xref: /netbsd-src/share/man/man7/sysctl.7 (revision 6deb2c22d20de1d75d538e8a5c57b573926fd157)
1.\"	$NetBSD: sysctl.7,v 1.27 2009/10/05 10:47:52 wiz Exp $
2.\"
3.\" Copyright (c) 1993
4.\"	The Regents of the University of California.  All rights reserved.
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.\" 3. Neither the name of the University nor the names of its contributors
15.\"    may be used to endorse or promote products derived from this software
16.\"    without specific prior written permission.
17.\"
18.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
19.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
22.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
24.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
25.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
26.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
27.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28.\" SUCH DAMAGE.
29.\"
30.\"	@(#)sysctl.3	8.4 (Berkeley) 5/9/95
31.\"
32.Dd October 4, 2009
33.Dt SYSCTL 7
34.Os
35.Sh NAME
36.Nm sysctl
37.Nd system information variables
38.Sh DESCRIPTION
39The
40.Xr sysctl 3
41library function and the
42.Xr sysctl 8
43utility are used to get and set values of system variables, maintained
44by the kernel.
45The variables are organized in a tree and identified by a sequence of
46numbers, conventionally separated by dots with the topmost identifier
47at the left side.
48The numbers have corresponding text names.
49The
50.Xr sysctlnametomib 3
51function or the
52.Fl M
53argument to the
54.Xr sysctl 8
55utility can be used to convert the text representation to the
56numeric one.
57.Pp
58The individual sysctl variables are described below, both the textual
59and numeric form where applicable.
60The textual names can be used as argument to the
61.Xr sysctl 8
62utility and in the file
63.Pa /etc/sysctl.conf .
64The numeric names are usually defined as preprocessor constants and
65are intended for use by programs.
66Every such constant expands to one integer, which identifies the
67sysctl variable relative to the upper level of the tree.
68See the
69.Xr sysctl 3
70manual page for programming examples.
71.Sh Top level names
72The top level names are defined with a CTL_ prefix in
73.Aq Pa sys/sysctl.h ,
74and are as follows.
75The next and subsequent levels down are found in the include files
76listed here, and described in separate sections below.
77.Bl -column security CTL_SECURITY "Next level names" "High kernel limits"
78.It Sy Name	Constant	Next level names	Description
79.It kern	CTL_KERN	sys/sysctl.h	High kernel limits
80.It vm	CTL_VM	uvm/uvm_param.h	Virtual memory
81.It vfs	CTL_VFS	sys/mount.h	Filesystem
82.It net	CTL_NET	sys/socket.h	Networking
83.It debug	CTL_DEBUG	sys/sysctl.h	Debugging
84.It hw	CTL_HW	sys/sysctl.h	Generic CPU, I/O
85.It machdep	CTL_MACHDEP	sys/sysctl.h	Machine dependent
86.It user	CTL_USER	sys/sysctl.h	User-level
87.It ddb	CTL_DDB	sys/sysctl.h	In-kernel debugger
88.It proc	CTL_PROC	sys/sysctl.h	Per-process
89.It vendor	CTL_VENDOR	?	Vendor specific
90.It emul	CTL_EMUL	sys/sysctl.h	Emulation settings
91.It security	CTL_SECURITY	sys/sysctl.h	Security settings
92.El
93.Sh The debug.* subtree
94The debugging variables vary from system to system.
95A debugging variable may be added or deleted without need to recompile
96.Nm
97to know about it.
98Each time it runs,
99.Nm
100gets the list of debugging variables from the kernel and
101displays their current values.
102The system defines twenty
103.Va ( struct ctldebug )
104variables named
105.Dv debug0
106through
107.Dv debug19 .
108They are declared as separate variables so that they can be
109individually initialized at the location of their associated variable.
110The loader prevents multiple use of the same variable by issuing errors
111if a variable is initialized in more than one place.
112For example, to export the variable
113.Dv dospecialcheck
114as a debugging variable, the following declaration would be used:
115.Bd -literal -offset indent -compact
116int dospecialcheck = 1;
117struct ctldebug debug5 = { "dospecialcheck", \*[Am]dospecialcheck };
118.Ed
119.Pp
120Note that the dynamic implementation of
121.Nm
122currently in use largely makes this particular
123.Nm
124interface obsolete.
125See
126.Xr sysctl 8
127.\" and
128.\" .Xr sysctl 9
129for more information.
