucket will be updated. The cir value is | |
3695 | calculated by (pps x packet data size). For example assuming a user | |
3696 | wishes to limit a stream consisting of 64 byte packets to 1 million | |
3697 | packets per second the CIR would be set to to to 46000000. This value | |
3698 | can be broken into '1,000,000 x 46'. Where 1,000,000 is the policing | |
3699 | rate for the number of packets per second and 46 represents the size | |
3700 | of the packet data for a 64 byte ip packet. | |
3701 | </column> | |
3702 | <column name="other_config" key="cbs" type='{"type": "integer"}'> | |
3703 | The Committed Burst Size (CBS) is measured in bytes and represents a | |
3704 | token bucket. At a minimum this value should be be set to the expected | |
3705 | largest size packet in the traffic stream. In practice larger values | |
3706 | may be used to increase the size of the token bucket. If a packet can | |
3707 | be transmitted then the cbs will be decremented by the number of | |
3708 | bytes/tokens of the packet. If there are not enough tokens in the cbs | |
3709 | bucket the packet will be dropped. | |
3710 | </column> | |
3711 | </group> | |
3712 | ||
3fd8d445 BP |
3713 | <group title="Common Columns"> |
3714 | The overall purpose of these columns is described under <code>Common | |
3715 | Columns</code> at the beginning of this document. | |
3716 | ||
3717 | <column name="other_config"/> | |
3718 | <column name="external_ids"/> | |
3719 | </group> | |
c1c9c9c4 BP |
3720 | </table> |
3721 | ||
3722 | <table name="Queue" title="QoS output queue."> | |
3723 | <p>A configuration for a port output queue, used in configuring Quality of | |
3fd8d445 BP |
3724 | Service (QoS) features. May be referenced by <ref column="queues" |
3725 | table="QoS"/> column in <ref table="QoS"/> table.</p> | |
13008eb3 | 3726 | |
8b36f51e EJ |
3727 | <column name="dscp"> |
3728 | If set, Open vSwitch will mark all traffic egressing this | |
3729 | <ref table="Queue"/> with the given DSCP bits. Traffic egressing the | |
3730 | default <ref table="Queue"/> is only marked if it was explicitly selected | |
3731 | as the <ref table="Queue"/> at the time the packet was output. If unset, | |
3732 | the DSCP bits of traffic egressing this <ref table="Queue"/> will remain | |
3733 | unchanged. | |
3734 | </column> | |
3735 | ||
3fd8d445 BP |
3736 | <group title="Configuration for linux-htb QoS"> |
3737 | <p> | |
69822b3c EJ |
3738 | <ref table="QoS"/> <ref table="QoS" column="type"/> |
3739 | <code>linux-htb</code> may use <code>queue_id</code>s less than 61440. | |
3740 | It has the following key-value pairs defined. | |
3fd8d445 | 3741 | </p> |
9cc6bf75 | 3742 | |
f9e5e5b3 BP |
3743 | <column name="other_config" key="min-rate" |
3744 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
3745 | Minimum guaranteed bandwidth, in bit/s. |
3746 | </column> | |
3747 | ||
f9e5e5b3 BP |
3748 | <column name="other_config" key="max-rate" |
3749 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
3750 | Maximum allowed bandwidth, in bit/s. Optional. If specified, the |
3751 | queue's rate will not be allowed to exceed the specified value, even | |
3752 | if excess bandwidth is available. If unspecified, defaults to no | |
3753 | limit. | |
3754 | </column> | |
3755 | ||
f9e5e5b3 BP |
3756 | <column name="other_config" key="burst" |
3757 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
3758 | Burst size, in bits. This is the maximum amount of ``credits'' that a |
3759 | queue can accumulate while it is idle. Optional. Details of the | |
3760 | <code>linux-htb</code> implementation require a minimum burst size, so | |
3761 | a too-small <code>burst</code> will be silently ignored. | |
3762 | </column> | |
3763 | ||
f9e5e5b3 BP |
3764 | <column name="other_config" key="priority" |
3765 | type='{"type": "integer", "minInteger": 0, "maxInteger": 4294967295}'> | |
3766 | A queue with a smaller <code>priority</code> will receive all the | |
3767 | excess bandwidth that it can use before a queue with a larger value | |
3768 | receives any. Specific priority values are unimportant; only relative | |
3769 | ordering matters. Defaults to 0 if unspecified. | |
3fd8d445 BP |
3770 | </column> |
3771 | </group> | |
3772 | ||
3773 | <group title="Configuration for linux-hfsc QoS"> | |
3774 | <p> | |
69822b3c EJ |
3775 | <ref table="QoS"/> <ref table="QoS" column="type"/> |
3776 | <code>linux-hfsc</code> may use <code>queue_id</code>s less than 61440. | |
3777 | It has the following key-value pairs defined. | |
3fd8d445 | 3778 | </p> |
9cc6bf75 | 3779 | |
f9e5e5b3 BP |
3780 | <column name="other_config" key="min-rate" |
3781 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
3782 | Minimum guaranteed bandwidth, in bit/s. |
3783 | </column> | |
9cc6bf75 | 3784 | |
f9e5e5b3 BP |
3785 | <column name="other_config" key="max-rate" |
3786 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
3787 | Maximum allowed bandwidth, in bit/s. Optional. If specified, the |
3788 | queue's rate will not be allowed to exceed the specified value, even if | |
3789 | excess bandwidth is available. If unspecified, defaults to no | |
3790 | limit. | |
3791 | </column> | |
3792 | </group> | |
3793 | ||
3794 | <group title="Common Columns"> | |
3795 | The overall purpose of these columns is described under <code>Common | |
3796 | Columns</code> at the beginning of this document. | |
3797 | ||
3798 | <column name="other_config"/> | |
3799 | <column name="external_ids"/> | |
3800 | </group> | |
c1c9c9c4 BP |
3801 | </table> |
3802 | ||
9ae7ddc0 | 3803 | <table name="Mirror" title="Port mirroring."> |
89365653 BP |
3804 | <p>A port mirror within a <ref table="Bridge"/>.</p> |
3805 | <p>A port mirror configures a bridge to send selected frames to special | |
92ada132 | 3806 | ``mirrored'' ports, in addition to their normal destinations. Mirroring |
9ae7ddc0 | 3807 | traffic may also be referred to as SPAN or RSPAN, depending on how |
92ada132 | 3808 | the mirrored traffic is sent.</p> |
89365653 | 3809 | |
7efbc3b7 BP |
3810 | <p> |
3811 | When a packet enters an Open vSwitch bridge, it becomes eligible for | |
3812 | mirroring based on its ingress port and VLAN. As the packet travels | |
3813 | through the flow tables, each time it is output to a port, it becomes | |
3814 | eligible for mirroring based on the egress port and VLAN. In Open | |
3815 | vSwitch 2.5 and later, mirroring occurs just after a packet first becomes | |
3816 | eligible, using the packet as it exists at that point; in Open vSwitch | |
3817 | 2.4 and earlier, mirroring occurs only after a packet has traversed all | |
3818 | the flow tables, using the original packet as it entered the bridge. | |
3819 | This makes a difference only when the flow table modifies the packet: in | |
3820 | Open vSwitch 2.4, the modifications are never visible to mirrors, whereas | |
3821 | in Open vSwitch 2.5 and later modifications made before the first output | |
3822 | that makes it eligible for mirroring to a particular destination are | |
3823 | visible. | |
3824 | </p> | |
3825 | ||
3826 | <p> | |
3827 | A packet that enters an Open vSwitch bridge is mirrored to a particular | |
3828 | destination only once, even if it is eligible for multiple reasons. For | |
3829 | example, a packet would be mirrored to a particular <ref | |
3830 | column="output_port"/> only once, even if it is selected for mirroring to | |
3831 | that port by <ref column="select_dst_port"/> and <ref | |
3832 | column="select_src_port"/> in the same or different <ref table="Mirror"/> | |
3833 | records. | |
3834 | </p> | |
3835 | ||
89365653 BP |
3836 | <column name="name"> |
3837 | Arbitrary identifier for the <ref table="Mirror"/>. | |
3838 | </column> | |
3839 | ||
3840 | <group title="Selecting Packets for Mirroring"> | |
3e519d8e BP |
3841 | <p> |
3842 | To be selected for mirroring, a given packet must enter or leave the | |
3843 | bridge through a selected port and it must also be in one of the | |
3844 | selected VLANs. | |
3845 | </p> | |
3846 | ||
939ff267 BP |
3847 | <column name="select_all"> |
3848 | If true, every packet arriving or departing on any port is | |
3849 | selected for mirroring. | |
3850 | </column> | |
3851 | ||
89365653 BP |
3852 | <column name="select_dst_port"> |
3853 | Ports on which departing packets are selected for mirroring. | |
3854 | </column> | |
3855 | ||
3856 | <column name="select_src_port"> | |
939ff267 | 3857 | Ports on which arriving packets are selected for mirroring. |
89365653 BP |
3858 | </column> |
3859 | ||
3860 | <column name="select_vlan"> | |
3861 | VLANs on which packets are selected for mirroring. An empty set | |
3862 | selects packets on all VLANs. | |
3863 | </column> | |
3864 | </group> | |
3865 | ||
3866 | <group title="Mirroring Destination Configuration"> | |
