xref: /freebsd/share/man/man4/tcp.4 (revision b43b8f81578cbb7bddbd6f7b8ebe06a219c88140)
1.\" Copyright (c) 1983, 1991, 1993
2.\"	The Regents of the University of California.
3.\" Copyright (c) 2010-2011 The FreeBSD Foundation
4.\" All rights reserved.
5.\"
6.\" Portions of this documentation were written at the Centre for Advanced
7.\" Internet Architectures, Swinburne University of Technology, Melbourne,
8.\" Australia by David Hayes under sponsorship from the FreeBSD Foundation.
9.\"
10.\" Redistribution and use in source and binary forms, with or without
11.\" modification, are permitted provided that the following conditions
12.\" are met:
13.\" 1. Redistributions of source code must retain the above copyright
14.\"    notice, this list of conditions and the following disclaimer.
15.\" 2. Redistributions in binary form must reproduce the above copyright
16.\"    notice, this list of conditions and the following disclaimer in the
17.\"    documentation and/or other materials provided with the distribution.
18.\" 3. Neither the name of the University nor the names of its contributors
19.\"    may be used to endorse or promote products derived from this software
20.\"    without specific prior written permission.
21.\"
22.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32.\" SUCH DAMAGE.
33.\"
34.\"     From: @(#)tcp.4	8.1 (Berkeley) 6/5/93
35.\" $FreeBSD$
36.\"
37.Dd July 20, 2022
38.Dt TCP 4
39.Os
40.Sh NAME
41.Nm tcp
42.Nd Internet Transmission Control Protocol
43.Sh SYNOPSIS
44.In sys/types.h
45.In sys/socket.h
46.In netinet/in.h
47.In netinet/tcp.h
48.Ft int
49.Fn socket AF_INET SOCK_STREAM 0
50.Sh DESCRIPTION
51The
52.Tn TCP
53protocol provides reliable, flow-controlled, two-way
54transmission of data.
55It is a byte-stream protocol used to
56support the
57.Dv SOCK_STREAM
58abstraction.
59.Tn TCP
60uses the standard
61Internet address format and, in addition, provides a per-host
62collection of
63.Dq "port addresses" .
64Thus, each address is composed
65of an Internet address specifying the host and network,
66with a specific
67.Tn TCP
68port on the host identifying the peer entity.
69.Pp
70Sockets utilizing the
71.Tn TCP
72protocol are either
73.Dq active
74or
75.Dq passive .
76Active sockets initiate connections to passive
77sockets.
78By default,
79.Tn TCP
80sockets are created active; to create a
81passive socket, the
82.Xr listen 2
83system call must be used
84after binding the socket with the
85.Xr bind 2
86system call.
87Only passive sockets may use the
88.Xr accept 2
89call to accept incoming connections.
90Only active sockets may use the
91.Xr connect 2
92call to initiate connections.
93.Pp
94Passive sockets may
95.Dq underspecify
96their location to match
97incoming connection requests from multiple networks.
98This technique, termed
99.Dq "wildcard addressing" ,
100allows a single
101server to provide service to clients on multiple networks.
102To create a socket which listens on all networks, the Internet
103address
104.Dv INADDR_ANY
105must be bound.
106The
107.Tn TCP
108port may still be specified
109at this time; if the port is not specified, the system will assign one.
110Once a connection has been established, the socket's address is
111fixed by the peer entity's location.
112The address assigned to the
113socket is the address associated with the network interface
114through which packets are being transmitted and received.
115Normally, this address corresponds to the peer entity's network.
116.Pp
117.Tn TCP
118supports a number of socket options which can be set with
119.Xr setsockopt 2
120and tested with
121.Xr getsockopt 2 :
122.Bl -tag -width ".Dv TCP_FUNCTION_BLK"
123.It Dv TCP_INFO
124Information about a socket's underlying TCP session may be retrieved
125by passing the read-only option
126.Dv TCP_INFO
127to
128.Xr getsockopt 2 .
129It accepts a single argument: a pointer to an instance of
130.Vt "struct tcp_info" .
131.Pp
132This API is subject to change; consult the source to determine
133which fields are currently filled out by this option.
134.Fx
135specific additions include
136send window size,
137receive window size,
138and
139bandwidth-controlled window space.
140.It Dv TCP_CCALGOOPT
141Set or query congestion control algorithm specific parameters.
142See
143.Xr mod_cc 4
144for details.
145.It Dv TCP_CONGESTION
146Select or query the congestion control algorithm that TCP will use for the
147connection.