130.Sh The vfs.* subtree
131A distinguished second level name,
132.Li vfs.generic ( VFS_GENERIC ) ,
133is used to get general information about all filesystems.
134It has the following third level identifiers:
135.Bl -tag -width compact
136.It vfs.generic.maxtypenum ( VFS_MAXTYPENUM )
137The highest valid filesystem type number.
138.It vfs.generic.conf ( VFS_CONF )
139Returns configuration information about the file-system type given as a fourth
140level identifier.
141.El
142.Pp
143The remaining second level identifiers are the file-system names, identified
144by the type number returned by a
145.Xr statvfs 2
146call or from
147.Li vfs.generic.conf .
148The third level identifiers available for each filesystem
149are given in the header file that defines the mount
150argument structure for that filesystem.
151.Sh The hw.* subtree
152The string and integer information available for the
153.Li hw
154level is detailed below.
155The changeable column shows whether a process with appropriate
156privilege may change the value.
157.Bl -column "hw.acpi.supported_states" "integer" "Changeable" -offset indent
158.It Sy Second level name	Type	Changeable
159.It hw.acpi.supported_states	string	no
160.It hw.alignbytes	integer	no
161.It hw.byteorder	integer	no
162.It hw.cnmagic	string	yes
163.It hw.disknames	string	no
164.It hw.diskstats	struct	no
165.It hw.machine	string	no
166.It hw.machine_arch	string	no
167.It hw.model	string	no
168.It hw.ncpu	integer	no
169.It hw.pagesize	integer	no
170.It hw.physmem	integer	no
171.It hw.physmem64	quad	no
172.It hw.usermem	integer	no
173.It hw.usermem64	quad	no
174.El
175.Pp
176.Bl -tag -width "123456"
177.It Li hw.acpi.support_states
178List of possible
179.Tn ACPI
180sleep states.
181The list can contain the following values:
182.Bl -tag -width XS1X
183.It S0
184fully running
185.It S1
186power on suspend (CPU and hard disks are off)
187.It S2
188similar to S3, usually not implemented
189.It S3
190suspend-to-RAM
191.It S4
192suspend-to-disk (needs BIOS support)
193.It S5
194power off
195.El
196.It Li hw.alignbytes ( HW_ALIGNBYTES )
197Alignment constraint for all possible data types.
198This shows the value
199.Dv ALIGNBYTES
200in
201.Pa /usr/include/machine/param.h ,
202at the kernel compilation time.
203.It Li hw.byteorder ( HW_BYTEORDER )
204The byteorder (4,321, or 1,234).
205.It Li hw.cnmagic ( HW_CNMAGIC )
206The console magic key sequence.
207.It Li hw.disknames ( HW_DISKNAMES )
208The list of (space separated) disk device names on the system.
209.It Li hw.iostatnames ( HW_IOSTATNAMES )
210A space separated list of devices that will have I/O statistics
211collected on them.
212.It Li hw.iostats ( HW_IOSTATS )
213Return statistical information on the NFS mounts, disk and tape
214devices on the system.
215An array of
216.Va struct io_sysctl
217structures is returned,
218whose size depends on the current number of such objects in the system.
219The third level name is the size of the
220.Va struct io_sysctl .
221The type of object can be determined by examining the
222.Va type
223element of
224.Va struct io_sysctl .
225Which can be
226.Dv IOSTAT_DISK
227(disk drive),
228.Dv IOSTAT_TAPE
229(tape drive), or
230.Dv IOSTAT_NFS
231(NFS mount).
232.It Li hw.machine ( HW_MACHINE )
233The machine class.
234.It Li hw.machine_arch ( HW_MACHINE_ARCH )
235The machine CPU class.
236.It Li hw.model ( HW_MODEL )
237The machine model.
238.It Li hw.ncpu ( HW_NCPU )
239The number of CPUs.
240.It Li hw.pagesize ( HW_PAGESIZE )
241The software page size.
242.It Li hw.physmem ( HW_PHYSMEM )
243The bytes of physical memory as a 32-bit integer.
244.It Li hw.physmem64 ( HW_PHYSMEM64 )
245The bytes of physical memory as a 64-bit integer.
246.It Li hw.usermem ( HW_USERMEM )
247The bytes of non-kernel memory as a 32-bit integer.
248.It Li hw.usermem64 ( HW_USERMEM64 )
249The bytes of non-kernel memory as a 64-bit integer.
