1 .TH dpctl 8 "May 2008" "OpenFlow" "OpenFlow Manual"
4 dpctl \- administer OpenFlow datapaths
8 [\fIoptions\fR] \fIcommand \fR[\fIswitch\fR] [\fIargs\fR&...]
13 program is a command line tool for monitoring and administering OpenFlow
14 datapaths. It is able to show the current state of a datapath,
15 including features, configuration, and tables entries. When using the
16 OpenFlow kernel module,
18 is used to add, delete, modify, and monitor datapaths.
20 Most \fBdpctl\fR commands take an argument that specifies the
21 method for connecting to an OpenFlow switch. The following connection
22 methods are supported:
26 The local Netlink datapath numbered \fIdp_idx\fR. This form requires
27 that the local host has the OpenFlow kernel module for Linux loaded.
30 \fBssl:\fIhost\fR[\fB:\fIport\fR]
31 The specified SSL \fIport\fR (default: 976) on the given remote
32 \fIhost\fR. The \fB--private-key\fR, \fB--certificate\fR, and
33 \fB--ca-cert\fR options are mandatory when this form is used.
36 \fBtcp:\fIhost\fR[\fB:\fIport\fR]
37 The specified TCP \fIport\fR (default: 975) on the given remote
42 The Unix domain server socket named \fIfile\fR.
46 With the \fBdpctl\fR program, datapaths running in the kernel can be
47 created, deleted, and modified. A single machine may
48 host up to 32 datapaths (numbered 0 to 31). In most situations,
49 a machine hosts only one datapath.
51 A newly created datapath is not associated with any of the
52 host's network devices thus does not process any incoming
53 traffic. To intercept and process traffic on a given network device, the
54 network device must be explicitly added to a datapath through the
57 The following commands manage local datapaths.
60 \fBadddp nl:\fIdp_idx\fR
61 Creates datapath numbered \fIdp_idx\fR on the local host. This will
62 fail if \fIdp_idx\fR is not in the range 0 to 31, or if the datapath
63 with that number already exists on the host.
66 \fBdeldp nl:\fIdp_idx\fR
67 Deletes datapath \fIdp_idx\fR on the local host. \fIdp_idx\fR must be
68 an existing datapath. All of a datapath's network devices must be
69 explicitly removed before the datapath can be deleted (see \fBdelif\fR
73 \fBaddif nl:\fIdp_idx netdev\fR...
74 Adds each \fInetdev\fR to the list of network devices datapath
75 \fIdp_idx\fR monitors, where \fIdp_idx\fR is the ID of an existing
76 datapath, and \fInetdev\fR is the name of one of the host's
77 network devices, e.g. \fBeth0\fR. Once a network device has been added
78 to a datapath, the datapath has complete ownership of the network device's
79 traffic and the network device appears silent to the rest of the system.
82 \fBdelif nl:\fIdp_idx netdev\fR...
83 Removes each \fInetdev\fR from the list of network devices datapath
84 \fIdp_idx\fR monitors.
87 The following commands can be apply to OpenFlow switches regardless of
88 the connection method.
92 Prints to the console information on datapath \fIswitch\fR including
93 information on its flow tables and ports.
96 \fBstatus \fIswitch\fR [\fIkey\fR]
97 Prints to the console a series of key-value pairs that report the
98 status of \fIswitch\fR. If \fIkey\fR is specified, only the key-value
99 pairs whose key names begin with \fIkey\fR are printed. If \fIkey\fR is
100 omitted, all key-value pairs are printed.
102 (In the OpenFlow reference implementation, the \fBstatus\fR command is
103 implemented in \fBsecchan\fR(8), not in the kernel module, so the
104 \fBnl:\fIdp_idx\fR connection method should not be used with this
105 command. Instead, specify \fB-l\fR or \fB--listen\fR on the
106 \fBsecchan\fR command line and tell \fBdpctl\fR to use the connection
107 method specified there.)
110 \fBdump-tables \fIswitch\fR
111 Prints to the console statistics for each of the flow tables used by
112 datapath \fIswitch\fR.
115 \fBdump-ports \fIswitch\fR
116 Prints to the console statistics for each of the network devices
117 associated with datapath \fIswitch\fR.
120 \fBmod-port \fIswitch\fR \fInetdev\fR \fIaction\fR
121 Modify characteristics of an interface monitored by \fIswitch\fR.
122 \fInetdev\fR can be referred to by its OpenFlow assigned port number or
123 the device name, e.g. \fBeth0\fR. The \fIaction\fR may be any one of the
128 Enables the interface. This is equivalent to ``ifconfig up'' on a Unix
132 Disables the interface. This is equivalent to ``ifconfig down'' on a Unix
136 When a \fIflood\fR action is specified, traffic will be sent out this
137 interface. This is the default posture for monitored ports.
