1IPERF3(1)                        User Manuals                        IPERF3(1)
2
3
4

NAME

6       iperf3 - perform network throughput tests
7

SYNOPSIS

9       iperf3 -s [ options ]
10       iperf3 -c server [ options ]
11
12

DESCRIPTION

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

GENERAL OPTIONS

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       -F, --file name
111              Use  a  file  as  the  source  (on  the  sender) or sink (on the
112              receiver) of data, rather than just generating  random  data  or
113              throwing  it  away.  This feature is used for finding whether or
114              not the storage subsystem is the bottleneck for file  transfers.
115              It  does not turn iperf3 into a file transfer tool.  The length,
116              attributes, and in some cases contents of the received file  may
117              not match those of the original file.
118
119       -A, --affinity n/n,m
120              Set  the  CPU affinity, if possible (Linux, FreeBSD, and Windows
121              only).  On both the client and server  you  can  set  the  local
122              affinity  by using the n form of this argument (where n is a CPU
123              number).  In addition, on the client side you can  override  the
124              server's  affinity for just that one test, using the n,m form of
125              argument.  Note that when using this  feature,  a  process  will
126              only  be  bound  to a single CPU (as opposed to a set containing
127              potentialy multiple CPUs).
128
129       -B, --bind host
130              bind to the specific interface associated with address host.
131
132       -V, --verbose
133              give more detailed output
134
135       -J, --json
136              output in JSON format
137
138       --logfile file
139              send output to a log file.
140
141       --forceflush
142              force flushing output at every interval.  Used to avoid  buffer‐
143              ing when sending output to pipe.
144
145       -d, --debug
146              emit  debugging  output.  Primarily (perhaps exclusively) of use
147              to developers.
148
149       -v, --version
150              show version information and quit
151
152       -h, --help
153              show a help synopsis
154
155

SERVER SPECIFIC OPTIONS

157       -s, --server
158              run in server mode
159
160       -D, --daemon
161              run the server in background as a daemon
162
163       -I, --pidfile file
164              write a file with the process ID, most useful when running as  a
165              daemon.
166
167       -1, --one-off
168              handle one client connection, then exit.
169
170       --rsa-private-key-path file
171              path  to  the  RSA  private key (not password-protected) used to
172              decrypt authentication credentials from  the  client  (if  built
173              with OpenSSL support).
174
175       --authorized-users-path file
176              path  to the configuration file containing authorized users cre‐
177              dentials to run iperf tests (if  built  with  OpenSSL  support).
178              The  file  is  a  comma separated list of usernames and password
179              hashes; more information on the structure of  the  file  can  be
180              found in the EXAMPLES section.
181