250.El
251.Sh The kern.* subtree
252The string and integer information available for the
253.Li kern
254level is detailed below.
255The changeable column shows whether a process with appropriate
256privilege may change the value.
257The types of data currently available are process information,
258system vnodes, the open file entries, routing table entries,
259virtual memory statistics, load average history, and clock rate
260information.
261.Bl -column "kern.posix_reader_writer_locks" "struct kinfo_drivers" "not applicable"
262.It Sy Second level name	Type	Changeable
263.It kern.argmax	integer	no
264.It kern.autonicetime	integer	yes
265.It kern.autoniceval	integer	yes
266.It kern.boottime	struct timeval	no
267.It kern.bufq	node	not applicable
268.It kern.ccpu	integer	no
269.It kern.clockrate	struct clockinfo	no
270.It kern.consdev	integer	no
271.It kern.cp_id	struct	no
272.It kern.cp_time	uint64_t[\|]	no
273.It kern.defcorename	string	yes
274.It kern.domainname	string	yes
275.It kern.drivers	struct kinfo_drivers	no
276.It kern.file	struct file	no
277.It kern.forkfsleep	integer	yes
278.It kern.fscale	integer	no
279.It kern.fsync	integer	no
280.It kern.hardclock_ticks	integer	no
281.It kern.hostid	integer	yes
282.It kern.hostname	string	yes
283.It kern.iov_max	integer	no
284.It kern.job_control	integer	no
285.It kern.labeloffset	integer	no
286.It kern.labelsector	integer	no
287.It kern.login_name_max	integer	no
288.It kern.logsigexit	integer	yes
289.It kern.mapped_files	integer	no
290.It kern.maxfiles	integer	yes
291.It kern.maxpartitions	integer	no
292.It kern.maxphys	integer	no
293.It kern.maxproc	integer	yes
294.It kern.maxptys	integer	yes
295.It kern.maxvnodes	integer	yes
296.It kern.mbuf	node	not applicable
297.It kern.memlock	integer	no
298.It kern.memlock_range	integer	no
299.It kern.memory_protection	integer	no
300.It kern.monotonic_clock	integer	no
301.It kern.msgbuf	integer	no
302.It kern.msgbufsize	integer	no
303.It kern.ngroups	integer	no
304.It kern.ntptime	struct ntptimeval	no
305.It kern.osrelease	string	no
306.It kern.osrev	integer	no
307.It kern.ostype	string	no
308.It kern.pipe	node	not applicable
309.It kern.posix1	integer	no
310.It kern.posix_barriers	integer	no
311.It kern.posix_reader_writer_locks	integer	no
312.It kern.posix_semaphores	integer	no
313.It kern.posix_spin_locks	integer	no
314.It kern.posix_threads	integer	no
315.It kern.posix_timers	integer	no
316.It kern.proc	struct kinfo_proc	no
317.It kern.proc2	struct kinfo_proc2	no
318.It kern.proc_args	string	no
319.It kern.prof	node	not applicable
320.It kern.rawpartition	integer	no
321.It kern.root_device	string	no
322.It kern.root_partition	integer	no
323.It kern.rtc_offset	integer	yes
324.It kern.saved_ids	integer	no
325.It kern.securelevel	integer	raise only
326.It kern.synchronized_io	integer	no
327.It kern.ipc	node	not applicable
328.It kern.timecounter	node	not applicable
329.It kern.timex	struct	no
330.It kern.tkstat	node	not applicable
331.It kern.urandom	integer	no
332.It kern.version	string	no
333.It kern.vnode	struct vnode	no
334.El
335.Bl -tag -width "123456"
336.It Li kern.argmax ( KERN_ARGMAX )
337The maximum bytes of argument to
338.Xr execve 2 .
339.It Li kern.autonicetime ( KERN_AUTONICETIME )
340The number of seconds of CPU-time a non-root process may accumulate before
341having its priority lowered from the default to the value of KERN_AUTONICEVAL.
342If set to 0, automatic lowering of priority is not performed, and if set to \-1
343all non-root processes are immediately lowered.
344.It Li kern.autoniceval ( KERN_AUTONICEVAL )
345The priority assigned for automatically niced processes.
346.It Li kern.boothowto
347Flags passed from the boot loader; see
348.Xr reboot 2
349for the meanings of the flags.
350.It Li kern.boottime ( KERN_BOOTTIME )
351A
352.Va struct timeval
353structure is returned.
354This structure contains the time that the system was booted.