3e519d8e BP |
3867 | <p> |
3868 | These columns are mutually exclusive. Exactly one of them must be | |
3869 | nonempty. | |
3870 | </p> | |
3871 | ||
89365653 | 3872 | <column name="output_port"> |
3e519d8e | 3873 | <p>Output port for selected packets, if nonempty.</p> |
89365653 | 3874 | <p>Specifying a port for mirror output reserves that port exclusively |
92ada132 | 3875 | for mirroring. No frames other than those selected for mirroring |
653fe3a3 | 3876 | via this column |
92ada132 BP |
3877 | will be forwarded to the port, and any frames received on the port |
3878 | will be discarded.</p> | |
3879 | <p> | |
3880 | The output port may be any kind of port supported by Open vSwitch. | |
9ae7ddc0 JP |
3881 | It may be, for example, a physical port (sometimes called SPAN) or a |
3882 | GRE tunnel. | |
92ada132 | 3883 | </p> |
89365653 BP |
3884 | </column> |
3885 | ||
3886 | <column name="output_vlan"> | |
3e519d8e | 3887 | <p>Output VLAN for selected packets, if nonempty.</p> |
89365653 | 3888 | <p>The frames will be sent out all ports that trunk |
3fd8d445 BP |
3889 | <ref column="output_vlan"/>, as well as any ports with implicit VLAN |
3890 | <ref column="output_vlan"/>. When a mirrored frame is sent out a | |
3891 | trunk port, the frame's VLAN tag will be set to | |
3892 | <ref column="output_vlan"/>, replacing any existing tag; when it is | |
3893 | sent out an implicit VLAN port, the frame will not be tagged. This | |
3894 | type of mirroring is sometimes called RSPAN.</p> | |
07817dfe | 3895 | <p> |
05be4e2c EJ |
3896 | See the documentation for |
3897 | <ref column="other_config" key="forward-bpdu"/> in the | |
3898 | <ref table="Interface"/> table for a list of destination MAC | |
3899 | addresses which will not be mirrored to a VLAN to avoid confusing | |
3900 | switches that interpret the protocols that they represent. | |
07817dfe | 3901 | </p> |
89365653 | 3902 | <p><em>Please note:</em> Mirroring to a VLAN can disrupt a network that |
3fd8d445 BP |
3903 | contains unmanaged switches. Consider an unmanaged physical switch |
3904 | with two ports: port 1, connected to an end host, and port 2, | |
3905 | connected to an Open vSwitch configured to mirror received packets | |
3906 | into VLAN 123 on port 2. Suppose that the end host sends a packet on | |
3907 | port 1 that the physical switch forwards to port 2. The Open vSwitch | |
3908 | forwards this packet to its destination and then reflects it back on | |
3909 | port 2 in VLAN 123. This reflected packet causes the unmanaged | |
3910 | physical switch to replace the MAC learning table entry, which | |
3911 | correctly pointed to port 1, with one that incorrectly points to port | |
3912 | 2. Afterward, the physical switch will direct packets destined for | |
3913 | the end host to the Open vSwitch on port 2, instead of to the end | |
3914 | host on port 1, disrupting connectivity. If mirroring to a VLAN is | |
3915 | desired in this scenario, then the physical switch must be replaced | |
3916 | by one that learns Ethernet addresses on a per-VLAN basis. In | |
3917 | addition, learning should be disabled on the VLAN containing mirrored | |
3918 | traffic. If this is not done then intermediate switches will learn | |
3919 | the MAC address of each end host from the mirrored traffic. If | |
3920 | packets being sent to that end host are also mirrored, then they will | |
3921 | be dropped since the switch will attempt to send them out the input | |
3922 | port. Disabling learning for the VLAN will cause the switch to | |
3923 | correctly send the packet out all ports configured for that VLAN. If | |
3924 | Open vSwitch is being used as an intermediate switch, learning can be | |
3925 | disabled by adding the mirrored VLAN to <ref column="flood_vlans"/> | |
3926 | in the appropriate <ref table="Bridge"/> table or tables.</p> | |
3927 | <p> | |
3928 | Mirroring to a GRE tunnel has fewer caveats than mirroring to a | |
3929 | VLAN and should generally be preferred. | |
3930 | </p> | |
89365653 | 3931 | </column> |
1356dbd1 WT |
3932 | |
3933 | <column name="snaplen"> | |
3934 | <p>Maximum per-packet number of bytes to mirror.</p> | |
3935 | <p>A mirrored packet with size larger than <ref column="snaplen"/> | |
3936 | will be truncated in datapath to <ref column="snaplen"/> bytes | |
3937 | before sending to the mirror output port. If omitted, packets | |
3938 | are not truncated. | |
3939 | </p> | |
3940 | </column> | |
89365653 | 3941 | </group> |
13008eb3 | 3942 | |
9d24de3b JP |
3943 | <group title="Statistics: Mirror counters"> |
3944 | <p> | |
12eb035b AW |
3945 | Key-value pairs that report mirror statistics. The update period |
3946 | is controlled by <ref column="other_config" | |
3947 | key="stats-update-interval"/> in the <code>Open_vSwitch</code> table. | |
9d24de3b JP |
3948 | </p> |
3949 | <column name="statistics" key="tx_packets"> | |
3950 | Number of packets transmitted through this mirror. | |
3951 | </column> | |
3952 | <column name="statistics" key="tx_bytes"> | |
3953 | Number of bytes transmitted through this mirror. | |
3954 | </column> | |
3955 | </group> | |
3956 | ||
3fd8d445 BP |
3957 | <group title="Common Columns"> |
3958 | The overall purpose of these columns is described under <code>Common | |
3959 | Columns</code> at the beginning of this document. | |
3960 | ||
3961 | <column name="external_ids"/> | |
13008eb3 | 3962 | </group> |
89365653 BP |
3963 | </table> |
3964 | ||
3965 | <table name="Controller" title="OpenFlow controller configuration."> | |
76ce9432 BP |
3966 | <p>An OpenFlow controller.</p> |
3967 | ||
7d674866 BP |
3968 | <p> |
3969 | Open vSwitch supports two kinds of OpenFlow controllers: | |
3970 | </p> | |
299a244b | 3971 | |
7d674866 BP |
3972 | <dl> |
3973 | <dt>Primary controllers</dt> | |
3974 | <dd> | |
3975 | <p> | |
3976 | This is the kind of controller envisioned by the OpenFlow 1.0 | |
3977 | specification. Usually, a primary controller implements a network | |
3978 | policy by taking charge of the switch's flow table. | |
3979 | </p> | |
3980 | ||
3981 | <p> | |
3982 | Open vSwitch initiates and maintains persistent connections to | |
3983 | primary controllers, retrying the connection each time it fails or | |
3984 | drops. The <ref table="Bridge" column="fail_mode"/> column in the | |
3985 | <ref table="Bridge"/> table applies to primary controllers. | |
3986 | </p> | |
3987 | ||
3988 | <p> | |
3989 | Open vSwitch permits a bridge to have any number of primary | |
3990 | controllers. When multiple controllers are configured, Open | |
3991 | vSwitch connects to all of them simultaneously. Because | |
3992 | OpenFlow 1.0 does not specify how multiple controllers | |
3993 | coordinate in interacting with a single switch, more than | |
3994 | one primary controller should be specified only if the | |
3995 | controllers are themselves designed to coordinate with each | |
3996 | other. (The Nicira-defined <code>NXT_ROLE</code> OpenFlow | |
3997 | vendor extension may be useful for this.) | |
3998 | </p> | |
3999 | </dd> | |
4000 | <dt>Service controllers</dt> | |
4001 | <dd> | |
4002 | <p> | |
4003 | These kinds of OpenFlow controller connections are intended for | |
4004 | occasional support and maintenance use, e.g. with | |
4005 | <code>ovs-ofctl</code>. Usually a service controller connects only | |
4006 | briefly to inspect or modify some of a switch's state. | |
4007 | </p> | |
4008 | ||
4009 | <p> | |
4010 | Open vSwitch listens for incoming connections from service | |
4011 | controllers. The service controllers initiate and, if necessary, | |
4012 | maintain the connections from their end. The <ref table="Bridge" | |
4013 | column="fail_mode"/> column in the <ref table="Bridge"/> table does | |
4014 | not apply to service controllers. | |
4015 | </p> | |
4016 | ||
4017 | <p> | |
4018 | Open vSwitch supports configuring any number of service controllers. | |
4019 | </p> | |
4020 | </dd> | |
4021 | </dl> | |
4022 | ||
4023 | <p> | |
4024 | The <ref column="target"/> determines the type of controller. | |
4025 | </p> | |
89365653 BP |
4026 | |
4027 | <group title="Core Features"> | |
4028 | <column name="target"> | |
7d674866 BP |
4029 | <p>Connection method for controller.</p> |
4030 | <p> | |
4031 | The following connection methods are currently supported for primary | |
4032 | controllers: | |
4033 | </p> | |
89365653 BP |
4034 | <dl> |
4035 | <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt> | |
4036 | <dd> | |
125b0291 JP |
4037 | <p>The specified SSL <var>port</var> on the host at the |
4038 | given <var>ip</var>, which must be expressed as an IP | |
4039 | address (not a DNS name). The <ref table="Open_vSwitch" | |
4040 | column="ssl"/> column in the <ref table="Open_vSwitch"/> | |
4041 | table must point to a valid SSL configuration when this form | |