148See
149.Xr mod_cc 4
150for details.
151.It Dv TCP_FASTOPEN
152Enable or disable TCP Fast Open (TFO).
153To use this option, the kernel must be built with the
154.Dv TCP_RFC7413
155option.
156.Pp
157This option can be set on the socket either before or after the
158.Xr listen 2
159is invoked.
160Clearing this option on a listen socket after it has been set has no effect on
161existing TFO connections or TFO connections in progress; it only prevents new
162TFO connections from being established.
163.Pp
164For passively-created sockets, the
165.Dv TCP_FASTOPEN
166socket option can be queried to determine whether the connection was established
167using TFO.
168Note that connections that are established via a TFO
169.Tn SYN ,
170but that fall back to using a non-TFO
171.Tn SYN|ACK
172will have the
173.Dv TCP_FASTOPEN
174socket option set.
175.Pp
176In addition to the facilities defined in RFC7413, this implementation supports a
177pre-shared key (PSK) mode of operation in which the TFO server requires the
178client to be in posession of a shared secret in order for the client to be able
179to successfully open TFO connections with the server.
180This is useful, for example, in environments where TFO servers are exposed to
181both internal and external clients and only wish to allow TFO connections from
182internal clients.
183.Pp
184In the PSK mode of operation, the server generates and sends TFO cookies to
185requesting clients as usual.
186However, when validating cookies received in TFO SYNs from clients, the server
187requires the client-supplied cookie to equal
188.Bd -literal -offset left
189SipHash24(key=\fI16-byte-psk\fP, msg=\fIcookie-sent-to-client\fP)
190.Ed
191.Pp
192Multiple concurrent valid pre-shared keys are supported so that time-based
193rolling PSK invalidation policies can be implemented in the system.
194The default number of concurrent pre-shared keys is 2.
195.Pp
196This can be adjusted with the
197.Dv TCP_RFC7413_MAX_PSKS
198kernel option.
199.It Dv TCP_FUNCTION_BLK
200Select or query the set of functions that TCP will use for this connection.
201This allows a user to select an alternate TCP stack.
202The alternate TCP stack must already be loaded in the kernel.
203To list the available TCP stacks, see
204.Va functions_available
205in the
206.Sx MIB Variables
207section further down.
208To list the default TCP stack, see
209.Va functions_default
210in the
211.Sx MIB Variables
212section.
213.It Dv TCP_KEEPINIT
214This
215.Xr setsockopt 2
216option accepts a per-socket timeout argument of
217.Vt "u_int"
218in seconds, for new, non-established
219.Tn TCP
220connections.
221For the global default in milliseconds see
222.Va keepinit
223in the
224.Sx MIB Variables
225section further down.
226.It Dv TCP_KEEPIDLE
227This
228.Xr setsockopt 2
229option accepts an argument of
230.Vt "u_int"
231for the amount of time, in seconds, that the connection must be idle
232before keepalive probes (if enabled) are sent for the connection of this
233socket.
234If set on a listening socket, the value is inherited by the newly created
235socket upon
236.Xr accept 2 .
237For the global default in milliseconds see
238.Va keepidle
239in the
240.Sx MIB Variables
241section further down.
242.It Dv TCP_KEEPINTVL
243This
244.Xr setsockopt 2
245option accepts an argument of
246.Vt "u_int"
247to set the per-socket interval, in seconds, between keepalive probes sent
248to a peer.
249If set on a listening socket, the value is inherited by the newly created
250socket upon
251.Xr accept 2 .
252For the global default in milliseconds see
253.Va keepintvl
254in the
255.Sx MIB Variables
256section further down.
257.It Dv TCP_KEEPCNT
258This
259.Xr setsockopt 2
260option accepts an argument of
261.Vt "u_int"
262and allows a per-socket tuning of the number of probes sent, with no response,
263before the connection will be dropped.
264If set on a listening socket, the value is inherited by the newly created
265socket upon
266.Xr accept 2 .
267For the global default see the
268.Va keepcnt
269in the
270.Sx MIB Variables
271section further down.
272.It Dv TCP_NODELAY
273Under most circumstances,
274.Tn TCP
275sends data when it is presented;
276when outstanding data has not yet been acknowledged, it gathers
277small amounts of output to be sent in a single packet once
278an acknowledgement is received.
279For a small number of clients, such as window systems
280that send a stream of mouse events which receive no replies,
281this packetization may cause significant delays.
282The boolean option
283.Dv TCP_NODELAY
284defeats this algorithm.
