Annotation of embedaddon/iperf/docs/invoking.rst, revision 1.1.1.1

1.1       misho       1: Invoking iperf3
                      2: ===============
                      3: 
                      4: iperf3 includes a manual page listing all of the command-line options.
                      5: The manual page is the most up-to-date reference to the various flags and parameters.
                      6: 
                      7: For sample command line usage, see: 
                      8: 
                      9: http://fasterdata.es.net/performance-testing/network-troubleshooting-tools/iperf-and-iperf3/
                     10: 
                     11: Using the default options, iperf3 is meant to show typical well
                     12: designed application performance.  "Typical well designed application"
                     13: means avoiding artificial enhancements that work only for testing
                     14: (such as ``splice()``-ing the data to ``/dev/null``).  iperf3 does
                     15: also have flags for "extreme best case" optimizations but they must be
                     16: explicitly activated.  These flags include the ``-Z`` (``--zerocopy``)
                     17: and ``-A`` (``--affinity``) options.
                     18: 
                     19: iperf3 Manual Page
                     20: ------------------
                     21: 
                     22: This section contains a plaintext rendering of the iperf3 manual page.
                     23: It is presented here only for convenience; the authoritative iperf3
                     24: manual page is included in the source tree and installed along with
                     25: the executable.
                     26: 
                     27: ::
                     28: 
                     29:    IPERF(1)                         User Manuals                         IPERF(1)
                     30: 
                     31: 
                     32: 
                     33:    NAME
                     34:    iperf3 − perform network throughput tests
                     35: 
                     36:    SYNOPSIS
                     37:    iperf3 ‐s [ options ]
                     38:    iperf3 ‐c server [ options ]
                     39: 
                     40: 
                     41:    DESCRIPTION
                     42:    iperf3  is  a  tool for performing network throughput measurements.  It
                     43:    can test either TCP or UDP throughput.  To perform an iperf3  test  the
                     44:    user must establish both a server and a client.
                     45: 
                     46: 
                     47:    GENERAL OPTIONS
                     48:    ‐p, ‐‐port n
                     49:    set server port to listen on/connect to to n (default 5201)
                     50: 
                     51:    ‐f, ‐‐format
                     52:    [kmKM]   format to report: Kbits, Mbits, KBytes, MBytes
                     53: 
                     54:    ‐i, ‐‐interval n
                     55:    pause  n  seconds between periodic bandwidth reports; default is
                     56:    1, use 0 to disable
                     57: 
                     58:    ‐F, ‐‐file name
                     59:    client‐side: read from  the  file  and  write  to  the  network,
                     60:    instead of using random data; server‐side: read from the network
                     61:    and write to the file, instead of throwing the data away
                     62: 
                     63:    ‐A, ‐‐affinity n/n,m
                     64:    Set the CPU affinity, if possible (Linux and FreeBSD only).   On
                     65:    both  the  client  and  server you can set the local affinity by
                     66:    using the n form of this argument (where n is a CPU number).  In
                     67:    addition,  on  the  client  side  you  can override the server’s
                     68:    affinity for just that one test, using the n,m form of argument.
                     69:    Note  that when using this feature, a process will only be bound
                     70:    to a single CPU (as opposed to a set containing potentialy  mul‐
                     71:    tiple CPUs).
                     72: 
                     73:    ‐B, ‐‐bind host
                     74:    bind to a specific interface
                     75: 
                     76:    ‐V, ‐‐verbose
                     77:    give more detailed output
                     78: 
                     79:    ‐J, ‐‐json
                     80:    output in JSON format
                     81: 
                     82:    ‐‐logfile file
                     83:    send output to a log file.
                     84: 
                     85:    ‐d, ‐‐debug
                     86:    emit  debugging  output.  Primarily (perhaps exclusively) of use
                     87:    to developers.
                     88: 
                     89:    ‐v, ‐‐version
                     90:    show version information and quit
                     91: 
                     92:    ‐h, ‐‐help
                     93:    show a help synopsis
                     94: 
                     95: 
                     96:    SERVER SPECIFIC OPTIONS
                     97:    ‐s, ‐‐server
                     98:    run in server mode
                     99: 
                    100:    ‐D, ‐‐daemon
                    101:    run the server in background as a daemon
                    102: 
                    103:    ‐I, ‐‐pidfile file
                    104:    write a file with the process ID, most useful when running as  a
                    105:    daemon.
