xref: /freebsd/sbin/sysctl/sysctl.8 (revision 5d3e7166)
1.\" Copyright (c) 1993
2.\"	The Regents of the University of California.  All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\" 3. Neither the name of the University nor the names of its contributors
13.\"    may be used to endorse or promote products derived from this software
14.\"    without specific prior written permission.
15.\"
16.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
17.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
20.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26.\" SUCH DAMAGE.
27.\"
28.\"	From: @(#)sysctl.8	8.1 (Berkeley) 6/6/93
29.\" $FreeBSD$
30.\"
31.Dd December 24, 2022
32.Dt SYSCTL 8
33.Os
34.Sh NAME
35.Nm sysctl
36.Nd get or set kernel state
37.Sh SYNOPSIS
38.Nm
39.Op Fl bdeFhilNnoTtqWx
40.Op Fl B Ar bufsize
41.Op Fl f Ar filename
42.Ar name Ns Op = Ns Ar value Ns Op , Ns Ar value
43.Ar ...
44.Nm
45.Op Fl bdeFhlNnoTtqWx
46.Op Fl B Ar bufsize
47.Fl a
48.Sh DESCRIPTION
49The
50.Nm
51utility retrieves kernel state and allows processes with appropriate
52privilege to set kernel state.
53The state to be retrieved or set is described using a
54.Dq Management Information Base
55.Pq Dq MIB
56style name, described as a dotted set of
57components.
58.Pp
59The following options are available:
60.Bl -tag -width indent
61.It Fl A
62Equivalent to
63.Fl o a
64(for compatibility).
65.It Fl a
66List all the currently available values except for those which are
67opaque or excluded from listing via the
68.Dv CTLFLAG_SKIP
69flag.
70This option is ignored if one or more variable names are specified on
71the command line.
72.It Fl b
73Force the value of the variable(s) to be output in raw, binary format.
74No names are printed and no terminating newlines are output.
75This is mostly useful with a single variable.
76.It Fl B Ar bufsize
77Set the buffer size to read from the
78.Nm
79to
80.Ar bufsize .
81This is necessary for a
82.Nm
83that has variable length, and the probe value of 0 is a valid length, such as
84.Va kern.arandom .
85.It Fl d
86Print the description of the variable instead of its value.
87.It Fl e
88Separate the name and the value of the variable(s) with
89.Ql = .
90This is useful for producing output which can be fed back to the
91.Nm
92utility.
93This option is ignored if either
94.Fl N
95or
96.Fl n
97is specified, or a variable is being set.
98.It Fl f Ar filename
99Specify a file which contains a pair of name and value in each line.
100.Nm
101reads and processes the specified file first and then processes the name
102and value pairs in the command line argument.
103.It Fl F
104Print the format of the variable.
105This is additional information to describe the type of the variable and
106most useful with struct types such as clockinfo, timeval, and loadavg.
107.It Fl h
108Format output for human, rather than machine, readability.
109.It Fl i
110Ignore unknown OIDs.
111The purpose is to make use of
112.Nm
113for collecting data from a variety of machines (not all of which
114are necessarily running exactly the same software) easier.
115.It Fl l
116Show the length of variables along with their values.
117This option cannot be combined with the
118.Fl N
119option.
120.It Fl N
121Show only variable names, not their values.
122This is particularly useful with shells that offer programmable
123completion.
124To enable completion of variable names in
125.Xr zsh 1 Pq Pa ports/shells/zsh ,
126use the following code:
127.Bd -literal -offset indent
128listsysctls () { set -A reply $(sysctl -AN ${1%.*}) }
129compctl -K listsysctls sysctl
130.Ed
131.Pp
132To enable completion of variable names in
133.Xr tcsh 1 ,
134use:
135.Pp
136.Dl "complete sysctl 'n/*/`sysctl -Na`/'"
137.It Fl n
138Do not show variable names.
139This option is useful for setting shell variables.
140For instance, to save the pagesize in variable
141.Va psize ,
142use:
143.Pp
144.Dl "set psize=`sysctl -n hw.pagesize`"
145.It Fl o
146Show opaque variables (which are normally suppressed).
147The format and length are printed, as well as a hex dump of the first
148sixteen bytes of the value.
149.It Fl q
150Suppress some warnings generated by
151.Nm
152to standard error.
153.It Fl T
154Display only variables that are settable via loader (CTLFLAG_TUN).
155.It Fl t
156Print the type of the variable.
157.It Fl W
158Display only writable variables that are not statistical.
159Useful for determining the set of runtime tunable sysctls.
160.It Fl X
161Equivalent to
162.Fl x a
163(for compatibility).
164.It Fl x
165As
166.Fl o ,
167but prints a hex dump of the entire value instead of just the first
168few bytes.
169.El
170.Pp
171The information available from
172.Nm
173consists of integers, strings, and opaque types.
174The
175.Nm
176utility
177only knows about a couple of opaque types, and will resort to hexdumps
178for the rest.
179The opaque information is much more useful if retrieved by special
180purpose programs such as
181.Xr ps 1 ,
182.Xr systat 1 ,
183and
184.Xr netstat 1 .
185.Pp
186Some of the variables which cannot be modified during normal system
187operation can be initialized via
188.Xr loader 8
189tunables.
190This can for example be done by setting them in
191.Xr loader.conf 5 .
192Please refer to
193.Xr loader.conf 5
194for more information on which tunables are available and how to set them.
195.Pp
196The string and integer information is summarized below.
197For a detailed description of these variables see
198.Xr sysctl 3 .
199.Pp
200The changeable column indicates whether a process with appropriate
201privilege can change the value.
