1IPERF3(1) User Manuals IPERF3(1)
2
3
4
6 iperf3 - perform network throughput tests
7
9 iperf3 -s [ options ]
10 iperf3 -c server [ options ]
11
12
14 iperf3 is a tool for performing network throughput measurements. It
15 can test TCP, UDP, or SCTP throughput. To perform an iperf3 test the
16 user must establish both a server and a client.
17
18 The iperf3 executable contains both client and server functionality.
19 An iperf3 server can be started using either of the -s or --server com‐
20 mand-line parameters, for example:
21
22 iperf3 -s
23
24 iperf3 --server
25
26 Note that many iperf3 parameters have both short (-s) and long
27 (--server) forms. In this section we will generally use the short form
28 of command-line flags, unless only the long form of a flag is avail‐
29 able.
30
31 By default, the iperf3 server listens on TCP port 5201 for connections
32 from an iperf3 client. A custom port can be specified by using the -p
33 flag, for example:
34
35 iperf3 -s -p 5002
36
37 After the server is started, it will listen for connections from iperf3
38 clients (in other words, the iperf3 program run in client mode). The
39 client mode can be started using the -c command-line option, which also
40 requires a host to which iperf3 should connect. The host can by speci‐
41 fied by hostname, IPv4 literal, or IPv6 literal:
42
43 iperf3 -c iperf3.example.com
44
45 iperf3 -c 192.0.2.1
46
47 iperf3 -c 2001:db8::1
48
49 If the iperf3 server is running on a non-default TCP port, that port
50 number needs to be specified on the client as well:
51
52 iperf3 -c iperf3.example.com -p 5002
53
54 The initial TCP connection is used to exchange test parameters, control
55 the start and end of the test, and to exchange test results. This is
56 sometimes referred to as the "control connection". The actual test
57 data is sent over a separate TCP connection, as a separate flow of UDP
58 packets, or as an independent SCTP connection, depending on what proto‐
59 col was specified by the client.
60
61 Normally, the test data is sent from the client to the server, and mea‐
62 sures the upload speed of the client. Measuring the download speed
63 from the server can be done by specifying the -R flag on the client.
64 This causes data to be sent from the server to the client.
65
66 iperf3 -c iperf3.example.com -p 5202 -R
67
68 Results are displayed on both the client and server. There will be at
69 least one line of output per measurement interval (by default a mea‐
70 surement interval lasts for one second, but this can be changed by the
71 -i option). Each line of output includes (at least) the time since the
72 start of the test, amount of data transferred during the interval, and
73 the average bitrate over that interval. Note that the values for each
74 measurement interval are taken from the point of view of the endpoint
75 process emitting that output (in other words, the output on the client
76 shows the measurement interval data for the client.
77
78 At the end of the test is a set of statistics that shows (at least as
79 much as possible) a summary of the test as seen by both the sender and
80 the receiver, with lines tagged accordingly. Recall that by default
81 the client is the sender and the server is the receiver, although as
82 indicated above, use of the -R flag will reverse these roles.
83
84 The client can be made to retrieve the server-side output for a given
85 test by specifying the --get-server-output flag.
86
87 Either the client or the server can produce its output in a JSON struc‐
88 ture, useful for integration with other programs, by passing it the -J
89 flag. Because the contents of the JSON structure are only competely
90 known after the test has finished, no JSON output will be emitted until
91 the end of the test.
92
93 iperf3 has a (overly) large set of command-line options that can be
94 used to set the parameters of a test. They are given in the "GENERAL
95 OPTIONS" section of the manual page below, as well as summarized in
96 iperf3's help output, which can be viewed by running iperf3 with the -h
97 flag.
98
100 -p, --port n
101 set server port to listen on/connect to to n (default 5201)
102
103 -f, --format
104 [kmgtKMGT] format to report: Kbits/Mbits/Gbits/Tbits
105
106 -i, --interval n
107 pause n seconds between periodic throughput reports; default is
108 1, use 0 to disable
109
110 -I, --pidfile file
111 write a file with the process ID, most useful when running as a
112 daemon.
113
114 -F, --file name
115 Use a file as the source (on the sender) or sink (on the re‐
116 ceiver) of data, rather than just generating random data or
117 throwing it away. This feature is used for finding whether or
118 not the storage subsystem is the bottleneck for file transfers.