4042 | is used.</p> | |
d4763d1d | 4043 | <p>If <var>port</var> is not specified, it defaults to 6653.</p> |
89365653 | 4044 | <p>SSL support is an optional feature that is not always built as |
3fd8d445 | 4045 | part of Open vSwitch.</p> |
89365653 BP |
4046 | </dd> |
4047 | <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt> | |
125b0291 | 4048 | <dd> |
e731d71b AS |
4049 | <p> |
4050 | The specified TCP <var>port</var> on the host at the given | |
4051 | <var>ip</var>, which must be expressed as an IP address (not a | |
4052 | DNS name), where <var>ip</var> can be IPv4 or IPv6 address. If | |
4053 | <var>ip</var> is an IPv6 address, wrap it in square brackets, | |
d4763d1d | 4054 | e.g. <code>tcp:[::1]:6653</code>. |
e731d71b AS |
4055 | </p> |
4056 | <p> | |
d4763d1d | 4057 | If <var>port</var> is not specified, it defaults to 6653. |
e731d71b | 4058 | </p> |
125b0291 | 4059 | </dd> |
7d674866 BP |
4060 | </dl> |
4061 | <p> | |
4062 | The following connection methods are currently supported for service | |
4063 | controllers: | |
4064 | </p> | |
4065 | <dl> | |
4066 | <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt> | |
4067 | <dd> | |
e731d71b AS |
4068 | <p> |
4069 | Listens for SSL connections on the specified TCP <var>port</var>. | |
4070 | If <var>ip</var>, which must be expressed as an IP address (not a | |
4071 | DNS name), is specified, then connections are restricted to the | |
4072 | specified local IP address (either IPv4 or IPv6). If | |
4073 | <var>ip</var> is an IPv6 address, wrap it in square brackets, | |
d4763d1d | 4074 | e.g. <code>pssl:6653:[::1]</code>. |
e731d71b AS |
4075 | </p> |
4076 | <p> | |
d4763d1d JP |
4077 | If <var>port</var> is not specified, it defaults to |
4078 | 6653. If <var>ip</var> is not specified then it listens only on | |
e731d71b AS |
4079 | IPv4 (but not IPv6) addresses. The |
4080 | <ref table="Open_vSwitch" column="ssl"/> | |
4081 | column in the <ref table="Open_vSwitch"/> table must point to a | |
4082 | valid SSL configuration when this form is used. | |
4083 | </p> | |
4084 | <p> | |
d4763d1d | 4085 | If <var>port</var> is not specified, it currently to 6653. |
e731d71b AS |
4086 | </p> |
4087 | <p> | |
4088 | SSL support is an optional feature that is not always built as | |
4089 | part of Open vSwitch. | |
4090 | </p> | |
7d674866 BP |
4091 | </dd> |
4092 | <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt> | |
4093 | <dd> | |
e731d71b AS |
4094 | <p> |
4095 | Listens for connections on the specified TCP <var>port</var>. If | |
4096 | <var>ip</var>, which must be expressed as an IP address (not a | |
4097 | DNS name), is specified, then connections are restricted to the | |
4098 | specified local IP address (either IPv4 or IPv6). If | |
4099 | <var>ip</var> is an IPv6 address, wrap it in square brackets, | |
d4763d1d | 4100 | e.g. <code>ptcp:6653:[::1]</code>. If <var>ip</var> is not |
e731d71b AS |
4101 | specified then it listens only on IPv4 addresses. |
4102 | </p> | |
4103 | <p> | |
d4763d1d | 4104 | If <var>port</var> is not specified, it defaults to 6653. |
e731d71b | 4105 | </p> |
7d674866 | 4106 | </dd> |
89365653 | 4107 | </dl> |
9a3f4a49 | 4108 | <p>When multiple controllers are configured for a single bridge, the |
3fd8d445 BP |
4109 | <ref column="target"/> values must be unique. Duplicate |
4110 | <ref column="target"/> values yield unspecified results.</p> | |
89365653 BP |
4111 | </column> |
4112 | ||
4113 | <column name="connection_mode"> | |
9a3f4a49 JP |
4114 | <p>If it is specified, this setting must be one of the following |
4115 | strings that describes how Open vSwitch contacts this OpenFlow | |
4116 | controller over the network:</p> | |
4117 | ||
4118 | <dl> | |
4119 | <dt><code>in-band</code></dt> | |
4120 | <dd>In this mode, this controller's OpenFlow traffic travels over the | |
3fd8d445 BP |
4121 | bridge associated with the controller. With this setting, Open |
4122 | vSwitch allows traffic to and from the controller regardless of the | |
4123 | contents of the OpenFlow flow table. (Otherwise, Open vSwitch | |
4124 | would never be able to connect to the controller, because it did | |
4125 | not have a flow to enable it.) This is the most common connection | |
4126 | mode because it is not necessary to maintain two independent | |
4127 | networks.</dd> | |
9a3f4a49 JP |
4128 | <dt><code>out-of-band</code></dt> |
4129 | <dd>In this mode, OpenFlow traffic uses a control network separate | |
3fd8d445 BP |
4130 | from the bridge associated with this controller, that is, the |
4131 | bridge does not use any of its own network devices to communicate | |
4132 | with the controller. The control network must be configured | |
4133 | separately, before or after <code>ovs-vswitchd</code> is started. | |
9a3f4a49 JP |
4134 | </dd> |
4135 | </dl> | |
76ce9432 | 4136 | |
195c8086 | 4137 | <p>If not specified, the default is implementation-specific.</p> |
89365653 BP |
4138 | </column> |
4139 | </group> | |
4140 | ||
4141 | <group title="Controller Failure Detection and Handling"> | |
4142 | <column name="max_backoff"> | |
4143 | Maximum number of milliseconds to wait between connection attempts. | |
4144 | Default is implementation-specific. | |
4145 | </column> | |
4146 | ||
4147 | <column name="inactivity_probe"> | |
4148 | Maximum number of milliseconds of idle time on connection to | |
4149 | controller before sending an inactivity probe message. If Open | |
4150 | vSwitch does not communicate with the controller for the specified | |
4151 | number of seconds, it will send a probe. If a response is not | |
4152 | received for the same additional amount of time, Open vSwitch | |
4153 | assumes the connection has been broken and attempts to reconnect. | |
2bb82bf0 BP |
4154 | Default is implementation-specific. A value of 0 disables |
4155 | inactivity probes. | |
89365653 | 4156 | </column> |
89365653 BP |
4157 | </group> |
4158 | ||
a413195e | 4159 | <group title="Asynchronous Messages"> |
9886b662 BP |
4160 | <p> |
4161 | OpenFlow switches send certain messages to controllers spontanenously, | |
4162 | that is, not in response to any request from the controller. These | |
4163 | messages are called ``asynchronous messages.'' These columns allow | |
4164 | asynchronous messages to be limited or disabled to ensure the best use | |
4165 | of network resources. | |
4166 | </p> | |
4167 | ||
4168 | <column name="enable_async_messages"> | |
4169 | The OpenFlow protocol enables asynchronous messages at time of | |
4170 | connection establishment, which means that a controller can receive | |
4171 | asynchronous messages, potentially many of them, even if it turns them | |
4172 | off immediately after connecting. Set this column to | |
4173 | <code>false</code> to change Open vSwitch behavior to disable, by | |
4174 | default, all asynchronous messages. The controller can use the | |
4175 | <code>NXT_SET_ASYNC_CONFIG</code> Nicira extension to OpenFlow to turn | |
4176 | on any messages that it does want to receive, if any. | |
4177 | </column> | |
4178 | ||
a413195e | 4179 | <group title="Controller Rate Limiting"> |
ebb65354 | 4180 | <p> |
a413195e BP |
4181 | A switch can forward packets to a controller over the OpenFlow |
4182 | protocol. Forwarding packets this way at too high a rate can | |
4183 | overwhelm a controller, frustrate use of the OpenFlow connection for | |
4184 | other purposes, increase the latency of flow setup, and use an | |
4185 | unreasonable amount of bandwidth. Therefore, Open vSwitch supports | |
4186 | limiting the rate of packet forwarding to a controller. | |
ebb65354 BP |
4187 | </p> |
4188 | ||
4189 | <p> | |
a413195e BP |
4190 | There are two main reasons in OpenFlow for a packet to be sent to a |
4191 | controller: either the packet ``misses'' in the flow table, that is, | |
4192 | there is no matching flow, or a flow table action says to send the | |
4193 | packet to the controller. Open vSwitch limits the rate of each kind | |
4194 | of packet separately at the configured rate. Therefore, the actual | |
4195 | rate that packets are sent to the controller can be up to twice the | |
4196 | configured rate, when packets are sent for both reasons. | |
ebb65354 BP |
4197 | </p> |
4198 | ||
4199 | <p> | |
a413195e BP |
4200 | This feature is specific to forwarding packets over an OpenFlow |
4201 | connection. It is not general-purpose QoS. See the <ref | |
4202 | table="QoS"/> table for quality of service configuration, and <ref | |
4203 | column="ingress_policing_rate" table="Interface"/> in the <ref | |
4204 | table="Interface"/> table for ingress policing configuration. | |
ebb65354 | 4205 | </p> |
3fd8d445 | 4206 | |
a413195e BP |
4207 | <column name="controller_rate_limit"> |
4208 | <p> | |
4209 | The maximum rate at which the switch will forward packets to the | |