202String and integer values can be set using
203.Nm .
204.Bl -column security.bsd.unprivileged_read_msgbuf integerxxx
205.It Sy "Name	Type	Changeable"
206.It "kern.ostype	string	no"
207.It "kern.osrelease	string	no"
208.It "kern.osrevision	integer	no"
209.It "kern.version	string	no"
210.It "kern.maxvnodes	integer	yes"
211.It "kern.maxproc	integer	no"
212.It "kern.maxprocperuid	integer	yes"
213.It "kern.maxfiles	integer	yes"
214.It "kern.maxfilesperproc	integer	yes"
215.It "kern.argmax	integer	no"
216.It "kern.securelevel	integer	raise only"
217.It "kern.hostname	string	yes"
218.It "kern.hostid	integer	yes"
219.It "kern.clockrate	struct	no"
220.It "kern.posix1version	integer	no"
221.It "kern.ngroups	integer	no"
222.It "kern.job_control	integer	no"
223.It "kern.saved_ids	integer	no"
224.It "kern.boottime	struct	no"
225.It "kern.domainname	string	yes"
226.It "kern.filedelay	integer	yes"
227.It "kern.dirdelay	integer	yes"
228.It "kern.metadelay	integer	yes"
229.It "kern.osreldate	integer	no"
230.It "kern.bootfile	string	yes"
231.It "kern.corefile	string	yes"
232.It "kern.logsigexit	integer	yes"
233.It "security.bsd.suser_enabled	integer	yes"
234.It "security.bsd.see_other_uids	integer	yes"
235.It "security.bsd.unprivileged_proc_debug	integer	yes"
236.It "security.bsd.unprivileged_read_msgbuf	integer	yes"
237.It "vm.loadavg	struct	no"
238.It "hw.machine	string	no"
239.It "hw.model	string	no"
240.It "hw.ncpu	integer	no"
241.It "hw.byteorder	integer	no"
242.It "hw.physmem	integer	no"
243.It "hw.usermem	integer	no"
244.It "hw.pagesize	integer	no"
245.It "hw.floatingpoint	integer	no"
246.It "hw.machine_arch	string	no"
247.It "hw.realmem	integer	no"
248.It "machdep.adjkerntz	integer	yes"
249.It "machdep.disable_rtc_set	integer	yes"
250.It "machdep.guessed_bootdev	string	no"
251.It "user.cs_path	string	no"
252.It "user.bc_base_max	integer	no"
253.It "user.bc_dim_max	integer	no"
254.It "user.bc_scale_max	integer	no"
255.It "user.bc_string_max	integer	no"
256.It "user.coll_weights_max	integer	no"
257.It "user.expr_nest_max	integer	no"
258.It "user.line_max	integer	no"
259.It "user.re_dup_max	integer	no"
260.It "user.posix2_version	integer	no"
261.It "user.posix2_c_bind	integer	no"
262.It "user.posix2_c_dev	integer	no"
263.It "user.posix2_char_term	integer	no"
264.It "user.posix2_fort_dev	integer	no"
265.It "user.posix2_fort_run	integer	no"
266.It "user.posix2_localedef	integer	no"
267.It "user.posix2_sw_dev	integer	no"
268.It "user.posix2_upe	integer	no"
269.It "user.stream_max	integer	no"
270.It "user.tzname_max	integer	no"
271.It "user.localbase	string	no"
272.El
273.Sh FILES
274.Bl -tag -width ".In netinet/icmp_var.h" -compact
275.It In sys/sysctl.h
276definitions for top level identifiers, second level kernel and hardware
277identifiers, and user level identifiers
278.It In sys/socket.h
279definitions for second level network identifiers
280.It In sys/gmon.h
281definitions for third level profiling identifiers
282.It In vm/vm_param.h
283definitions for second level virtual memory identifiers
284.It In netinet/in.h
285definitions for third level Internet identifiers and
286fourth level IP identifiers
287.It In netinet/icmp_var.h
288definitions for fourth level ICMP identifiers
289.It In netinet/udp_var.h
290definitions for fourth level UDP identifiers
291.El
292.Sh EXIT STATUS
293.Ex -std
294.Sh EXAMPLES
295For example, to retrieve the maximum number of processes allowed
296in the system, one would use the following request:
297.Pp
298.Dl "sysctl kern.maxproc"
299.Pp
300To set the maximum number of processes allowed
301per uid to 1000, one would use the following request:
302.Pp
303.Dl "sysctl kern.maxprocperuid=1000"
304.Pp
305Information about the system clock rate may be obtained with:
306.Pp
307.Dl "sysctl kern.clockrate"
308.Pp
309Information about the load average history may be obtained with:
310.Pp
311.Dl "sysctl vm.loadavg"
312.Pp
313More variables than these exist, and the best and likely only place
314to search for their deeper meaning is undoubtedly the source where
315they are defined.
316.Sh COMPATIBILITY
317The
318.Fl w
319option has been deprecated and is silently ignored.
320.Sh SEE ALSO
321.Xr sysctl 3 ,
322.Xr loader.conf 5 ,
323.Xr sysctl.conf 5 ,
324.Xr loader 8
325.Sh HISTORY
326A
327.Nm
328utility first appeared in
329.Bx 4.4 .
330.Pp
331In
332.Fx 2.2 ,
333.Nm
334was significantly remodeled.
335.Sh BUGS
336The
337.Nm
338utility presently exploits an undocumented interface to the kernel
339.Xr sysctl 9
340facility to traverse the sysctl tree and to retrieve format
341and name information.
342This correct interface is being thought about for the time being.
343