CLIENT SPECIFIC OPTIONS

183       -c, --client host
184              run  in  client  mode,  connecting  to the specified server.  By
185              default, a test consists of sending data from the client to  the
186              server, unless the -R flag is specified.
187
188       --sctp use SCTP rather than TCP (FreeBSD and Linux)
189
190       -u, --udp
191              use UDP rather than TCP
192
193       --connect-timeout n
194              set  timeout  for establishing the initial control connection to
195              the server, in milliseconds.  The default behavior is the  oper‐
196              ating  system's  timeout for TCP connection establishment.  Pro‐
197              viding a shorter value may speed up detection of a  down  iperf3
198              server.
199
200       -b, --bitrate n[KM]
201              set  target  bitrate  to n bits/sec (default 1 Mbit/sec for UDP,
202              unlimited for TCP/SCTP).  If  there  are  multiple  streams  (-P
203              flag),  the  throughput  limit  is  applied  separately  to each
204              stream.  You can also add a '/' and  a  number  to  the  bitrate
205              specifier.  This is called "burst mode".  It will send the given
206              number of packets without  pausing,  even  if  that  temporarily
207              exceeds  the  specified  throughput  limit.   Setting the target
208              bitrate to 0 will disable bitrate  limits  (particularly  useful
209              for UDP tests).  This throughput limit is implemented internally
210              inside iperf3, and is available on all platforms.  Compare  with
211              the  --fq-rate flag.  This option replaces the --bandwidth flag,
212              which is now deprecated but (at least for now) still accepted.
213
214       --pacing-timer n[KMG]
215              set  pacing  timer  interval  in  microseconds   (default   1000
216              microseconds,  or 1 ms).  This controls iperf3's internal pacing
217              timer for the -b/--bitrate  option.   The  timer  fires  at  the
218              interval  set  by  this parameter.  Smaller values of the pacing
219              timer parameter smooth out the traffic emitted  by  iperf3,  but
220              potentially  at  the  cost  of  performance due to more frequent
221              timer processing.
222
223       --fq-rate n[KM]
224              Set a rate to be used with fair-queueing based socket-level pac‐
225              ing,  in bits per second.  This pacing (if specified) will be in
226              addition to any pacing due to iperf3's internal throughput  pac‐
227              ing  (-b/--bitrate flag), and both can be specified for the same
228              test.  Only available on platforms  supporting  the  SO_MAX_PAC‐
229              ING_RATE  socket  option (currently only Linux).  The default is
230              no fair-queueing based pacing.
231
232       --no-fq-socket-pacing
233              This option is deprecated and will be removed.  It is equivalent
234              to specifying --fq-rate=0.
235
236       -t, --time n
237              time in seconds to transmit for (default 10 secs)
238
239       -n, --bytes n[KM]
240              number of bytes to transmit (instead of -t)
241
242       -k, --blockcount n[KM]
243              number of blocks (packets) to transmit (instead of -t or -n)
244
245       -l, --length n[KM]
246              length  of  buffer to read or write.  For TCP tests, the default
247              value is 128KB.  In the case of UDP, iperf3 tries to dynamically
248              determine  a  reasonable  sending size based on the path MTU; if
249              that cannot be determined it uses 1460 bytes as a sending  size.
250              For SCTP tests, the default size is 64KB.
251
252       --cport port
253              bind  data  streams  to  a specific client port (for TCP and UDP
254              only, default is to use an ephemeral port)
255
256       -P, --parallel n
257              number of parallel client streams to run. Note  that  iperf3  is
258              single  threaded,  so  if you are CPU bound, this will not yield
259              higher throughput.
260
261       -R, --reverse
262              reverse the direction of a test, so that the server  sends  data
263              to the client
264
265       -w, --window n[KM]
266              window  size  / socket buffer size (this gets sent to the server
267              and used on that side too)
268
269       -M, --set-mss n
270              set TCP/SCTP maximum segment size (MTU - 40 bytes)
271
272       -N, --no-delay
273              set TCP/SCTP no delay, disabling Nagle's Algorithm
274
275       -4, --version4
276              only use IPv4
277
278       -6, --version6
279              only use IPv6
280
281       -S, --tos n
282              set the IP type of service. The usual prefixes for octal and hex
283              can be used, i.e. 52, 064 and 0x34 all specify the same value.
284
285       --dscp dscp
286              set  the  IP  DSCP  bits.   Both numeric and symbolic values are
287              accepted. Numeric values can be specified in decimal, octal  and
288              hex (see --tos above).
289
290       -L, --flowlabel n
291              set the IPv6 flow label (currently only supported on Linux)
292
293       -X, --xbind name
294              Bind  SCTP  associations  to  a  specific  subset of links using
295              sctp_bindx(3).  The --B flag will be ignored  if  this  flag  is
296              specified.  Normally SCTP will include the protocol addresses of
297              all active links on the local host when setting up  an  associa‐
298              tion.  Specifying at least one --X name will disable this behav‐
299              iour.  This flag must be specified for each link to be  included
300              in  the association, and is supported for both iperf servers and
301              clients (the latter are supported by passing the first --X argu‐
302              ment  to  bind(2)).  Hostnames are accepted as arguments and are
303              resolved using getaddrinfo(3).  If the  --4  or  --6  flags  are
304              specified,  names  which  do not resolve to addresses within the
305              specified protocol family will be ignored.
306
307       --nstreams n
308              Set number of SCTP streams.
309
310       -Z, --zerocopy
311              Use a "zero copy" method of sending data, such  as  sendfile(2),
312              instead of the usual write(2).
313
314       -O, --omit n
315              Omit the first n seconds of the test, to skip past the TCP slow-
316              start period.
317
318       -T, --title str
319              Prefix every output line with this string.
320
321       --extra-data str
322              Specify an extra data string field to be included in  JSON  out‐
323              put.
324
325       -C, --congestion algo
326              Set  the  congestion control algorithm (Linux and FreeBSD only).
327              An older --linux-congestion synonym for this  flag  is  accepted
328              but is deprecated.
329
330       --get-server-output
331              Get the output from the server.  The output format is determined
332              by the server (in particular, if the server was invoked with the
333              --json  flag,  the  output  will be in JSON format, otherwise it
334              will be in human-readable format).  If the client  is  run  with
335              --json,  the  server output is included in a JSON object; other‐
336              wise it is appended at the bottom of the human-readable output.
337
338       --udp-counters-64bit
339              Use 64-bit counters in UDP test packets.  The use of this option
340              can  help  prevent counter overflows during long or high-bitrate
341              UDP tests.  Both client and server need to be running  at  least
342              version  3.1 for this option to work.  It may become the default
343              behavior at some point in the future.
344
345       --repeating-payload
346              Use repeating pattern in payload, instead of random bytes.   The
347              same  payload  is  used  in iperf2 (ASCII '0..9' repeating).  It
348              might help to test and reveal problems in networking  gear  with
349              hardware  compression (including some WiFi access points), where
350              iperf2 and iperf3 perform differently,  just  based  on  payload
351              entropy.
352
353       --username username
354              username to use for authentication to the iperf server (if built
355              with OpenSSL support).  The password will be prompted for inter‐
356              actively  when  the  test is run.  Note, the password to use can
357              also be specified via the IPERF3_PASSWORD environment  variable.
358              If  this  variable  is  present,  the  password  prompt  will be
359              skipped.
360
361       --rsa-public-key-path file
362              path to the RSA public key used to encrypt  authentication  cre‐
363              dentials (if built with OpenSSL support)
364
365