355.It Li kern.ccpu ( KERN_CCPU )
356The scheduler exponential decay value.
357.It Li kern.clockrate ( KERN_CLOCKRATE )
358A
359.Va struct clockinfo
360structure is returned.
361This structure contains the clock, statistics clock and profiling clock
362frequencies, the number of micro-seconds per hz tick, and the clock
363skew rate.
364.It Li kern.consdev ( KERN_CONSDEV )
365Console device.
366.It Li kern.cp_id ( KERN_CP_ID )
367Mapping of CPU number to CPU id.
368.It Li kern.cp_time ( KERN_CP_TIME )
369Returns an array of CPUSTATES uint64_ts.
370This array contains the
371number of clock ticks spent in different CPU states.
372On multi-processor systems, the sum across all CPUs is returned unless
373appropriate space is given for one data set for each CPU.
374Data for a specific CPU can also be obtained by adding the number of the
375CPU at the end of the MIB, enlarging it by one.
376.It Li kern.defcorename ( KERN_DEFCORENAME )
377Default template for the name of core dump files (see also
378.Li proc.pid.corename
379in the per-process variables
380.Li proc.* ,
381and
382.Xr core 5
383for format of this template).
384The default value is
385.Nm %n.core
386and can be changed with the kernel configuration option
387.Cd options DEFCORENAME
388(see
389.Xr options 4
390).
391.It Li kern.domainname ( KERN_DOMAINNAME )
392Get or set the YP domain name.
393.It Li kern.dump_on_panic ( KERN_DUMP_ON_PANIC )
394Perform a crash dump on system panic.
395.It Li kern.drivers ( KERN_DRIVERS )
396Return an array of
397.Va struct kinfo_drivers
398that contains the name and major device numbers of all the device drivers
399in the current kernel.
400The
401.Va d_name
402field is always a NUL terminated string.
403The
404.Va d_bmajor
405field will be set to \-1 if the driver doesn't have a block device.
406.It Li kern.file ( KERN_FILE )
407Return the entire file table.
408The returned data consists of a single
409.Va struct filelist
410followed by an array of
411.Va struct file ,
412whose size depends on the current number of such objects in the system.
413.It Li kern.forkfsleep ( KERN_FORKFSLEEP )
414If
415.Xr fork 2
416system call fails due to limit on number of processes (either
417the global maxproc limit or user's one), wait for this many
418milliseconds before returning
419.Er EAGAIN
420error to process.
421Useful to keep heavily forking runaway processes in bay.
422Default zero (no sleep).
423Maximum is 20 seconds.
424.It Li kern.fscale ( KERN_FSCALE )
425The kernel fixed-point scale factor.
426.It Li kern.fsync ( KERN_FSYNC )
427Return 1 if the POSIX 1003.1b File Synchronization Option is available
428on this system,
429otherwise 0.
430.It Li kern.hardclock_ticks ( KERN_HARDCLOCK_TICKS )
431Returns the number of
432.Xr hardclock 9
433ticks.
434.It Li kern.hostid ( KERN_HOSTID )
435Get or set the host id.
436.It Li kern.hostname ( KERN_HOSTNAME )
437Get or set the hostname.
438.It Li kern.iov_max ( KERN_IOV_MAX )
439Return the maximum number of
440.Va iovec
441structures that a process has available for use with
442.Xr preadv 2 ,
443.Xr pwritev 2 ,
444.Xr readv 2 ,
445.Xr recvmsg 2 ,
446.Xr sendmsg 2
447and
448.Xr writev 2 .
449.It Li kern.job_control ( KERN_JOB_CONTROL )
450Return 1 if job control is available on this system, otherwise 0.
451.It Li kern.labeloffset ( KERN_LABELOFFSET )
452The offset within the sector specified by KERN_LABELSECTOR of the
453.Xr disklabel 5 .
454.It Li kern.labelsector ( KERN_LABELSECTOR )
455The sector number containing the
456.Xr disklabel 5 .
457.It Li kern.login_name_max ( KERN_LOGIN_NAME_MAX )
458The size of the storage required for a login name, in bytes,
459including the terminating NUL.
460.It Li kern.logsigexit ( KERN_LOGSIGEXIT )
461If this flag is non-zero, the kernel will
462.Xr log 9
463all process exits due to signals which create a
464.Xr core 5
465file, and whether the coredump was created.