                    106: 
                    107: 
                    108:    CLIENT SPECIFIC OPTIONS
                    109:    ‐c, ‐‐client host
                    110:    run in client mode, connecting to the specified server
                    111: 
                    112:    ‐‐sctp use SCTP rather than TCP (FreeBSD and Linux)
                    113: 
                    114:    ‐u, ‐‐udp
                    115:    use UDP rather than TCP
                    116: 
                    117:    ‐b, ‐‐bandwidth n[KM]
                    118:    set  target bandwidth to n bits/sec (default 1 Mbit/sec for UDP,
                    119:    unlimited for TCP).  If there are multiple  streams  (‐P  flag),
                    120:    the  bandwidth  limit is applied separately to each stream.  You
                    121:    can also add a ’/’ and a  number  to  the  bandwidth  specifier.
                    122:    This  is  called "burst mode".  It will send the given number of
                    123:    packets without pausing, even if that  temporarily  exceeds  the
                    124:    specified bandwidth limit.
                    125: 
                    126:    ‐t, ‐‐time n
                    127:    time in seconds to transmit for (default 10 secs)
                    128: 
                    129:    ‐n, ‐‐bytes n[KM]
                    130:    number of bytes to transmit (instead of ‐t)
                    131: 
                    132:    ‐k, ‐‐blockcount n[KM]
                    133:    number of blocks (packets) to transmit (instead of ‐t or ‐n)
                    134: 
                    135:    ‐l, ‐‐length n[KM]
                    136:    length  of  buffer to read or write (default 128 KB for TCP, 8KB
                    137:    for UDP)
                    138: 
                    139:    ‐P, ‐‐parallel n
                    140:    number of parallel client streams to run
                    141: 
                    142:    ‐R, ‐‐reverse
                    143:    run in reverse mode (server sends, client receives)
                    144: 
                    145:    ‐w, ‐‐window n[KM]
                    146:    TCP window size / socket buffer size  (this  gets  sent  to  the
                    147:    server and used on that side too)
                    148: 
                    149:    ‐M, ‐‐set‐mss n
                    150:    set TCP maximum segment size (MTU ‐ 40 bytes)
                    151: 
                    152:    ‐N, ‐‐no‐delay
                    153:    set TCP no delay, disabling Nagle’s Algorithm
                    154: 
                    155:    ‐4, ‐‐version4
                    156:    only use IPv4
                    157: 
                    158:    ‐6, ‐‐version6
                    159:    only use IPv6
                    160: 
                    161:    ‐S, ‐‐tos n
                    162:    set the IP ’type of service’
                    163: 
                    164:    ‐L, ‐‐flowlabel n
                    165:    set the IPv6 flow label (currently only supported on Linux)
                    166: 
                    167:    ‐Z, ‐‐zerocopy
                    168:    Use  a  "zero copy" method of sending data, such as sendfile(2),
                    169:    instead of the usual write(2).
                    170: 
                    171:    ‐O, ‐‐omit n
                    172:    Omit the first n seconds of the test, to skip past the TCP slow‐
                    173:    start period.
                    174: 
                    175:    ‐T, ‐‐title str
                    176:    Prefix every output line with this string.
                    177: 
                    178:    ‐C, ‐‐linux‐congestion algo
                    179:    Set the congestion control algorithm (linux only).
                    180: 
                    181: 
                    182:    AUTHORS
                    183:    Iperf  was  originally  written by Mark Gates and Alex Warshavsky.  Man
                    184:    page and maintence by Jon Dugan <jdugan at x1024 dot net>.  Other  con‐
                    185:    tributions  from  Ajay  Tirumala,  Jim Ferguson, Feng Qin, Kevin Gibbs,
                    186:    John Estabrook <jestabro at ncsa.uiuc.edu>, Andrew  Gallatin  <gallatin
                    187:    at gmail.com>, Stephen Hemminger <shemminger at linux‐foundation.org>
                    188: 
                    189: 
                    190:    SEE ALSO
                    191:    libiperf(3), https://github.com/esnet/iperf
                    192: 
                    193: 
                    194: 
                    195:    ESnet                            February 2014                        IPERF(1)
                    196: 
                    197: The iperf3 manual page will typically be installed in manual
                    198: section 1.
                    199: 

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>