140 When a \fIflood\fR action is specified, traffic will not be sent out
141 this interface. This is primarily useful to prevent loops when a
142 spanning tree protocol is not in use.
147 \fBdump-flows \fIswitch \fR[\fIflows\fR]
148 Prints to the console all flow entries in datapath \fIswitch\fR's
149 tables that match \fIflows\fR. If \fIflows\fR is omitted, all flows
150 in the datapath are retrieved. See \fBFLOW SYNTAX\fR, below, for the
151 syntax of \fIflows\fR.
154 \fBdump-aggregate \fIswitch \fR[\fIflows\fR]
155 Prints to the console aggregate statistics for flows in datapath
156 \fSWITCH\fR's tables that match \fIflows\fR. If \fIflows\fR is omitted,
157 the statistics are aggregated across all flows in the datapath's flow
158 tables. See \fBFLOW SYNTAX\fR, below, for the syntax of \fIflows\fR.
161 \fBadd-flow \fIswitch flow\fR
162 Add the flow entry as described by \fIflow\fR to the datapath \fIswitch\fR's
163 tables. The flow entry is in the format described in \fBFLOW SYNTAX\fR,
167 \fBadd-flows \fIswitch file\fR
168 Add flow entries as described in \fIfile\fR to the datapath \fIswitch\fR's
169 tables. Each line in \fIfile\fR is a flow entry in the format
170 described in \fBFLOW SYNTAX\fR, below.
173 \fBdel-flows \fIswitch \fR[\fIflow\fR]
174 Deletes entries from the datapath \fIswitch\fR's tables that match
175 \fIflow\fR. If \fIflow\fR is omitted, all flows in the datapath's
176 tables are removed. See \fBFLOW SYNTAX\fR, below, for the syntax of
180 \fBmonitor \fIswitch\fR
181 Connects to \fIswitch\fR and prints to the console all OpenFlow
182 messages received. Usually, \fIswitch\fR should specify a connection
183 named on \fBsecchan\fR(8)'s \fB-m\fR or \fB--monitor\fR command line
184 option, in which the messages printed will be all those sent or
185 received by \fBsecchan\fR to or from the kernel datapath module. A
186 \fIswitch\fR of the form \fBnl:\fIdp_idx\fR will print all
187 asynchronously generated OpenFlow messages (such as packet-in
188 messages), but it will not print any messages sent to the kernel by
189 \fBsecchan\fR and other processes, nor will it print replies sent by
190 the kernel in response to those messages.
193 The following commands can be used regardless of the connection
194 method. They apply to OpenFlow switches and controllers.
198 Connects to \fIvconn\fR and sends a single OpenFlow echo-request
199 packet and waits for the response. With the \fB-t\fR or
200 \fB--timeout\fR option, this command can test whether an OpenFlow
201 switch or controller is up and running.
204 \fBping \fIvconn \fR[\fIn\fR]
205 Sends a series of 10 echo request packets to \fIvconn\fR and times
206 each reply. The echo request packets consist of an OpenFlow header
207 plus \fIn\fR bytes (default: 64) of randomly generated payload. This
208 measures the latency of individual requests.
211 \fBbenchmark \fIvconn n count\fR
212 Sends \fIcount\fR echo request packets that each consist of an
213 OpenFlow header plus \fIn\fR bytes of payload and waits for each
214 response. Reports the total time required. This is a measure of the
215 maximum bandwidth to \fIvconn\fR for round-trips of \fIn\fR-byte
220 Some \fBdpctl\fR commands accept an argument that describes a flow or
221 flows. Such flow descriptions comprise a series
222 \fIfield\fB=\fIvalue\fR assignments, separated by commas or white
225 The following field assignments describe how a flow matches a packet.
226 If any of these assignments is omitted from the flow syntax, the field
227 is treated as a wildcard; thus, if all of them are omitted, the
228 resulting flow matches all packets. The string \fB*\fR or \fBANY\fR
229 may be specified a value to explicitly mark any of these fields as a
232 .IP \fBin_port=\fIport_no\fR
233 Matches physical port \fIport_no\fR. Switch ports are numbered as
234 displayed by \fBdpctl show\fR.