285.It Dv TCP_MAXSEG
286By default, a sender- and
287.No receiver- Ns Tn TCP
288will negotiate among themselves to determine the maximum segment size
289to be used for each connection.
290The
291.Dv TCP_MAXSEG
292option allows the user to determine the result of this negotiation,
293and to reduce it if desired.
294.It Dv TCP_NOOPT
295.Tn TCP
296usually sends a number of options in each packet, corresponding to
297various
298.Tn TCP
299extensions which are provided in this implementation.
300The boolean option
301.Dv TCP_NOOPT
302is provided to disable
303.Tn TCP
304option use on a per-connection basis.
305.It Dv TCP_NOPUSH
306By convention, the
307.No sender- Ns Tn TCP
308will set the
309.Dq push
310bit, and begin transmission immediately (if permitted) at the end of
311every user call to
312.Xr write 2
313or
314.Xr writev 2 .
315When this option is set to a non-zero value,
316.Tn TCP
317will delay sending any data at all until either the socket is closed,
318or the internal send buffer is filled.
319.It Dv TCP_MD5SIG
320This option enables the use of MD5 digests (also known as TCP-MD5)
321on writes to the specified socket.
322Outgoing traffic is digested;
323digests on incoming traffic are verified.
324When this option is enabled on a socket, all inbound and outgoing
325TCP segments must be signed with MD5 digests.
326.Pp
327One common use for this in a
328.Fx
329router deployment is to enable
330based routers to interwork with Cisco equipment at peering points.
331Support for this feature conforms to RFC 2385.
332.Pp
333In order for this option to function correctly, it is necessary for the
334administrator to add a tcp-md5 key entry to the system's security
335associations database (SADB) using the
336.Xr setkey 8
337utility.
338This entry can only be specified on a per-host basis at this time.
339.Pp
340If an SADB entry cannot be found for the destination,
341the system does not send any outgoing segments and drops any inbound segments.
342However, during connection negotiation, a non-signed segment will be accepted if
343an SADB entry does not exist between hosts.
344When a non-signed segment is accepted, the established connection is not
345protected with MD5 digests.
346.It Dv TCP_STATS
347Manage collection of connection level statistics using the
348.Xr stats 3
349framework.
350.Pp
351Each dropped segment is taken into account in the TCP protocol statistics.
352.It Dv TCP_TXTLS_ENABLE
353Enable in-kernel Transport Layer Security (TLS) for data written to this
354socket.
355See
356.Xr ktls 4
357for more details.
358.It Dv TCP_TXTLS_MODE
359The integer argument can be used to get or set the current TLS transmit mode
360of a socket.
361See
362.Xr ktls 4
363for more details.
364.It Dv TCP_RXTLS_ENABLE
365Enable in-kernel TLS for data read from this socket.
366See
367.Xr ktls 4
368for more details.
369.It Dv TCP_REUSPORT_LB_NUMA
370Changes NUMA affinity filtering for an established TCP listen
371socket.
372This option takes a single integer argument which specifies
373the NUMA domain to filter on for this listen socket.
374The argument can also have the follwing special values:
375.Bl -tag -width "Dv TCP_REUSPORT_LB_NUMA"
376.It Dv TCP_REUSPORT_LB_NUMA_NODOM
377Remove NUMA filtering for this listen socket.
378.It Dv TCP_REUSPORT_LB_NUMA_CURDOM
379Filter traffic associated with the domain where the calling thread is
380currently executing.
381This is typically used after a process or thread inherits a listen
382socket from its parent, and sets its CPU affinity to a particular core.
383.El
384.It Dv TCP_REMOTE_UDP_ENCAPS_PORT
385Set and get the remote UDP encapsulation port.
386It can only be set on a closed TCP socket.
387.El
388.Pp
389The option level for the
390.Xr setsockopt 2
391call is the protocol number for
392.Tn TCP ,
393available from
394.Xr getprotobyname 3 ,
395or
396.Dv IPPROTO_TCP .
397All options are declared in
398.In netinet/tcp.h .
399.Pp
400Options at the
401.Tn IP
402transport level may be used with
403.Tn TCP ;
404see
405.Xr ip 4 .
406Incoming connection requests that are source-routed are noted,
407and the reverse source route is used in responding.
408.Pp
409The default congestion control algorithm for
410.Tn TCP
411is
412.Xr cc_newreno 4 .
413Other congestion control algorithms can be made available using the
414.Xr mod_cc 4
415framework.