4210 | OpenFlow controller, in packets per second. If no value is | |
4211 | specified, rate limiting is disabled. | |
4212 | </p> | |
4213 | </column> | |
4214 | ||
4215 | <column name="controller_burst_limit"> | |
4216 | <p> | |
4217 | When a high rate triggers rate-limiting, Open vSwitch queues | |
4218 | packets to the controller for each port and transmits them to the | |
4219 | controller at the configured rate. This value limits the number of | |
4220 | queued packets. Ports on a bridge share the packet queue fairly. | |
4221 | </p> | |
4222 | ||
4223 | <p> | |
4224 | This value has no effect unless <ref | |
4225 | column="controller_rate_limit"/> is configured. The current | |
4226 | default when this value is not specified is one-quarter of <ref | |
4227 | column="controller_rate_limit"/>, meaning that queuing can delay | |
4228 | forwarding a packet to the controller by up to 250 ms. | |
4229 | </p> | |
4230 | </column> | |
4231 | ||
4232 | <group title="Controller Rate Limiting Statistics"> | |
4233 | <p> | |
4234 | These values report the effects of rate limiting. Their values are | |
4235 | relative to establishment of the most recent OpenFlow connection, | |
4236 | or since rate limiting was enabled, whichever happened more | |
4237 | recently. Each consists of two values, one with <code>TYPE</code> | |
4238 | replaced by <code>miss</code> for rate limiting flow table misses, | |
4239 | and the other with <code>TYPE</code> replaced by | |
4240 | <code>action</code> for rate limiting packets sent by OpenFlow | |
4241 | actions. | |
4242 | </p> | |
4243 | ||
4244 | <p> | |
4245 | These statistics are reported only when controller rate limiting is | |
4246 | enabled. | |
4247 | </p> | |
4248 | ||
4249 | <column name="status" key="packet-in-TYPE-bypassed" | |
4250 | type='{"type": "integer", "minInteger": 0}'> | |
4251 | Number of packets sent directly to the controller, without queuing, | |
4252 | because the rate did not exceed the configured maximum. | |
4253 | </column> | |
4254 | ||
4255 | <column name="status" key="packet-in-TYPE-queued" | |
4256 | type='{"type": "integer", "minInteger": 0}'> | |
4257 | Number of packets added to the queue to send later. | |
4258 | </column> | |
4259 | ||
4260 | <column name="status" key="packet-in-TYPE-dropped" | |
4261 | type='{"type": "integer", "minInteger": 0}'> | |
4262 | Number of packets added to the queue that were later dropped due to | |
4263 | overflow. This value is less than or equal to <ref column="status" | |
4264 | key="packet-in-TYPE-queued"/>. | |
4265 | </column> | |
4266 | ||
4267 | <column name="status" key="packet-in-TYPE-backlog" | |
4268 | type='{"type": "integer", "minInteger": 0}'> | |
4269 | Number of packets currently queued. The other statistics increase | |
4270 | monotonically, but this one fluctuates between 0 and the <ref | |
4271 | column="controller_burst_limit"/> as conditions change. | |
4272 | </column> | |
4273 | </group> | |
4274 | </group> | |
89365653 BP |
4275 | </group> |
4276 | ||
76ce9432 BP |
4277 | <group title="Additional In-Band Configuration"> |
4278 | <p>These values are considered only in in-band control mode (see | |
3fd8d445 | 4279 | <ref column="connection_mode"/>).</p> |
76ce9432 BP |
4280 | |
4281 | <p>When multiple controllers are configured on a single bridge, there | |
3fd8d445 BP |
4282 | should be only one set of unique values in these columns. If different |
4283 | values are set for these columns in different controllers, the effect | |
4284 | is unspecified.</p> | |
89365653 BP |
4285 | |
4286 | <column name="local_ip"> | |
76ce9432 BP |
4287 | The IP address to configure on the local port, |
4288 | e.g. <code>192.168.0.123</code>. If this value is unset, then | |
4289 | <ref column="local_netmask"/> and <ref column="local_gateway"/> are | |
4290 | ignored. | |
89365653 BP |
4291 | </column> |
4292 | ||
4293 | <column name="local_netmask"> | |
76ce9432 BP |
4294 | The IP netmask to configure on the local port, |
4295 | e.g. <code>255.255.255.0</code>. If <ref column="local_ip"/> is set | |
4296 | but this value is unset, then the default is chosen based on whether | |
4297 | the IP address is class A, B, or C. | |
4298 | </column> | |
4299 | ||
4300 | <column name="local_gateway"> | |
4301 | The IP address of the gateway to configure on the local port, as a | |
4302 | string, e.g. <code>192.168.0.1</code>. Leave this column unset if | |
4303 | this network has no gateway. | |
89365653 BP |
4304 | </column> |
4305 | </group> | |
13008eb3 | 4306 | |
bffc0589 AE |
4307 | <group title="Controller Status"> |
4308 | <column name="is_connected"> | |
4309 | <code>true</code> if currently connected to this controller, | |
4310 | <code>false</code> otherwise. | |
4311 | </column> | |
4312 | ||
f9e5e5b3 BP |
4313 | <column name="role" |
4314 | type='{"type": "string", "enum": ["set", ["other", "master", "slave"]]}'> | |
bffc0589 | 4315 | <p>The level of authority this controller has on the associated |
3fd8d445 | 4316 | bridge. Possible values are:</p> |
bffc0589 AE |
4317 | <dl> |
4318 | <dt><code>other</code></dt> | |
4319 | <dd>Allows the controller access to all OpenFlow features.</dd> | |
bffc0589 AE |
4320 | <dt><code>master</code></dt> |
4321 | <dd>Equivalent to <code>other</code>, except that there may be at | |
3fd8d445 BP |
4322 | most one master controller at a time. When a controller configures |
4323 | itself as <code>master</code>, any existing master is demoted to | |
9f90ed90 | 4324 | the <code>slave</code> role.</dd> |
bffc0589 AE |
4325 | <dt><code>slave</code></dt> |
4326 | <dd>Allows the controller read-only access to OpenFlow features. | |
3fd8d445 BP |
4327 | Attempts to modify the flow table will be rejected with an |
4328 | error. Slave controllers do not receive OFPT_PACKET_IN or | |
4329 | OFPT_FLOW_REMOVED messages, but they do receive OFPT_PORT_STATUS | |
4330 | messages.</dd> | |
bffc0589 AE |
4331 | </dl> |
4332 | </column> | |
4333 | ||
3fd8d445 BP |
4334 | <column name="status" key="last_error"> |
4335 | A human-readable description of the last error on the connection | |
4336 | to the controller; i.e. <code>strerror(errno)</code>. This key | |
4337 | will exist only if an error has occurred. | |
4338 | </column> | |
4339 | ||
f9e5e5b3 BP |
4340 | <column name="status" key="state" |
4341 | type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'> | |
3fd8d445 | 4342 | <p> |
f9e5e5b3 | 4343 | The state of the connection to the controller: |
3fd8d445 | 4344 | </p> |
bffc0589 | 4345 | <dl> |
3fd8d445 BP |
4346 | <dt><code>VOID</code></dt> |
4347 | <dd>Connection is disabled.</dd> | |
4348 | ||
4349 | <dt><code>BACKOFF</code></dt> | |
4350 | <dd>Attempting to reconnect at an increasing period.</dd> | |
4351 | ||
4352 | <dt><code>CONNECTING</code></dt> | |
4353 | <dd>Attempting to connect.</dd> | |
4354 | ||
4355 | <dt><code>ACTIVE</code></dt> | |
4356 | <dd>Connected, remote host responsive.</dd> | |
4357 | ||
4358 | <dt><code>IDLE</code></dt> | |
4359 | <dd>Connection is idle. Waiting for response to keep-alive.</dd> | |
bffc0589 | 4360 | </dl> |
3fd8d445 BP |
4361 | <p> |
4362 | These values may change in the future. They are provided only for | |
4363 | human consumption. | |
4364 | </p> | |
4365 | </column> | |
9cc6bf75 | 4366 | |
f9e5e5b3 BP |
4367 | <column name="status" key="sec_since_connect" |
4368 | type='{"type": "integer", "minInteger": 0}'> | |
3fd8d445 BP |
4369 | The amount of time since this controller last successfully connected to |
4370 | the switch (in seconds). Value is empty if controller has never | |
4371 | successfully connected. | |
bffc0589 | 4372 | </column> |
9cc6bf75 | 4373 | |
f9e5e5b3 BP |
4374 | <column name="status" key="sec_since_disconnect" |
4375 | type='{"type": "integer", "minInteger": 1}'> | |
3fd8d445 BP |
4376 | The amount of time since this controller last disconnected from |
4377 | the switch (in seconds). Value is empty if controller has never | |
4378 | disconnected. | |
4379 | </column> | |
4380 | </group> | |
4381 | ||
f125905c MM |
4382 | <group title="Connection Parameters"> |
4383 | <p> | |
4384 | Additional configuration for a connection between the controller | |
4385 | and the Open vSwitch. | |
4386 | </p> | |
4387 | ||
4388 | <column name="other_config" key="dscp" | |
039a8ccd | 4389 | type='{"type": "integer"}'> |
cea15768 EJ |
4390 | The Differentiated Service Code Point (DSCP) is specified using 6 bits |
4391 | in the Type of Service (TOS) field in the IP header. DSCP provides a | |
4392 | mechanism to classify the network traffic and provide Quality of | |
4393 | Service (QoS) on IP networks. | |
4394 | ||
4395 | The DSCP value specified here is used when establishing the connection | |
0442efd9 MM |
4396 | between the controller and the Open vSwitch. If no value is specified, |