236 .IP \fBdl_vlan=\fIvlan\fR
237 Matches IEEE 802.1q virtual LAN tag \fIvlan\fR. Specify \fB0xffff\fR
238 as \fIvlan\fR to match packets that are not tagged with a virtual LAN;
239 otherwise, specify a number between 0 and 4095, inclusive, as the
240 12-bit VLAN ID to match.
242 .IP \fBdl_src=\fImac\fR
243 Matches Ethernet source address \fImac\fR, which should be specified
244 as 6 pairs of hexadecimal digits delimited by colons,
245 e.g. \fB00:0A:E4:25:6B:B0\fR.
247 .IP \fBdl_dst=\fImac\fR
248 Matches Ethernet destination address \fImac\fR.
250 .IP \fBdl_type=\fIethertype\fR
251 Matches Ethernet protocol type \fIethertype\fR, which should be
252 specified as a integer between 0 and 65535, inclusive, either in
253 decimal or as a hexadecimal number prefixed by \fB0x\fR,
254 e.g. \fB0x0806\fR to match ARP packets.
256 .IP \fBnw_src=\fIip\fR[\fB/\fInetmask\fR]
257 Matches IPv4 source address \fIip\fR, which should be specified as an
258 IP address or host name, e.g. \fB192.168.1.1\fR or
259 \fBwww.example.com\fR. The optional \fInetmask\fR allows matching
260 only on an IPv4 address prefix. It may be specified as a dotted quad
261 (e.g. \fB192.168.1.0/255.255.255.0\fR) or as a count of bits
262 (e.g. \fB192.168.1.0/24\fR).
264 .IP \fBnw_dst=\fIip\fR[\fB/\fInetmask\fR]
265 Matches IPv4 destination address \fIip\fR.
267 .IP \fBnw_proto=\fIproto\fR
268 Matches IP protocol type \fIproto\fR, which should be specified as a
269 decimal number between 0 and 255, inclusive, e.g. 6 to match TCP
272 .IP \fBtp_src=\fIport\fR
273 Matches UDP or TCP source port \fIport\fR, which should be specified
274 as a decimal number between 0 and 65535, inclusive, e.g. 80 to match
275 packets originating from a HTTP server.
277 .IP \fBtp_dst=\fIport\fR
278 Matches UDP or TCP destination port \fIport\fR.
281 The following shorthand notations are also available:
284 Same as \fBdl_type=0x0800\fR.
287 Same as \fBdl_type=0x0800,nw_proto=1\fR.
290 Same as \fBdl_type=0x0800,nw_proto=6\fR.
293 Same as \fBdl_type=0x0800,nw_proto=17\fR.
296 Same as \fBdl_type=0x0806\fR.
299 The \fBadd-flow\fR and \fBadd-flows\fR commands require an additional field:
301 .IP \fIactions\fB=\fItarget\fR[\fB,\fItarget\fR...]\fR
302 Specifies a comma-separated list of actions to take on a packet when the
303 flow entry matches. The \fItarget\fR may be a decimal port number
304 designating the physical port on which to output the packet, or one of
305 the following keywords:
308 .IP \fBoutput\fR:\fIport\fR
309 Outputs the packet on the port specified by \fIport\fR.
312 Subjects the packet to the device's normal L2/L3 processing. (This
313 action is not implemented by all OpenFlow switches.)
316 Outputs the packet on all switch physical ports other than the port on
317 which it was received and any ports on which flooding is disabled
318 (typically, these would be ports disabled by the IEEE 802.1D spanning
322 Outputs the packet on all switch physical ports other than the port on
323 which it was received.
325 .IP \fBcontroller\fR:\fImax_len\fR
326 Sends the packet to the OpenFlow controller as a ``packet in''
327 message. If \fImax_len\fR is a number, then it specifies the maximum
328 number of bytes that should be sent. If \fImax_len\fR is \fBALL\fR or
329 omitted, then the entire packet is sent.
332 Outputs the packet on the ``local port,'' which corresponds to the
333 \fBof\fIn\fR network device (see \fBCONTACTING THE CONTROLLER\fR in
334 \fBsecchan\fR(8) for information on the \fBof\fIn\fR network device).
336 .IP \fBmod_vlan\fR:\fIvlan_id\fR
337 Modifies the VLAN tag on a packet. If \fIvlan_id\fR is a number, then
338 the VLAN tag is added or modified as necessary to match the value
339 specified. If \fIvlan_id\fR is \fBSTRIP\fR, then the VLAN tag is
340 stripped from the packet if one is present. (This action is not
341 implemented by all OpenFlow switches.)