416.Ss MIB Variables
417The
418.Tn TCP
419protocol implements a number of variables in the
420.Va net.inet.tcp
421branch of the
422.Xr sysctl 3
423MIB.
424.Bl -tag -width ".Va v6pmtud_blackhole_mss"
425.It Va always_keepalive
426Assume that
427.Dv SO_KEEPALIVE
428is set on all
429.Tn TCP
430connections, the kernel will
431periodically send a packet to the remote host to verify the connection
432is still up.
433.It Va blackhole
434If enabled, disable sending of RST when a connection is attempted
435to a port where there is not a socket accepting connections.
436See
437.Xr blackhole 4 .
438.It Va delacktime
439Maximum amount of time, in milliseconds, before a delayed ACK is sent.
440.It Va delayed_ack
441Delay ACK to try and piggyback it onto a data packet.
442.It Va do_lrd
443Enable Lost Retransmission Detection for SACK-enabled sessions, disabled by
444default.
445Under severe congestion, a retransmission can be lost which then leads to a
446mandatory Retransmission Timeout (RTO), followed by slow-start.
447LRD will try to resend the repeatedly lost packet, preventing the time-consuming
448RTO and performance reducing slow-start.
449.It Va do_prr
450Perform SACK loss recovery using the Proportional Rate Reduction (PRR) algorithm
451described in RFC6937.
452This improves the effectiveness of retransmissions particular in environments
453with ACK thinning or burst loss events, as chances to run out of the ACK clock
454are reduced, preventing lengthy and performance reducing RTO based loss recovery
455(default is true).
456.It Va do_prr_conservative
457While doing Proportional Rate Reduction, remain strictly in a packet conserving
458mode, sending only one new packet for each ACK received.
459Helpful when a misconfigured token bucket traffic policer causes persistent
460high losses leading to RTO, but reduces PRR effectiveness in more common settings
461(default is false).
462.It Va do_tcpdrain
463Flush packets in the
464.Tn TCP
465reassembly queue if the system is low on mbufs.
466.It Va ecn.enable
467Enable support for TCP Explicit Congestion Notification (ECN).
468ECN allows a TCP sender to reduce the transmission rate in order to
469avoid packet drops.
470.Bl -tag -compact
471.It 0
472Disable ECN.
473.It 1
474Allow incoming connections to request ECN.
475Outgoing connections will request ECN.
476.It 2
477Allow incoming connections to request ECN.
478Outgoing connections will not request ECN.
479(default)
480.El
481.It Va ecn.maxretries
482Number of retries (SYN or SYN/ACK retransmits) before disabling ECN on a
483specific connection.
484This is needed to help with connection establishment
485when a broken firewall is in the network path.
486.It Va fast_finwait2_recycle
487Recycle
488.Tn TCP
489.Dv FIN_WAIT_2
490connections faster when the socket is marked as
491.Dv SBS_CANTRCVMORE
492(no user process has the socket open, data received on
493the socket cannot be read).
494The timeout used here is
495.Va finwait2_timeout .
496.It Va fastopen.acceptany
497When non-zero, all client-supplied TFO cookies will be considered to be valid.
498The default is 0.
499.It Va fastopen.autokey
500When this and
501.Va net.inet.tcp.fastopen.server_enable
502are non-zero, a new key will be automatically generated after this specified
503seconds.
504The default is 120.
505.It Va fastopen.ccache_bucket_limit
506The maximum number of entries in a client cookie cache bucket.
507The default value can be tuned with the
508.Dv TCP_FASTOPEN_CCACHE_BUCKET_LIMIT_DEFAULT
509kernel option or by setting
510.Va net.inet.tcp.fastopen_ccache_bucket_limit
511in the
512.Xr loader 8 .
513.It Va fastopen.ccache_buckets
514The number of client cookie cache buckets.
515Read-only.
516The value can be tuned with the
517.Dv TCP_FASTOPEN_CCACHE_BUCKETS_DEFAULT
518kernel option or by setting
519.Va fastopen.ccache_buckets
520in the
521.Xr loader 8 .
522.It Va fastopen.ccache_list
523Print the client cookie cache.
524Read-only.
525.It Va fastopen.client_enable
526When zero, no new active (i.e., client) TFO connections can be created.
527On the transition from enabled to disabled, the client cookie cache is cleared
528and disabled.
529The transition from enabled to disabled does not affect any active TFO
530connections in progress; it only prevents new ones from being established.
531The default is 0.
532.It Va fastopen.keylen
533The key length in bytes.
534Read-only.
535.It Va fastopen.maxkeys
536The maximum number of keys supported.