119 It does not turn iperf3 into a file transfer tool. The length,
120 attributes, and in some cases contents of the received file may
121 not match those of the original file.
122
123 -A, --affinity n/n,m
124 Set the CPU affinity, if possible (Linux, FreeBSD, and Windows
125 only). On both the client and server you can set the local
126 affinity by using the n form of this argument (where n is a CPU
127 number). In addition, on the client side you can override the
128 server's affinity for just that one test, using the n,m form of
129 argument. Note that when using this feature, a process will
130 only be bound to a single CPU (as opposed to a set containing
131 potentialy multiple CPUs).
132
133 -B, --bind host
134 bind to the specific interface associated with address host.
135 --bind-dev dev.ft R bind to the specified network interface.
136 This option uses SO_BINDTODEVICE, and may require root permis‐
137 sions. (Available on Linux and possibly other systems.)
138
139 -V, --verbose
140 give more detailed output
141
142 -J, --json
143 output in JSON format
144
145 --logfile file
146 send output to a log file.
147
148 --forceflush
149 force flushing output at every interval. Used to avoid buffer‐
150 ing when sending output to pipe.
151
152 --timestamps[=format]
153 prepend a timestamp at the start of each output line. By de‐
154 fault, timestamps have the format emitted by ctime(1). Option‐
155 ally, = followed by a format specification can be passed to cus‐
156 tomize the timestamps, see strftime(3). If this optional format
157 is given, the = must immediately follow the --timestamps option
158 with no whitespace intervening.
159
160 --rcv-timeout #
161 set idle timeout for receiving data during active tests. The re‐
162 ceiver will halt a test if no data is received from the sender
163 for this number of ms (default to 12000 ms, or 2 minutes).
164
165 -d, --debug
166 emit debugging output. Primarily (perhaps exclusively) of use
167 to developers.
168
169 -v, --version
170 show version information and quit
171
172 -h, --help
173 show a help synopsis
174
175
177 -s, --server
178 run in server mode
179
180 -D, --daemon
181 run the server in background as a daemon
182
183 -1, --one-off
184 handle one client connection, then exit.
185
186 --server-bitrate-limit n[KMGT]
187 set a limit on the server side, which will cause a test to abort
188 if the client specifies a test of more than n bits per second,
189 or if the average data sent or received by the client (including
190 all data streams) is greater than n bits per second. The de‐
191 fault limit is zero, which implies no limit. The interval over
192 which to average the data rate is 5 seconds by default, but can
193 be specified by adding a '/' and a number to the bitrate speci‐
194 fier.
195
196 --rsa-private-key-path file
197 path to the RSA private key (not password-protected) used to de‐
198 crypt authentication credentials from the client (if built with
199 OpenSSL support).
200
201 --authorized-users-path file
202 path to the configuration file containing authorized users cre‐
203 dentials to run iperf tests (if built with OpenSSL support).
204 The file is a comma separated list of usernames and password
205 hashes; more information on the structure of the file can be
206 found in the EXAMPLES section.
207
208 --time-skew-thresholdsecond seconds
209 time skew threshold (in seconds) between the server and client
210 during the authentication process.
211
213 -c, --client host
214 run in client mode, connecting to the specified server. By de‐
215 fault, a test consists of sending data from the client to the
216 server, unless the -R flag is specified.
217
218 --sctp use SCTP rather than TCP (FreeBSD and Linux)
219
220 -u, --udp
221 use UDP rather than TCP
222
223 --connect-timeout n
224 set timeout for establishing the initial control connection to
225 the server, in milliseconds. The default behavior is the oper‐
226 ating system's timeout for TCP connection establishment. Pro‐
227 viding a shorter value may speed up detection of a down iperf3
228 server.
229
230 -b, --bitrate n[KMGT]
231 set target bitrate to n bits/sec (default 1 Mbit/sec for UDP,
232 unlimited for TCP/SCTP). If there are multiple streams (-P
233 flag), the throughput limit is applied separately to each
234 stream. You can also add a '/' and a number to the bitrate
235 specifier. This is called "burst mode". It will send the given
236 number of packets without pausing, even if that temporarily ex‐
237 ceeds the specified throughput limit. Setting the target bi‐
238 trate to 0 will disable bitrate limits (particularly useful for
239 UDP tests). This throughput limit is implemented internally in‐
240 side iperf3, and is available on all platforms. Compare with
241 the --fq-rate flag. This option replaces the --bandwidth flag,
242 which is now deprecated but (at least for now) still accepted.