4397 | a default value of 48 is chosen. Valid DSCP values must be in the | |
4398 | range 0 to 63. | |
f125905c MM |
4399 | </column> |
4400 | </group> | |
4401 | ||
4402 | ||
3fd8d445 BP |
4403 | <group title="Common Columns"> |
4404 | The overall purpose of these columns is described under <code>Common | |
4405 | Columns</code> at the beginning of this document. | |
4406 | ||
4407 | <column name="external_ids"/> | |
f125905c | 4408 | <column name="other_config"/> |
bffc0589 | 4409 | </group> |
89365653 BP |
4410 | </table> |
4411 | ||
94db5407 BP |
4412 | <table name="Manager" title="OVSDB management connection."> |
4413 | <p> | |
4414 | Configuration for a database connection to an Open vSwitch database | |
4415 | (OVSDB) client. | |
4416 | </p> | |
4417 | ||
4418 | <p> | |
4419 | This table primarily configures the Open vSwitch database | |
4420 | (<code>ovsdb-server</code>), not the Open vSwitch switch | |
4421 | (<code>ovs-vswitchd</code>). The switch does read the table to determine | |
4422 | what connections should be treated as in-band. | |
4423 | </p> | |
4424 | ||
4425 | <p> | |
4426 | The Open vSwitch database server can initiate and maintain active | |
4427 | connections to remote clients. It can also listen for database | |
4428 | connections. | |
4429 | </p> | |
4430 | ||
4431 | <group title="Core Features"> | |
4432 | <column name="target"> | |
4433 | <p>Connection method for managers.</p> | |
4434 | <p> | |
4435 | The following connection methods are currently supported: | |
4436 | </p> | |
4437 | <dl> | |
4438 | <dt><code>ssl:<var>ip</var></code>[<code>:<var>port</var></code>]</dt> | |
4439 | <dd> | |
4440 | <p> | |
efc295d2 JP |
4441 | The specified SSL <var>port</var> on the host at the given |
4442 | <var>ip</var>, which must be expressed as an IP address | |
4443 | (not a DNS name). The <ref table="Open_vSwitch" | |
4444 | column="ssl"/> column in the <ref table="Open_vSwitch"/> | |
4445 | table must point to a valid SSL configuration when this | |
4446 | form is used. | |
94db5407 BP |
4447 | </p> |
4448 | <p> | |
d4763d1d | 4449 | If <var>port</var> is not specified, it defaults to 6640. |
efc295d2 JP |
4450 | </p> |
4451 | <p> | |
4452 | SSL support is an optional feature that is not always | |
4453 | built as part of Open vSwitch. | |
94db5407 BP |
4454 | </p> |
4455 | </dd> | |
4456 | ||
4457 | <dt><code>tcp:<var>ip</var></code>[<code>:<var>port</var></code>]</dt> | |
4458 | <dd> | |
efc295d2 JP |
4459 | <p> |
4460 | The specified TCP <var>port</var> on the host at the given | |
e731d71b AS |
4461 | <var>ip</var>, which must be expressed as an IP address (not a |
4462 | DNS name), where <var>ip</var> can be IPv4 or IPv6 address. If | |
4463 | <var>ip</var> is an IPv6 address, wrap it in square brackets, | |
d4763d1d | 4464 | e.g. <code>tcp:[::1]:6640</code>. |
efc295d2 JP |
4465 | </p> |
4466 | <p> | |
d4763d1d | 4467 | If <var>port</var> is not specified, it defaults to 6640. |
efc295d2 | 4468 | </p> |
94db5407 BP |
4469 | </dd> |
4470 | <dt><code>pssl:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt> | |
4471 | <dd> | |
4472 | <p> | |
e731d71b AS |
4473 | Listens for SSL connections on the specified TCP <var>port</var>. |
4474 | Specify 0 for <var>port</var> to have the kernel automatically | |
4475 | choose an available port. If <var>ip</var>, which must be | |
4476 | expressed as an IP address (not a DNS name), is specified, then | |
4477 | connections are restricted to the specified local IP address | |
4478 | (either IPv4 or IPv6 address). If <var>ip</var> is an IPv6 | |
4479 | address, wrap in square brackets, | |
d4763d1d | 4480 | e.g. <code>pssl:6640:[::1]</code>. If <var>ip</var> is not |
e731d71b AS |
4481 | specified then it listens only on IPv4 (but not IPv6) addresses. |
4482 | The <ref table="Open_vSwitch" column="ssl"/> column in the <ref | |
94db5407 BP |
4483 | table="Open_vSwitch"/> table must point to a valid SSL |
4484 | configuration when this form is used. | |
4485 | </p> | |
efc295d2 | 4486 | <p> |
d4763d1d | 4487 | If <var>port</var> is not specified, it defaults to 6640. |
efc295d2 | 4488 | </p> |
94db5407 BP |
4489 | <p> |
4490 | SSL support is an optional feature that is not always built as | |
4491 | part of Open vSwitch. | |
4492 | </p> | |
4493 | </dd> | |
4494 | <dt><code>ptcp:</code>[<var>port</var>][<code>:<var>ip</var></code>]</dt> | |
4495 | <dd> | |
efc295d2 | 4496 | <p> |
e731d71b AS |
4497 | Listens for connections on the specified TCP <var>port</var>. |
4498 | Specify 0 for <var>port</var> to have the kernel automatically | |
4499 | choose an available port. If <var>ip</var>, which must be | |
4500 | expressed as an IP address (not a DNS name), is specified, then | |
4501 | connections are restricted to the specified local IP address | |
4502 | (either IPv4 or IPv6 address). If <var>ip</var> is an IPv6 | |
4503 | address, wrap it in square brackets, | |
d4763d1d | 4504 | e.g. <code>ptcp:6640:[::1]</code>. If <var>ip</var> is not |
e731d71b | 4505 | specified then it listens only on IPv4 addresses. |
efc295d2 JP |
4506 | </p> |
4507 | <p> | |
d4763d1d | 4508 | If <var>port</var> is not specified, it defaults to 6640. |
efc295d2 | 4509 | </p> |
94db5407 BP |
4510 | </dd> |
4511 | </dl> | |
4512 | <p>When multiple managers are configured, the <ref column="target"/> | |
4513 | values must be unique. Duplicate <ref column="target"/> values yield | |
4514 | unspecified results.</p> | |
4515 | </column> | |
4516 | ||
4517 | <column name="connection_mode"> | |
4518 | <p> | |
4519 | If it is specified, this setting must be one of the following strings | |
4520 | that describes how Open vSwitch contacts this OVSDB client over the | |
4521 | network: | |
4522 | </p> | |
299a244b | 4523 | |
94db5407 BP |
4524 | <dl> |
4525 | <dt><code>in-band</code></dt> | |
4526 | <dd> | |
4527 | In this mode, this connection's traffic travels over a bridge | |
4528 | managed by Open vSwitch. With this setting, Open vSwitch allows | |
4529 | traffic to and from the client regardless of the contents of the | |
4530 | OpenFlow flow table. (Otherwise, Open vSwitch would never be able | |
4531 | to connect to the client, because it did not have a flow to enable | |
4532 | it.) This is the most common connection mode because it is not | |
4533 | necessary to maintain two independent networks. | |
4534 | </dd> | |
4535 | <dt><code>out-of-band</code></dt> | |
4536 | <dd> | |
4537 | In this mode, the client's traffic uses a control network separate | |
4538 | from that managed by Open vSwitch, that is, Open vSwitch does not | |
4539 | use any of its own network devices to communicate with the client. | |
4540 | The control network must be configured separately, before or after | |
4541 | <code>ovs-vswitchd</code> is started. | |
4542 | </dd> | |
4543 | </dl> | |
4544 | ||
4545 | <p> | |
4546 | If not specified, the default is implementation-specific. | |
4547 | </p> | |
4548 | </column> | |
4549 | </group> | |
4550 | ||
4551 | <group title="Client Failure Detection and Handling"> | |
4552 | <column name="max_backoff"> | |
4553 | Maximum number of milliseconds to wait between connection attempts. | |
4554 | Default is implementation-specific. | |
4555 | </column> | |
4556 | ||
4557 | <column name="inactivity_probe"> | |
4558 | Maximum number of milliseconds of idle time on connection to the client | |
4559 | before sending an inactivity probe message. If Open vSwitch does not | |
4560 | communicate with the client for the specified number of seconds, it | |
4561 | will send a probe. If a response is not received for the same | |
4562 | additional amount of time, Open vSwitch assumes the connection has been | |
4563 | broken and attempts to reconnect. Default is implementation-specific. | |
2bb82bf0 | 4564 | A value of 0 disables inactivity probes. |
94db5407 BP |
4565 | </column> |
4566 | </group> | |
4567 | ||
0b3e7a8b | 4568 | <group title="Status"> |
600766e8 AZ |
4569 | <p> |
4570 | Key-value pair of <ref column="is_connected"/> is always updated. | |
4571 | Other key-value pairs in the status columns may be updated depends | |
4572 | on the <ref column="target"/> type. | |
4573 | </p> | |
4574 | ||
4575 | <p> | |
4576 | When <ref column="target"/> specifies a connection method that | |
4577 | listens for inbound connections (e.g. <code>ptcp:</code> or | |
4578 | <code>punix:</code>), both <ref column="n_connections"/> and | |
4579 | <ref column="is_connected"/> may also be updated while the | |
4580 | remaining key-value pairs are omitted. | |
4581 | </p> | |
4582 | ||
4583 | <p> | |
4584 | On the other hand, when <ref column="target"/> specifies an | |
4585 | outbound connection, all key-value pairs may be updated, except | |
4586 | the above-mentioned two key-value pairs associated with inbound | |