537Read-only,
538.It Va fastopen.maxpsks
539The maximum number of pre-shared keys supported.
540Read-only.
541.It Va fastopen.numkeys
542The current number of keys installed.
543Read-only.
544.It Va fastopen.numpsks
545The current number of pre-shared keys installed.
546Read-only.
547.It Va fastopen.path_disable_time
548When a failure occurs while trying to create a new active (i.e., client) TFO
549connection, new active connections on the same path, as determined by the tuple
550.Brq client_ip, server_ip, server_port ,
551will be forced to be non-TFO for this many seconds.
552Note that the path disable mechanism relies on state stored in client cookie
553cache entries, so it is possible for the disable time for a given path to be
554reduced if the corresponding client cookie cache entry is reused due to resource
555pressure before the disable period has elapsed.
556The default is
557.Dv TCP_FASTOPEN_PATH_DISABLE_TIME_DEFAULT .
558.It Va fastopen.psk_enable
559When non-zero, pre-shared key (PSK) mode is enabled for all TFO servers.
560On the transition from enabled to disabled, all installed pre-shared keys are
561removed.
562The default is 0.
563.It Va fastopen.server_enable
564When zero, no new passive (i.e., server) TFO connections can be created.
565On the transition from enabled to disabled, all installed keys and pre-shared
566keys are removed.
567On the transition from disabled to enabled, if
568.Va fastopen.autokey
569is non-zero and there are no keys installed, a new key will be generated
570immediately.
571The transition from enabled to disabled does not affect any passive TFO
572connections in progress; it only prevents new ones from being established.
573The default is 0.
574.It Va fastopen.setkey
575Install a new key by writing
576.Va net.inet.tcp.fastopen.keylen
577bytes to this sysctl.
578.It Va fastopen.setpsk
579Install a new pre-shared key by writing
580.Va net.inet.tcp.fastopen.keylen
581bytes to this sysctl.
582.It Va finwait2_timeout
583Timeout to use for fast recycling of
584.Tn TCP
585.Dv FIN_WAIT_2
586connections
587.Pq Va fast_finwait2_recycle .
588Defaults to 60 seconds.
589.It Va functions_available
590List of available TCP function blocks (TCP stacks).
591.It Va functions_default
592The default TCP function block (TCP stack).
593.It Va functions_inherit_listen_socket_stack
594Determines whether to inherit listen socket's TCP stack or use the current
595system default TCP stack, as defined by
596.Va functions_default .
597Default is true.
598.It Va hostcache
599The TCP host cache is used to cache connection details and metrics to
600improve future performance of connections between the same hosts.
601At the completion of a TCP connection, a host will cache information
602for the connection for some defined period of time.
603There are a number of
604.Va hostcache
605variables under this node.
606See
607.Va hostcache.enable .
608.It Va hostcache.bucketlimit
609The maximum number of entries for the same hash.
610Defaults to 30.
611.It Va hostcache.cachelimit
612Overall entry limit for hostcache.
613Defaults to
614.Va hashsize
615*
616.Va bucketlimit .
617.It Va hostcache.count
618The current number of entries in the host cache.
619.It Va hostcache.enable
620Enable/disable the host cache:
621.Bl -tag -compact
622.It 0
623Disable the host cache.
624.It 1
625Enable the host cache. (default)
626.El
627.It Va hostcache.expire
628Time in seconds, how long a entry should be kept in the
629host cache since last accessed.
630Defaults to 3600 (1 hour).
631.It Va hostcache.hashsize
632Size of TCP hostcache hashtable.
633This number has to be a power of two, or will be rejected.
634Defaults to 512.
635.It Va hostcache.histo
636Provide a Histogram of the hostcache hash utilization.
637.It Va hostcache.list
638Provide a complete list of all current entries in the host
639cache.
640.It Va hostcache.prune
641Time in seconds between pruning expired host cache entries.
642Defaults to 300 (5 minutes).
643.It Va hostcache.purge
644Expire all entires on next pruning of host cache entries.
645Any non-zero setting will be reset to zero, once the purge
646is running.
647.Bl -tag -compact
648.It 0
649Do not purge all entries when pruning the host cache (default).
650.It 1
651Purge all entries when doing the next pruning.
652.It 2
653Purge all entries and also reseed the hash salt.
654.El
655.It Va hostcache.purgenow
656Immediately purge all entries once set to any value.
657Setting this to 2 will also reseed the hash salt.