243
244 --pacing-timer n[KMGT]
245 set pacing timer interval in microseconds (default 1000 mi‐
246 croseconds, or 1 ms). This controls iperf3's internal pacing
247 timer for the -b/--bitrate option. The timer fires at the in‐
248 terval set by this parameter. Smaller values of the pacing
249 timer parameter smooth out the traffic emitted by iperf3, but
250 potentially at the cost of performance due to more frequent
251 timer processing.
252
253 --fq-rate n[KMGT]
254 Set a rate to be used with fair-queueing based socket-level pac‐
255 ing, in bits per second. This pacing (if specified) will be in
256 addition to any pacing due to iperf3's internal throughput pac‐
257 ing (-b/--bitrate flag), and both can be specified for the same
258 test. Only available on platforms supporting the SO_MAX_PAC‐
259 ING_RATE socket option (currently only Linux). The default is
260 no fair-queueing based pacing.
261
262 --no-fq-socket-pacing
263 This option is deprecated and will be removed. It is equivalent
264 to specifying --fq-rate=0.
265
266 -t, --time n
267 time in seconds to transmit for (default 10 secs)
268
269 -n, --bytes n[KMGT]
270 number of bytes to transmit (instead of -t)
271
272 -k, --blockcount n[KMGT]
273 number of blocks (packets) to transmit (instead of -t or -n)
274
275 -l, --length n[KMGT]
276 length of buffer to read or write. For TCP tests, the default
277 value is 128KB. In the case of UDP, iperf3 tries to dynamically
278 determine a reasonable sending size based on the path MTU; if
279 that cannot be determined it uses 1460 bytes as a sending size.
280 For SCTP tests, the default size is 64KB.
281
282 --cport port
283 bind data streams to a specific client port (for TCP and UDP
284 only, default is to use an ephemeral port)
285
286 -P, --parallel n
287 number of parallel client streams to run. Note that iperf3 is
288 single threaded, so if you are CPU bound, this will not yield
289 higher throughput.
290
291 -R, --reverse
292 reverse the direction of a test, so that the server sends data
293 to the client
294
295 --bidir
296 test in both directions (normal and reverse), with both the
297 client and server sending and receiving data simultaneously
298
299 -w, --window n[KMGT]
300 window size / socket buffer size (this gets sent to the server
301 and used on that side too)
302
303 -M, --set-mss n
304 set TCP/SCTP maximum segment size (MTU - 40 bytes)
305
306 -N, --no-delay
307 set TCP/SCTP no delay, disabling Nagle's Algorithm
308
309 -4, --version4
310 only use IPv4
311
312 -6, --version6
313 only use IPv6
314
315 -S, --tos n
316 set the IP type of service. The usual prefixes for octal and hex
317 can be used, i.e. 52, 064 and 0x34 all specify the same value.
318
319 --dscp dscp
320 set the IP DSCP bits. Both numeric and symbolic values are ac‐
321 cepted. Numeric values can be specified in decimal, octal and
322 hex (see --tos above).
323
324 -L, --flowlabel n
325 set the IPv6 flow label (currently only supported on Linux)
326
327 -X, --xbind name
328 Bind SCTP associations to a specific subset of links using
329 sctp_bindx(3). The --B flag will be ignored if this flag is
330 specified. Normally SCTP will include the protocol addresses of
331 all active links on the local host when setting up an associa‐
332 tion. Specifying at least one --X name will disable this behav‐
333 iour. This flag must be specified for each link to be included
334 in the association, and is supported for both iperf servers and
335 clients (the latter are supported by passing the first --X argu‐
336 ment to bind(2)). Hostnames are accepted as arguments and are
337 resolved using getaddrinfo(3). If the --4 or --6 flags are
338 specified, names which do not resolve to addresses within the
339 specified protocol family will be ignored.