4587 | connection targets. They are omitted. | |
4588 | </p> | |
4589 | ||
4590 | <column name="is_connected"> | |
0b3e7a8b AE |
4591 | <code>true</code> if currently connected to this manager, |
4592 | <code>false</code> otherwise. | |
4593 | </column> | |
4594 | ||
3fd8d445 BP |
4595 | <column name="status" key="last_error"> |
4596 | A human-readable description of the last error on the connection | |
4597 | to the manager; i.e. <code>strerror(errno)</code>. This key | |
4598 | will exist only if an error has occurred. | |
4599 | </column> | |
4600 | ||
f9e5e5b3 BP |
4601 | <column name="status" key="state" |
4602 | type='{"type": "string", "enum": ["set", ["VOID", "BACKOFF", "CONNECTING", "ACTIVE", "IDLE"]]}'> | |
3fd8d445 | 4603 | <p> |
f9e5e5b3 | 4604 | The state of the connection to the manager: |
3fd8d445 | 4605 | </p> |
a11f6164 | 4606 | <dl> |
3fd8d445 BP |
4607 | <dt><code>VOID</code></dt> |
4608 | <dd>Connection is disabled.</dd> | |
4609 | ||
4610 | <dt><code>BACKOFF</code></dt> | |
4611 | <dd>Attempting to reconnect at an increasing period.</dd> | |
4612 | ||
4613 | <dt><code>CONNECTING</code></dt> | |
4614 | <dd>Attempting to connect.</dd> | |
4615 | ||
4616 | <dt><code>ACTIVE</code></dt> | |
4617 | <dd>Connected, remote host responsive.</dd> | |
4618 | ||
4619 | <dt><code>IDLE</code></dt> | |
4620 | <dd>Connection is idle. Waiting for response to keep-alive.</dd> | |
a11f6164 | 4621 | </dl> |
3fd8d445 BP |
4622 | <p> |
4623 | These values may change in the future. They are provided only for | |
4624 | human consumption. | |
4625 | </p> | |
4626 | </column> | |
4627 | ||
f9e5e5b3 BP |
4628 | <column name="status" key="sec_since_connect" |
4629 | type='{"type": "integer", "minInteger": 0}'> | |
3fd8d445 BP |
4630 | The amount of time since this manager last successfully connected |
4631 | to the database (in seconds). Value is empty if manager has never | |
4632 | successfully connected. | |
4633 | </column> | |
4634 | ||
f9e5e5b3 BP |
4635 | <column name="status" key="sec_since_disconnect" |
4636 | type='{"type": "integer", "minInteger": 0}'> | |
3fd8d445 BP |
4637 | The amount of time since this manager last disconnected from the |
4638 | database (in seconds). Value is empty if manager has never | |
4639 | disconnected. | |
4640 | </column> | |
4641 | ||
4642 | <column name="status" key="locks_held"> | |
4643 | Space-separated list of the names of OVSDB locks that the connection | |
4644 | holds. Omitted if the connection does not hold any locks. | |
4645 | </column> | |
4646 | ||
4647 | <column name="status" key="locks_waiting"> | |
4648 | Space-separated list of the names of OVSDB locks that the connection is | |
4649 | currently waiting to acquire. Omitted if the connection is not waiting | |
4650 | for any locks. | |
4651 | </column> | |
4652 | ||
4653 | <column name="status" key="locks_lost"> | |
4654 | Space-separated list of the names of OVSDB locks that the connection | |
4655 | has had stolen by another OVSDB client. Omitted if no locks have been | |
4656 | stolen from this connection. | |
4657 | </column> | |
4658 | ||
f9e5e5b3 BP |
4659 | <column name="status" key="n_connections" |
4660 | type='{"type": "integer", "minInteger": 2}'> | |
600766e8 AZ |
4661 | When <ref column="target"/> specifies a connection method that |
4662 | listens for inbound connections (e.g. <code>ptcp:</code> or | |
4663 | <code>pssl:</code>) and more than one connection is actually active, | |
4664 | the value is the number of active connections. Otherwise, this | |
4665 | key-value pair is omitted. | |
0b3e7a8b | 4666 | </column> |
798e1352 BP |
4667 | |
4668 | <column name="status" key="bound_port" type='{"type": "integer"}'> | |
039a8ccd BP |
4669 | When <ref column="target"/> is <code>ptcp:</code> or |
4670 | <code>pssl:</code>, this is the TCP port on which the OVSDB server is | |
6c75e208 | 4671 | listening. (This is particularly useful when <ref |
039a8ccd BP |
4672 | column="target"/> specifies a port of 0, allowing the kernel to |
4673 | choose any available port.) | |
798e1352 | 4674 | </column> |
0b3e7a8b | 4675 | </group> |
3fd8d445 | 4676 | |
f125905c MM |
4677 | <group title="Connection Parameters"> |
4678 | <p> | |
4679 | Additional configuration for a connection between the manager | |
4680 | and the Open vSwitch Database. | |
4681 | </p> | |
4682 | ||
4683 | <column name="other_config" key="dscp" | |
039a8ccd | 4684 | type='{"type": "integer"}'> |
cea15768 EJ |
4685 | The Differentiated Service Code Point (DSCP) is specified using 6 bits |
4686 | in the Type of Service (TOS) field in the IP header. DSCP provides a | |
4687 | mechanism to classify the network traffic and provide Quality of | |
4688 | Service (QoS) on IP networks. | |
4689 | ||
4690 | The DSCP value specified here is used when establishing the connection | |
0442efd9 MM |
4691 | between the manager and the Open vSwitch. If no value is specified, a |
4692 | default value of 48 is chosen. Valid DSCP values must be in the range | |
4693 | 0 to 63. | |
f125905c MM |
4694 | </column> |
4695 | </group> | |
4696 | ||
3fd8d445 BP |
4697 | <group title="Common Columns"> |
4698 | The overall purpose of these columns is described under <code>Common | |
4699 | Columns</code> at the beginning of this document. | |
4700 | ||
4701 | <column name="external_ids"/> | |
f125905c | 4702 | <column name="other_config"/> |
3fd8d445 | 4703 | </group> |
94db5407 BP |
4704 | </table> |
4705 | ||
89365653 BP |
4706 | <table name="NetFlow"> |
4707 | A NetFlow target. NetFlow is a protocol that exports a number of | |
4708 | details about terminating IP flows, such as the principals involved | |
4709 | and duration. | |
4710 | ||
4711 | <column name="targets"> | |
4712 | NetFlow targets in the form | |
4713 | <code><var>ip</var>:<var>port</var></code>. The <var>ip</var> | |
4714 | must be specified numerically, not as a DNS name. | |
4715 | </column> | |
4716 | ||
4717 | <column name="engine_id"> | |
4718 | Engine ID to use in NetFlow messages. Defaults to datapath index | |
4719 | if not specified. | |
4720 | </column> | |
4721 | ||
4722 | <column name="engine_type"> | |
4723 | Engine type to use in NetFlow messages. Defaults to datapath | |
4724 | index if not specified. | |
4725 | </column> | |
4726 | ||
4727 | <column name="active_timeout"> | |
a70f8b11 | 4728 | <p> |
039a8ccd BP |
4729 | The interval at which NetFlow records are sent for flows that |
4730 | are still active, in seconds. A value of <code>0</code> | |
4731 | requests the default timeout (currently 600 seconds); a value | |
4732 | of <code>-1</code> disables active timeouts. | |
a70f8b11 BP |
4733 | </p> |
4734 | ||
4735 | <p> | |
039a8ccd BP |
4736 | The NetFlow passive timeout, for flows that become inactive, |
4737 | is not configurable. It will vary depending on the Open | |
4738 | vSwitch version, the forms and contents of the OpenFlow flow | |
4739 | tables, CPU and memory usage, and network activity. A typical | |
4740 | passive timeout is about a second. | |
a70f8b11 | 4741 | </p> |
89365653 BP |
4742 | </column> |
4743 | ||
4744 | <column name="add_id_to_interface"> | |
4745 | <p>If this column's value is <code>false</code>, the ingress and egress | |
3fd8d445 BP |
4746 | interface fields of NetFlow flow records are derived from OpenFlow port |
4747 | numbers. When it is <code>true</code>, the 7 most significant bits of | |
4748 | these fields will be replaced by the least significant 7 bits of the | |
4749 | engine id. This is useful because many NetFlow collectors do not | |
4750 | expect multiple switches to be sending messages from the same host, so | |
4751 | they do not store the engine information which could be used to | |
4752 | disambiguate the traffic.</p> | |
89365653 BP |
4753 | <p>When this option is enabled, a maximum of 508 ports are supported.</p> |
4754 | </column> | |
13008eb3 | 4755 | |
3fd8d445 BP |
4756 | <group title="Common Columns"> |
4757 | The overall purpose of these columns is described under <code>Common | |
4758 | Columns</code> at the beginning of this document. | |
4759 | ||
4760 | <column name="external_ids"/> | |
4761 | </group> | |
89365653 BP |
4762 | </table> |
4763 | ||
4764 | <table name="SSL"> | |
4765 | SSL configuration for an Open_vSwitch. | |
4766 | ||
4767 | <column name="private_key"> | |
4768 | Name of a PEM file containing the private key used as the switch's | |
4769 | identity for SSL connections to the controller. | |
4770 | </column> | |
4771 | ||
4772 | <column name="certificate"> | |
4773 | Name of a PEM file containing a certificate, signed by the | |
4774 | certificate authority (CA) used by the controller and manager, | |
4775 | that certifies the switch's private key, identifying a trustworthy | |
4776 | switch. | |
4777 | </column> | |
4778 | ||
4779 | <column name="ca_cert"> | |