658.It Va icmp_may_rst
659Certain
660.Tn ICMP
661unreachable messages may abort connections in
662.Tn SYN-SENT
663state.
664.It Va initcwnd_segments
665Enable the ability to specify initial congestion window in number of segments.
666The default value is 10 as suggested by RFC 6928.
667Changing the value on the fly would not affect connections
668using congestion window from the hostcache.
669Caution:
670This regulates the burst of packets allowed to be sent in the first RTT.
671The value should be relative to the link capacity.
672Start with small values for lower-capacity links.
673Large bursts can cause buffer overruns and packet drops if routers have small
674buffers or the link is experiencing congestion.
675.It Va insecure_rst
676Use criteria defined in RFC793 instead of RFC5961 for accepting RST segments.
677Default is false.
678.It Va insecure_syn
679Use criteria defined in RFC793 instead of RFC5961 for accepting SYN segments.
680Default is false.
681.It Va isn_reseed_interval
682The interval (in seconds) specifying how often the secret data used in
683RFC 1948 initial sequence number calculations should be reseeded.
684By default, this variable is set to zero, indicating that
685no reseeding will occur.
686Reseeding should not be necessary, and will break
687.Dv TIME_WAIT
688recycling for a few minutes.
689.It Va keepcnt
690Number of keepalive probes sent, with no response, before a connection
691is dropped.
692The default is 8 packets.
693.It Va keepidle
694Amount of time, in milliseconds, that the connection must be idle
695before sending keepalive probes (if enabled).
696The default is 7200000 msec (7.2M msec, 2 hours).
697.It Va keepinit
698Timeout, in milliseconds, for new, non-established
699.Tn TCP
700connections.
701The default is 75000 msec (75K msec, 75 sec).
702.It Va keepintvl
703The interval, in milliseconds, between keepalive probes sent to remote
704machines, when no response is received on a
705.Va keepidle
706probe.
707The default is 75000 msec (75K msec, 75 sec).
708.It Va log_in_vain
709Log any connection attempts to ports where there is not a socket
710accepting connections.
711The value of 1 limits the logging to
712.Tn SYN
713(connection establishment) packets only.
714A value of 2 results in any
715.Tn TCP
716packets to closed ports being logged.
717Any value not listed above disables the logging
718(default is 0, i.e., the logging is disabled).
719.It Va maxtcptw
720When a TCP connection enters the
721.Dv TIME_WAIT
722state, its associated socket structure is freed, since it is of
723negligible size and use, and a new structure is allocated to contain a
724minimal amount of information necessary for sustaining a connection in
725this state, called the compressed TCP
726.Dv TIME_WAIT
727state.
728Since this structure is smaller than a socket structure, it can save
729a significant amount of system memory.
730The
731.Va net.inet.tcp.maxtcptw
732MIB variable controls the maximum number of these structures allocated.
733By default, it is initialized to
734.Va kern.ipc.maxsockets
735/ 5.
736.It Va msl
737The Maximum Segment Lifetime, in milliseconds, for a packet.
738.It Va mssdflt
739The default value used for the maximum segment size
740.Pq Dq MSS
741when no advice to the contrary is received from MSS negotiation.
742.It Va newcwd
743Enable the New Congestion Window Validation mechanism as described in RFC 7661.
744This gently reduces the congestion window during periods, where TCP is
745application limited and the network bandwidth is not utilized completely.
746That prevents self-inflicted packet losses once the application starts to
747transmit data at a higher speed.
748.It Va nolocaltimewait
749Suppress creation of compressed TCP
750.Dv TIME_WAIT
751states for connections in
752which both endpoints are local.
753.It Va path_mtu_discovery
754Enable Path MTU Discovery.
755.It Va pcbcount
756Number of active process control blocks
757(read-only).
758.It Va perconn_stats_enable
759Controls the default collection of statistics for all connections using the
760.Xr stats 3
761framework.
7620 disables, 1 enables, 2 enables random sampling across log id connection
763groups with all connections in a group receiving the same setting.
764.It Va perconn_stats_sample_rates
765A CSV list of template_spec=percent key-value pairs which controls the per
766template sampling rates when
767.Xr stats 3
768sampling is enabled.
769.It Va pmtud_blackhole_detection
770Enable automatic path MTU blackhole detection.
771In case of retransmits of MSS sized segments,
772the OS will lower the MSS to check if it's an MTU problem.