340
341 --nstreams n
342 Set number of SCTP streams.
343
344 -Z, --zerocopy
345 Use a "zero copy" method of sending data, such as sendfile(2),
346 instead of the usual write(2).
347
348 -O, --omit n
349 Omit the first n seconds of the test, to skip past the TCP slow-
350 start period.
351
352 -T, --title str
353 Prefix every output line with this string.
354
355 --extra-data str
356 Specify an extra data string field to be included in JSON out‐
357 put.
358
359 -C, --congestion algo
360 Set the congestion control algorithm (Linux and FreeBSD only).
361 An older --linux-congestion synonym for this flag is accepted
362 but is deprecated.
363
364 --get-server-output
365 Get the output from the server. The output format is determined
366 by the server (in particular, if the server was invoked with the
367 --json flag, the output will be in JSON format, otherwise it
368 will be in human-readable format). If the client is run with
369 --json, the server output is included in a JSON object; other‐
370 wise it is appended at the bottom of the human-readable output.
371
372 --udp-counters-64bit
373 Use 64-bit counters in UDP test packets. The use of this option
374 can help prevent counter overflows during long or high-bitrate
375 UDP tests. Both client and server need to be running at least
376 version 3.1 for this option to work. It may become the default
377 behavior at some point in the future.
378
379 --repeating-payload
380 Use repeating pattern in payload, instead of random bytes. The
381 same payload is used in iperf2 (ASCII '0..9' repeating). It
382 might help to test and reveal problems in networking gear with
383 hardware compression (including some WiFi access points), where
384 iperf2 and iperf3 perform differently, just based on payload en‐
385 tropy.
386
387 --dont-fragment
388 Set the IPv4 Don't Fragment (DF) bit on outgoing packets. Only
389 applicable to tests doing UDP over IPv4.
390
391 --username username
392 username to use for authentication to the iperf server (if built
393 with OpenSSL support). The password will be prompted for inter‐
394 actively when the test is run. Note, the password to use can
395 also be specified via the IPERF3_PASSWORD environment variable.
396 If this variable is present, the password prompt will be
397 skipped.
398
399 --rsa-public-key-path file
400 path to the RSA public key used to encrypt authentication cre‐
401 dentials (if built with OpenSSL support)
402
403
405 Authentication - RSA Keypair
406 The authentication feature of iperf3 requires an RSA public keypair.
407 The public key is used to encrypt the authentication token containing
408 the user credentials, while the private key is used to decrypt the au‐
409 thentication token. The private key must be in PEM format and addi‐
410 tionally must not have a password set. The public key must be in PEM
411 format and use SubjectPrefixKeyInfo encoding. An example of a set of
412 UNIX/Linux commands using OpenSSL to generate a correctly-formed key‐
413 pair follows:
414
415 > openssl genrsa -des3 -out private.pem 2048
416 > openssl rsa -in private.pem -outform PEM -pubout -out public.pem
417 > openssl rsa -in private.pem -out private_not_protected.pem -out‐
418 form PEM
419
420 After these commands, the public key will be contained in the file pub‐
421 lic.pem and the private key will be contained in the file pri‐
422 vate_not_protected.pem.
423
424 Authentication - Authorized users configuration file
425 A simple plaintext file must be provided to the iperf3 server in order
426 to specify the authorized user credentials. The file is a simple list
427 of comma-separated pairs of a username and a corresponding password
428 hash. The password hash is a SHA256 hash of the string "{$user}$pass‐
429 word". The file can also contain commented lines (starting with the #
430 character). An example of commands to generate the password hash on a
431 UNIX/Linux system is given below:
432
433 > S_USER=mario S_PASSWD=rossi
434 > echo -n "{$S_USER}$S_PASSWD" | sha256sum | awk '{ print $1 }'
435
436 An example of a password file (with an entry corresponding to the above
437 username and password) is given below:
438 > cat credentials.csv
439 # file format: username,sha256
440 mario,bf7a49a846d44b454a5d11e7ac‐
441 faf13d138bbe0b7483aa3e050879700572709b
442
443
444
446 A list of the contributors to iperf3 can be found within the documenta‐
447 tion located at https://software.es.net/iperf/dev.html#authors.
448
449
451 libiperf(3), https://software.es.net/iperf
452
453
454
455ESnet February 2021 IPERF3(1)