4780 | Name of a PEM file containing the CA certificate used to verify | |
4781 | that the switch is connected to a trustworthy controller. | |
4782 | </column> | |
4783 | ||
4784 | <column name="bootstrap_ca_cert"> | |
4785 | If set to <code>true</code>, then Open vSwitch will attempt to | |
4786 | obtain the CA certificate from the controller on its first SSL | |
4787 | connection and save it to the named PEM file. If it is successful, | |
4788 | it will immediately drop the connection and reconnect, and from then | |
4789 | on all SSL connections must be authenticated by a certificate signed | |
4790 | by the CA certificate thus obtained. <em>This option exposes the | |
3fd8d445 BP |
4791 | SSL connection to a man-in-the-middle attack obtaining the initial |
4792 | CA certificate.</em> It may still be useful for bootstrapping. | |
89365653 | 4793 | </column> |
13008eb3 | 4794 | |
3fd8d445 BP |
4795 | <group title="Common Columns"> |
4796 | The overall purpose of these columns is described under <code>Common | |
4797 | Columns</code> at the beginning of this document. | |
4798 | ||
4799 | <column name="external_ids"/> | |
4800 | </group> | |
89365653 BP |
4801 | </table> |
4802 | ||
4803 | <table name="sFlow"> | |
29089a54 RL |
4804 | <p>A set of sFlow(R) targets. sFlow is a protocol for remote |
4805 | monitoring of switches.</p> | |
89365653 BP |
4806 | |
4807 | <column name="agent"> | |
e723ff43 | 4808 | Name of the network device whose IP address should be reported as the |
733adf2a LG |
4809 | ``agent address'' to collectors. If not specified, the agent device is |
4810 | figured from the first target address and the routing table. If the | |
4811 | routing table does not contain a route to the target, the IP address | |
e723ff43 BP |
4812 | defaults to the <ref table="Controller" column="local_ip"/> in the |
4813 | collector's <ref table="Controller"/>. If an agent IP address cannot be | |
733adf2a | 4814 | determined any of these ways, sFlow is disabled. |
89365653 BP |
4815 | </column> |
4816 | ||
4817 | <column name="header"> | |
4818 | Number of bytes of a sampled packet to send to the collector. | |
4819 | If not specified, the default is 128 bytes. | |
4820 | </column> | |
4821 | ||
4822 | <column name="polling"> | |
4823 | Polling rate in seconds to send port statistics to the collector. | |
4824 | If not specified, defaults to 30 seconds. | |
4825 | </column> | |
4826 | ||
4827 | <column name="sampling"> | |
4828 | Rate at which packets should be sampled and sent to the collector. | |
4829 | If not specified, defaults to 400, which means one out of 400 | |
4830 | packets, on average, will be sent to the collector. | |
4831 | </column> | |
4832 | ||
4833 | <column name="targets"> | |
4834 | sFlow targets in the form | |
4835 | <code><var>ip</var>:<var>port</var></code>. | |
4836 | </column> | |
13008eb3 | 4837 | |
3fd8d445 BP |
4838 | <group title="Common Columns"> |
4839 | The overall purpose of these columns is described under <code>Common | |
4840 | Columns</code> at the beginning of this document. | |
4841 | ||
4842 | <column name="external_ids"/> | |
4843 | </group> | |
89365653 | 4844 | </table> |
c1c9c9c4 | 4845 | |
29089a54 | 4846 | <table name="IPFIX"> |
99ec8f05 | 4847 | <p>Configuration for sending packets to IPFIX collectors.</p> |
29089a54 | 4848 | |
99ec8f05 BP |
4849 | <p> |
4850 | IPFIX is a protocol that exports a number of details about flows. The | |
4851 | IPFIX implementation in Open vSwitch samples packets at a configurable | |
4852 | rate, extracts flow information from those packets, optionally caches and | |
4853 | aggregates the flow information, and sends the result to one or more | |
4854 | collectors. | |
4855 | </p> | |
29089a54 | 4856 | |
99ec8f05 BP |
4857 | <p> |
4858 | IPFIX in Open vSwitch can be configured two different ways: | |
4859 | </p> | |
29089a54 | 4860 | |
99ec8f05 BP |
4861 | <ul> |
4862 | <li> | |
4863 | With <em>per-bridge sampling</em>, Open vSwitch performs IPFIX sampling | |
4864 | automatically on all packets that pass through a bridge. To configure | |
4865 | per-bridge sampling, create an <ref table="IPFIX"/> record and point a | |
4866 | <ref table="Bridge"/> table's <ref table="Bridge" column="ipfix"/> | |
4867 | column to it. The <ref table="Flow_Sample_Collector_Set"/> table is | |
4868 | not used for per-bridge sampling. | |
4869 | </li> | |
4870 | ||
4871 | <li> | |
4872 | <p> | |
4873 | With <em>flow-based sampling</em>, <code>sample</code> actions in the | |
4874 | OpenFlow flow table drive IPFIX sampling. See | |
4875 | <code>ovs-ofctl</code>(8) for a description of the | |
4876 | <code>sample</code> action. | |
4877 | </p> | |
4878 | ||
4879 | <p> | |
4880 | Flow-based sampling also requires database configuration: create a | |
4881 | <ref table="IPFIX"/> record that describes the IPFIX configuration | |
4882 | and a <ref table="Flow_Sample_Collector_Set"/> record that points to | |
4883 | the <ref table="Bridge"/> whose flow table holds the | |
4884 | <code>sample</code> actions and to <ref table="IPFIX"/> record. The | |
4885 | <ref table="Bridge" column="ipfix"/> in the <ref table="Bridge"/> | |
4886 | table is not used for flow-based sampling. | |
4887 | </p> | |
4888 | </li> | |
4889 | </ul> | |
29089a54 | 4890 | |
99ec8f05 BP |
4891 | <column name="targets"> |
4892 | IPFIX target collectors in the form | |
4893 | <code><var>ip</var>:<var>port</var></code>. | |
29089a54 RL |
4894 | </column> |
4895 | ||
978427a5 RL |
4896 | <column name="cache_active_timeout"> |
4897 | The maximum period in seconds for which an IPFIX flow record is | |
4898 | cached and aggregated before being sent. If not specified, | |
4899 | defaults to 0. If 0, caching is disabled. | |
4900 | </column> | |
4901 | ||
4902 | <column name="cache_max_flows"> | |
4903 | The maximum number of IPFIX flow records that can be cached at a | |
4904 | time. If not specified, defaults to 0. If 0, caching is | |
4905 | disabled. | |
4906 | </column> | |
4907 | ||
f69f713b BY |
4908 | <column name="other_config" key="enable-tunnel-sampling" |
4909 | type='{"type": "boolean"}'> | |
4910 | <p> | |
4911 | Set to <code>true</code> to enable sampling and reporting tunnel | |
4912 | header 7-tuples in IPFIX flow records. Tunnel sampling is enabled | |
4913 | by default. | |
4914 | </p> | |
4915 | ||
4916 | <p> | |
4917 | The following enterprise entities report the sampled tunnel info: | |
4918 | </p> | |
4919 | ||
4920 | <dl> | |
4921 | <dt>tunnelType:</dt> | |
4922 | <dd> | |
4923 | <p>ID: 891, and enterprise ID 6876 (VMware).</p> | |
4924 | <p>type: unsigned 8-bit integer.</p> | |
4925 | <p>data type semantics: identifier.</p> | |
4926 | <p>description: Identifier of the layer 2 network overlay network | |
2b02d770 | 4927 | encapsulation type: 0x01 VxLAN, 0x02 GRE, 0x03 LISP, 0x07 GENEVE.</p> |
f69f713b BY |
4928 | </dd> |
4929 | <dt>tunnelKey:</dt> | |
4930 | <dd> | |
4931 | <p>ID: 892, and enterprise ID 6876 (VMware).</p> | |
4932 | <p>type: variable-length octetarray.</p> | |
4933 | <p>data type semantics: identifier.</p> | |
4934 | <p>description: Key which is used for identifying an individual | |
4935 | traffic flow within a VxLAN (24-bit VNI), GENEVE (24-bit VNI), | |
4936 | GRE (32-bit key), or LISP (24-bit instance ID) tunnel. The | |
4937 | key is encoded in this octetarray as a 3-, 4-, or 8-byte integer | |
4938 | ID in network byte order.</p> | |
4939 | </dd> | |
4940 | <dt>tunnelSourceIPv4Address:</dt> | |
4941 | <dd> | |
4942 | <p>ID: 893, and enterprise ID 6876 (VMware).</p> | |
4943 | <p>type: unsigned 32-bit integer.</p> | |
4944 | <p>data type semantics: identifier.</p> | |
4945 | <p>description: The IPv4 source address in the tunnel IP packet | |
4946 | header.</p> | |
4947 | </dd> | |
4948 | <dt>tunnelDestinationIPv4Address:</dt> | |
4949 | <dd> | |
4950 | <p>ID: 894, and enterprise ID 6876 (VMware).</p> | |
4951 | <p>type: unsigned 32-bit integer.</p> | |
4952 | <p>data type semantics: identifier.</p> | |
4953 | <p>description: The IPv4 destination address in the tunnel IP | |
4954 | packet header.</p> | |
4955 | </dd> | |
4956 | <dt>tunnelProtocolIdentifier:</dt> | |
4957 | <dd> | |
4958 | <p>ID: 895, and enterprise ID 6876 (VMware).</p> | |
4959 | <p>type: unsigned 8-bit integer.</p> | |
4960 | <p>data type semantics: identifier.</p> | |
4961 | <p>description: The value of the protocol number in the tunnel | |
4962 | IP packet header. The protocol number identifies the tunnel IP | |
4963 | packet payload type.</p> | |
4964 | </dd> | |
4965 | <dt>tunnelSourceTransportPort:</dt> | |
4966 | <dd> | |
4967 | <p>ID: 896, and enterprise ID 6876 (VMware).</p> | |
4968 | <p>type: unsigned 16-bit integer.</p> | |
4969 | <p>data type semantics: identifier.</p> | |
4970 | <p>description: The source port identifier in the tunnel transport | |