773If the current MSS is greater than the configured value to try
774.Po Va net.inet.tcp.pmtud_blackhole_mss
775and
776.Va net.inet.tcp.v6pmtud_blackhole_mss
777.Pc ,
778it will be set to this value, otherwise,
779the MSS will be set to the default values
780.Po Va net.inet.tcp.mssdflt
781and
782.Va net.inet.tcp.v6mssdflt
783.Pc .
784Settings:
785.Bl -tag -compact
786.It 0
787Disable path MTU blackhole detection.
788.It 1
789Enable path MTU blackhole detection for IPv4 and IPv6.
790.It 2
791Enable path MTU blackhole detection only for IPv4.
792.It 3
793Enable path MTU blackhole detection only for IPv6.
794.El
795.It Va pmtud_blackhole_mss
796MSS to try for IPv4 if PMTU blackhole detection is turned on.
797.It Va reass.cursegments
798The current total number of segments present in all reassembly queues.
799.It Va reass.maxqueuelen
800The maximum number of segments allowed in each reassembly queue.
801By default, the system chooses a limit based on each TCP connection's
802receive buffer size and maximum segment size (MSS).
803The actual limit applied to a session's reassembly queue will be the lower of
804the system-calculated automatic limit and the user-specified
805.Va reass.maxqueuelen
806limit.
807.It Va reass.maxsegments
808The maximum limit on the total number of segments across all reassembly
809queues.
810The limit can be adjusted as a tunable.
811.It Va recvspace
812Maximum
813.Tn TCP
814receive window.
815.It Va rexmit_initial , rexmit_min , rexmit_slop
816Adjust the retransmit timer calculation for
817.Tn TCP .
818The slop is
819typically added to the raw calculation to take into account
820occasional variances that the
821.Tn SRTT
822(smoothed round-trip time)
823is unable to accommodate, while the minimum specifies an
824absolute minimum.
825While a number of
826.Tn TCP
827RFCs suggest a 1
828second minimum, these RFCs tend to focus on streaming behavior,
829and fail to deal with the fact that a 1 second minimum has severe
830detrimental effects over lossy interactive connections, such
831as a 802.11b wireless link, and over very fast but lossy
832connections for those cases not covered by the fast retransmit
833code.
834For this reason, we use 200ms of slop and a near-0
835minimum, which gives us an effective minimum of 200ms (similar to
836.Tn Linux ) .
837The initial value is used before an RTT measurement has been performed.
838.It Va rfc1323
839Implement the window scaling and timestamp options of RFC 1323/RFC 7323
840(default is true).
841.It Va rfc3042
842Enable the Limited Transmit algorithm as described in RFC 3042.
843It helps avoid timeouts on lossy links and also when the congestion window
844is small, as happens on short transfers.
845.It Va rfc3390
846Enable support for RFC 3390, which allows for a variable-sized
847starting congestion window on new connections, depending on the
848maximum segment size.
849This helps throughput in general, but
850particularly affects short transfers and high-bandwidth large
851propagation-delay connections.
852.It Va rfc6675_pipe
853Deprecated and superseded by
854.Va sack.revised
855.It Va sack.enable
856Enable support for RFC 2018, TCP Selective Acknowledgment option,
857which allows the receiver to inform the sender about all successfully
858arrived segments, allowing the sender to retransmit the missing segments
859only.
860.It Va sack.globalmaxholes
861Maximum number of SACK holes per system, across all connections.
862Defaults to 65536.
863.It Va sack.maxholes
864Maximum number of SACK holes per connection.
865Defaults to 128.
866.It Va sack.revised
867Enables three updated mechanisms from RFC6675 (default is true).
868Calculate the bytes in flight using the algorithm described in RFC 6675, and
869is also an improvement when Proportional Rate Reduction is enabled.
870Next, Rescue Retransmission helps timely loss recovery, when the trailing segments
871of a transmission are lost, while no additional data is ready to be sent.
872In case a partial ACK without a SACK block is received during SACK loss
873recovery, the trailing segment is immediately resent, rather than waiting
874for a Retransmission timeout.
875Finally, SACK loss recovery is also engaged, once two segments plus one byte are
876SACKed - even if no traditional duplicate ACKs were observed.
877.It Va sendspace
878Maximum
879.Tn TCP
880send window.
881.It Va syncookies
882Determines whether or not
883.Tn SYN
884cookies should be generated for outbound
885.Tn SYN-ACK
886packets.
887.Tn SYN
888cookies are a great help during
889.Tn SYN
890flood attacks, and are enabled by default.
891(See
892.Xr syncookies 4 . )
893.It Va tcbhashsize
894Size of the
895.Tn TCP
896control-block hash table
897(read-only).