345 (The OpenFlow protocol supports other actions that \fBdpctl\fR does
346 not yet expose to the user.)
349 The \fBadd-flow\fR, \fBadd-flows\fR, and \fBdel-flows\fR commands
350 support an additional optional field:
352 .IP \fBpriority=\fIvalue\fR
353 Sets the priority of the flow to be added or deleted to \fIvalue\fR,
354 which should be a number between 0 and 65535, inclusive. If this
355 field is not specified, it defaults to 32768.
358 The \fBadd-flow\fR and \fBadd-flows\fR commands support additional
362 \fBidle_timeout=\fIseconds\fR
363 Causes the flow to expire after the given number of seconds of
364 inactivity. A value of 0 prevents a flow from expiring due to
365 inactivity. The default is 60 seconds.
367 .IP \fBhard_timeout=\fIseconds\fR
368 Causes the flow to expire after the given number of seconds,
369 regardless of activity. A value of 0 (the default) gives the flow no
370 hard expiration deadline.
373 The \fBdump-flows\fR and \fBdump-aggregate\fR commands support an
374 additional optional field:
376 .IP \fBtable=\fInumber\fR
377 If specified, limits the flows about which statistics are gathered to
378 those in the table with the given \fInumber\fR. Tables are numbered
379 as shown by the \fBdump-tables\fR command.
381 If this field is not specified, or if \fInumber\fR is given as
382 \fB255\fR, statistics are gathered about flows from all tables.
386 \fB-t\fR, \fB--timeout=\fIsecs\fR
387 Limits \fBdpctl\fR runtime to approximately \fIsecs\fR seconds. If
388 the timeout expires, \fBdpctl\fR will exit with a \fBSIGALRM\fR
392 \fB-p\fR, \fB--private-key=\fIprivkey.pem\fR
393 Specifies a PEM file containing the private key used as the
394 identity for SSL connections to a switch.
397 \fB-c\fR, \fB--certificate=\fIcert.pem\fR
398 Specifies a PEM file containing a certificate, signed by the
399 controller's certificate authority (CA), that certifies the
400 private key to identify a trustworthy controller.
403 \fB-C\fR, \fB--ca-cert=\fIcacert.pem\fR
404 Specifies a PEM file containing the CA certificate used to verify that
405 a switch is trustworthy.
408 .BR \-h ", " \-\^\-help
409 Prints a brief help message to the console.
412 \fB-v\fImodule\fR[\fB:\fIfacility\fR[\fB:\fIlevel\fR]], \fB--verbose=\fImodule\fR[\fB:\fIfacility\fR[\fB:\fIlevel\fR]]
413 Sets the logging level for \fImodule\fR in \fIfacility\fR to
414 \fIlevel\fR. The \fImodule\fR may be any valid module name (as
415 displayed by the \fB--list\fR action on \fBvlogconf\fR(8)), or the
416 special name \fBANY\fR to set the logging levels for all modules. The
417 \fIfacility\fR may be \fBsyslog\fR or \fBconsole\fR to set the levels
418 for logging to the system log or to the console, respectively, or
419 \fBANY\fR to set the logging levels for both facilities. If it is
420 omitted, \fIfacility\fR defaults to \fBANY\fR. The \fIlevel\fR must
421 be one of \fBemer\fR, \fBerr\fR, \fBwarn\fR, or \fBdbg\fR, designating
422 the minimum severity of a message for it to be logged. If it is
423 omitted, \fIlevel\fR defaults to \fBdbg\fR.
426 \fB-v\fR, \fB--verbose\fR
427 Sets the maximum logging verbosity level, equivalent to
428 \fB--verbose=ANY:ANY:dbg\fR.
431 .BR \-V ", " \-\^\-version
432 Prints version information to the console.
436 A typical dpctl command sequence for controlling an OpenFlow kernel module:
439 Create datapath numbered 0:
441 .B % dpctl adddp nl:0
444 Add two network devices to the new datapath:
446 .B % dpctl addif nl:0 eth0
447 .B % dpctl addif nl:0 eth1
450 Monitor traffic received by the datapath (exit with control-C):
452 .B % dpctl monitor nl:0
456 View the datapath's table stats after some traffic has passed through:
458 .B % dpctl dump-tables nl:0
461 View the flow entries in the datapath:
463 .B % dpctl dump-flows nl:0
466 Remove network devices from the datapath when finished:
468 .B % dpctl delif nl:0 eth0
469 .B % dpctl delif nl:0 eth1
474 .B % dpctl deldp nl:0