4971 | header. For the transport protocols UDP, TCP, and SCTP, this is | |
4972 | the source port number given in the respective header.</p> | |
4973 | </dd> | |
4974 | <dt>tunnelDestinationTransportPort:</dt> | |
4975 | <dd> | |
4976 | <p>ID: 897, and enterprise ID 6876 (VMware).</p> | |
4977 | <p>type: unsigned 16-bit integer.</p> | |
4978 | <p>data type semantics: identifier.</p> | |
4979 | <p>description: The destination port identifier in the tunnel | |
4980 | transport header. For the transport protocols UDP, TCP, and SCTP, | |
4981 | this is the destination port number given in the respective header. | |
4982 | </p> | |
4983 | </dd> | |
4984 | </dl> | |
4985 | ||
4986 | <p> | |
4987 | Before Open vSwitch 2.5.90, <ref column="other_config" | |
4988 | key="enable-tunnel-sampling"/> was only supported with per-bridge | |
4989 | sampling, and ignored otherwise. Open vSwitch 2.5.90 and later support | |
4990 | <ref column="other_config" key="enable-tunnel-sampling"/> for | |
4991 | per-bridge and per-flow sampling. | |
4992 | </p> | |
4993 | </column> | |
4994 | ||
c97320eb WZ |
4995 | <column name="other_config" key="virtual_obs_id" |
4996 | type='{"type": "string"}'> | |
4997 | <p> | |
4998 | A string that accompanies each IPFIX flow record. Its intended use is | |
4999 | for the ``virtual observation ID,'' an identifier of a virtual | |
5000 | observation point that is locally unique in a virtual network. It | |
5001 | describes a location in the virtual network where IP packets can be | |
5002 | observed. The maximum length is 254 bytes. If not specified, the | |
5003 | field is omitted from the IPFIX flow record. | |
5004 | </p> | |
5005 | ||
5006 | <p> | |
5007 | The following enterprise entity reports the specified virtual | |
5008 | observation ID: | |
5009 | </p> | |
5010 | ||
5011 | <dl> | |
5012 | <dt>virtualObsID:</dt> | |
5013 | <dd> | |
5014 | <p>ID: 898, and enterprise ID 6876 (VMware).</p> | |
5015 | <p>type: variable-length string.</p> | |
5016 | <p>data type semantics: identifier.</p> | |
5017 | <p>description: A virtual observation domain ID that is locally | |
5018 | unique in a virtual network. | |
5019 | </p> | |
5020 | </dd> | |
5021 | </dl> | |
5022 | ||
5023 | <p> | |
5024 | This feature was introduced in Open vSwitch 2.5.90. | |
5025 | </p> | |
5026 | </column> | |
5027 | ||
99ec8f05 BP |
5028 | <group title="Per-Bridge Sampling"> |
5029 | <p> | |
5030 | These values affect only per-bridge sampling. See above for a | |
5031 | description of the differences between per-bridge and flow-based | |
5032 | sampling. | |
5033 | </p> | |
8b7ea2d4 | 5034 | |
99ec8f05 BP |
5035 | <column name="sampling"> |
5036 | The rate at which packets should be sampled and sent to each target | |
5037 | collector. If not specified, defaults to 400, which means one out of | |
5038 | 400 packets, on average, will be sent to each target collector. | |
5039 | </column> | |
8b7ea2d4 | 5040 | |
99ec8f05 BP |
5041 | <column name="obs_domain_id"> |
5042 | The IPFIX Observation Domain ID sent in each IPFIX packet. If not | |
5043 | specified, defaults to 0. | |
5044 | </column> | |
5045 | ||
5046 | <column name="obs_point_id"> | |
5047 | The IPFIX Observation Point ID sent in each IPFIX flow record. If not | |
5048 | specified, defaults to 0. | |
5049 | </column> | |
5050 | ||
99ec8f05 BP |
5051 | <column name="other_config" key="enable-input-sampling" |
5052 | type='{"type": "boolean"}'> | |
5053 | By default, Open vSwitch samples and reports flows at bridge port input | |
5054 | in IPFIX flow records. Set this column to <code>false</code> to | |
5055 | disable input sampling. | |
5056 | </column> | |
5057 | ||
5058 | <column name="other_config" key="enable-output-sampling" | |
5059 | type='{"type": "boolean"}'> | |
5060 | By default, Open vSwitch samples and reports flows at bridge port | |
5061 | output in IPFIX flow records. Set this column to <code>false</code> to | |
5062 | disable output sampling. | |
5063 | </column> | |
5064 | </group> | |
8b7ea2d4 | 5065 | |
29089a54 RL |
5066 | <group title="Common Columns"> |
5067 | The overall purpose of these columns is described under <code>Common | |
5068 | Columns</code> at the beginning of this document. | |
5069 | ||
5070 | <column name="external_ids"/> | |
5071 | </group> | |
5072 | </table> | |
5073 | ||
5074 | <table name="Flow_Sample_Collector_Set"> | |
99ec8f05 BP |
5075 | <p> |
5076 | A set of IPFIX collectors of packet samples generated by OpenFlow | |
5077 | <code>sample</code> actions. This table is used only for IPFIX | |
5078 | flow-based sampling, not for per-bridge sampling (see the <ref | |
5079 | table="IPFIX"/> table for a description of the two forms). | |
5080 | </p> | |
29089a54 RL |
5081 | |
5082 | <column name="id"> | |
5083 | The ID of this collector set, unique among the bridge's | |
5084 | collector sets, to be used as the <code>collector_set_id</code> | |
5085 | in OpenFlow <code>sample</code> actions. | |
5086 | </column> | |
5087 | ||
5088 | <column name="bridge"> | |
5089 | The bridge into which OpenFlow <code>sample</code> actions can | |
5090 | be added to send packet samples to this set of IPFIX collectors. | |
5091 | </column> | |
5092 | ||
5093 | <column name="ipfix"> | |
5094 | Configuration of the set of IPFIX collectors to send one flow | |
5095 | record per sampled packet to. | |
5096 | </column> | |
5097 | ||
5098 | <group title="Common Columns"> | |
5099 | The overall purpose of these columns is described under <code>Common | |
5100 | Columns</code> at the beginning of this document. | |
5101 | ||
5102 | <column name="external_ids"/> | |
5103 | </group> | |
5104 | </table> | |
5105 | ||
99eef98b | 5106 | <table name="AutoAttach"> |
039a8ccd BP |
5107 | <p> |
5108 | Auto Attach configuration within a bridge. The IETF Auto-Attach SPBM | |
5109 | draft standard describes a compact method of using IEEE 802.1AB Link | |
5110 | Layer Discovery Protocol (LLDP) together with a IEEE 802.1aq Shortest | |
5111 | Path Bridging (SPB) network to automatically attach network devices | |
5112 | to individual services in a SPB network. The intent here is to allow | |
5113 | network applications and devices using OVS to be able to easily take | |
5114 | advantage of features offered by industry standard SPB networks. | |
5115 | </p> | |
5116 | ||
5117 | <p> | |
5118 | Auto Attach (AA) uses LLDP to communicate between a directly connected | |
5119 | Auto Attach Client (AAC) and Auto Attach Server (AAS). The LLDP protocol | |
5120 | is extended to add two new Type-Length-Value tuples (TLVs). The first | |
5121 | new TLV supports the ongoing discovery of directly connected AA | |
5122 | correspondents. Auto Attach operates by regularly transmitting AA | |
5123 | discovery TLVs between the AA client and AA server. By exchanging these | |
5124 | discovery messages, both the AAC and AAS learn the system name and | |
5125 | system description of their peer. In the OVS context, OVS operates as | |
5126 | the AA client and the AA server resides on a switch at the edge of the | |
5127 | SPB network. | |
5128 | </p> | |
5129 | ||
5130 | <p> | |
5131 | Once AA discovery has been completed the AAC then uses the second new TLV | |
5132 | to deliver identifier mappings from the AAC to the AAS. A primary feature | |
5133 | of Auto Attach is to facilitate the mapping of VLANs defined outside the | |
5134 | SPB network onto service ids (ISIDs) defined within the SPM network. By | |
5135 | doing so individual external VLANs can be mapped onto specific SPB | |
5136 | network services. These VLAN id to ISID mappings can be configured and | |
5137 | managed locally using new options added to the ovs-vsctl command. | |
5138 | </p> | |
5139 | ||
5140 | <p> | |
5141 | The Auto Attach OVS feature does not provide a full implementation of | |
5142 | the LLDP protocol. Support for the mandatory TLVs as defined by the LLDP | |
5143 | standard and support for the AA TLV extensions is provided. LLDP | |
5144 | protocol support in OVS can be enabled or disabled on a port by port | |
5145 | basis. LLDP support is disabled by default. | |
5146 | </p> | |
99eef98b DF |
5147 | |
5148 | <column name="system_name"> | |
5149 | The system_name string is exported in LLDP messages. It should uniquely | |
5150 | identify the bridge in the network. | |
5151 | </column> | |
5152 | ||
5153 | <column name="system_description"> | |
5154 | The system_description string is exported in LLDP messages. It should | |
5155 | describe the type of software and hardware. | |
5156 | </column> | |
5157 | ||
5158 | <column name="mappings"> | |
039a8ccd BP |
5159 | A mapping from SPB network Individual Service Identifier (ISID) to VLAN |
5160 | id. | |
99eef98b DF |
5161 | </column> |
5162 | </table> | |
89365653 | 5163 | </database> |