898This is tuned using the kernel option
899.Dv TCBHASHSIZE
900or by setting
901.Va net.inet.tcp.tcbhashsize
902in the
903.Xr loader 8 .
904.It Va tolerate_missing_ts
905Tolerate the missing of timestamps (RFC 1323/RFC 7323) for
906.Tn TCP
907segments belonging to
908.Tn TCP
909connections for which support of
910.Tn TCP
911timestamps has been negotiated.
912As of June 2021, several TCP stacks are known to violate RFC 7323, including
913modern widely deployed ones.
914Therefore the default is 1, i.e., the missing of timestamps is tolerated.
915.It Va ts_offset_per_conn
916When initializing the TCP timestamps, use a per connection offset instead of a
917per host pair offset.
918Default is to use per connection offsets as recommended in RFC 7323.
919.It Va udp_tunneling_overhead
920The overhead taken into account when using UDP encapsulation.
921Since MSS clamping by middleboxes will most likely not work, values larger than
9228 (the size of the UDP header) are also supported.
923Supported values are between 8 and 1024.
924The default is 8.
925.It Va udp_tunneling_port
926The local UDP encapsulation port.
927A value of 0 indicates that UDP encapsulation is disabled.
928The default is 0.
929.It Va v6pmtud_blackhole_mss
930MSS to try for IPv6 if PMTU blackhole detection is turned on.
931See
932.Va pmtud_blackhole_detection .
933.El
934.Sh ERRORS
935A socket operation may fail with one of the following errors returned:
936.Bl -tag -width Er
937.It Bq Er EISCONN
938when trying to establish a connection on a socket which
939already has one;
940.It Bo Er ENOBUFS Bc or Bo Er ENOMEM Bc
941when the system runs out of memory for
942an internal data structure;
943.It Bq Er ETIMEDOUT
944when a connection was dropped
945due to excessive retransmissions;
946.It Bq Er ECONNRESET
947when the remote peer
948forces the connection to be closed;
949.It Bq Er ECONNREFUSED
950when the remote
951peer actively refuses connection establishment (usually because
952no process is listening to the port);
953.It Bq Er EADDRINUSE
954when an attempt
955is made to create a socket with a port which has already been
956allocated;
957.It Bq Er EADDRNOTAVAIL
958when an attempt is made to create a
959socket with a network address for which no network interface
960exists;
961.It Bq Er EAFNOSUPPORT
962when an attempt is made to bind or connect a socket to a multicast
963address.
964.It Bq Er EINVAL
965when trying to change TCP function blocks at an invalid point in the session;
966.It Bq Er ENOENT
967when trying to use a TCP function block that is not available;
968.El
969.Sh SEE ALSO
970.Xr getsockopt 2 ,
971.Xr socket 2 ,
972.Xr stats 3 ,
973.Xr sysctl 3 ,
974.Xr blackhole 4 ,
975.Xr inet 4 ,
976.Xr intro 4 ,
977.Xr ip 4 ,
978.Xr ktls 4 ,
979.Xr mod_cc 4 ,
980.Xr siftr 4 ,
981.Xr syncache 4 ,
982.Xr tcp_bbr 4 ,
983.Xr setkey 8 ,
984.Xr tcp_functions 9
985.Rs
986.%A "V. Jacobson"
987.%A "B. Braden"
988.%A "D. Borman"
989.%T "TCP Extensions for High Performance"
990.%O "RFC 1323"
991.Re
992.Rs
993.%A "D. Borman"
994.%A "B. Braden"
995.%A "V. Jacobson"
996.%A "R. Scheffenegger"
997.%T "TCP Extensions for High Performance"
998.%O "RFC 7323"
999.Re
1000.Rs
1001.%A "A. Heffernan"
1002.%T "Protection of BGP Sessions via the TCP MD5 Signature Option"
1003.%O "RFC 2385"
1004.Re
1005.Rs
1006.%A "K. Ramakrishnan"
1007.%A "S. Floyd"
1008.%A "D. Black"
1009.%T "The Addition of Explicit Congestion Notification (ECN) to IP"
1010.%O "RFC 3168"
1011.Re
1012.Sh HISTORY
1013The
1014.Tn TCP
1015protocol appeared in
1016.Bx 4.2 .
1017The RFC 1323 extensions for window scaling and timestamps were added
1018in
1019.Bx 4.4 .
1020The
1021.Dv TCP_INFO
1022option was introduced in
1023.Tn Linux 2.6
1024and is
1025.Em subject to change .
1026