EXAMPLES

367   Authentication - RSA Keypair
368       The  authentication  feature  of iperf3 requires an RSA public keypair.
369       The public key is used to encrypt the authentication  token  containing
370       the  user  credentials,  while  the  private key is used to decrypt the
371       authentication token.  An example of a set of  UNIX/Linux  commands  to
372       generate correct keypair follows:
373
374            > openssl genrsa -des3 -out private.pem 2048
375            > openssl rsa -in private.pem -outform PEM -pubout -out public.pem
376            > openssl rsa -in private.pem -out private_not_protected.pem -out‐
377            form PEM
378
379       After these commands, the public key will be contained in the file pub‐
380       lic.pem  and  the  private  key  will  be  contained  in  the file pri‐
381       vate_not_protected.pem.
382
383   Authentication - Authorized users configuration file
384       A simple plaintext file must be provided to the iperf3 server in  order
385       to  specify the authorized user credentials.  The file is a simple list
386       of comma-separated pairs of a username  and  a  corresponding  password
387       hash.   The password hash is a SHA256 hash of the string "{$user}$pass‐
388       word".  The file can also contain commented lines (starting with the  #
389       character).   An example of commands to generate the password hash on a
390       UNIX/Linux system is given below:
391
392            > S_USER=mario S_PASSWD=rossi
393            > echo -n "{$S_USER}$S_PASSWD" | sha256sum | awk '{ print $1 }'
394
395       An example of a password file (with an entry corresponding to the above
396       username and password) is given below:
397            > cat credentials.csv
398            # file format: username,sha256
399            mario,bf7a49a846d44b454a5d11e7acfaf13d138bbe0b7483aa3e050879700572709b
400
401
402

AUTHORS

404       A list of the contributors to iperf3 can be found within the documenta‐
405       tion located at https://software.es.net/iperf/dev.html#authors.
406
407

SEE ALSO

409       libiperf(3), https://software.es.net/iperf
410
411
412
413ESnet                              June 2018                         IPERF3(1)
Impressum