466.It Li kern.mapped_files ( KERN_MAPPED_FILES )
467Returns 1 if the POSIX 1003.1b Memory Mapped Files Option is available
468on this system,
469otherwise 0.
470.It Li kern.maxfiles ( KERN_MAXFILES )
471The maximum number of open files that may be open in the system.
472.It Li kern.maxpartitions ( KERN_MAXPARTITIONS )
473The maximum number of partitions allowed per disk.
474.It Li kern.maxphys ( KERN_MAXPHYS )
475Maximum raw I/O transfer size.
476.It Li kern.maxproc ( KERN_MAXPROC )
477The maximum number of simultaneous processes the system will allow.
478.It Li kern.maxptys ( KERN_MAXPTYS )
479The maximum number of pseudo terminals.
480This value can be both raised and lowered, though it cannot
481be set lower than number of currently used ptys.
482See also
483.Xr pty 4 .
484.It Li kern.maxvnodes ( KERN_MAXVNODES )
485The maximum number of vnodes available on the system.
486This can only be raised.
487.It Li kern.mbuf ( KERN_MBUF )
488Return information about the mbuf control variables.
489Mbufs are data structures which store network packets and other data
490structures in the networking code, see
491.Xr mbuf 9 .
492The third level names for the mbuf variables are detailed below.
493The changeable column shows whether a process with appropriate
494privilege may change the value.
495.Bl -column "kern.mbuf.nmbclusters" "integer" "Changeable" -offset indent
496.It Sy Third level name	Type	Changeable
497.\" XXX Changeable? really?
498.It kern.mbuf.mblowat	integer	yes
499.It kern.mbuf.mclbytes	integer	yes
500.It kern.mbuf.mcllowat	integer	yes
501.It kern.mbuf.msize	integer	yes
502.It kern.mbuf.nmbclusters	integer	yes
503.El
504.Pp
505The variables are as follows:
506.Bl -tag -width "123456"
507.It Li kern.mbuf.mblowat ( MBUF_MBLOWAT )
508The mbuf low water mark.
509.It Li kern.mbuf.mclbytes ( MBUF_MCLBYTES )
510The mbuf cluster size.
511.It Li kern.mbuf.mcllowat ( MBUF_MCLLOWAT )
512The mbuf cluster low water mark.
513.It Li kern.mbuf.msize ( MBUF_MSIZE )
514The mbuf base size.
515.It Li kern.mbuf.nmbclusters ( MBUF_NMBCLUSTERS )
516The limit on the number of mbuf clusters.
517The variable can only be increased, and only increased on machines with
518direct-mapped pool pages.
519.El
520.It Li kern.memlock ( KERN_MEMLOCK )
521Returns 1 if the POSIX 1003.1b Process Memory Locking Option is available
522on this system,
523otherwise 0.
524.It Li kern.memlock_range ( KERN_MEMLOCK_RANGE )
525Returns 1 if the POSIX 1003.1b Range Memory Locking Option is available
526on this system,
527otherwise 0.
528.It Li kern.memory_protection ( KERN_MEMORY_PROTECTION )
529Returns 1 if the POSIX 1003.1b Memory Protection Option is available
530on this system,
531otherwise 0.
532.It Li kern.monotonic_clock ( KERN_MONOTONIC_CLOCK )
533Returns the standard version the implementation of the POSIX 1003.1b
534Monotonic Clock Option conforms to,
535otherwise 0.
536.It Li kern.msgbuf ( KERN_MSGBUF )
537The kernel message buffer, rotated so that the head of the circular kernel
538message buffer is at the start of the returned data.
539The returned data may contain NUL bytes.
540.It Li kern.msgbufsize ( KERN_MSGBUFSIZE )
541The maximum number of characters that the kernel message buffer can hold.
542.It Li kern.ngroups ( KERN_NGROUPS )
543The maximum number of supplemental groups.
544.It Li kern.ntptime ( KERN_NTPTIME )
545A
546.Va struct ntptimeval
547structure is returned.
548This structure contains data used by the
549.Xr ntpd 8
550program.
551.It Li kern.osrelease ( KERN_OSRELEASE )
552The system release string.
553.It Li kern.osrevision ( KERN_OSREV )
554The system revision string.
555.It Li kern.ostype ( KERN_OSTYPE )
556The system type string.
557.It Li kern.pipe ( KERN_PIPE )
558Pipe settings.
559The third level names for the  integer pipe settings is detailed below.
560The changeable column shows whether a process with appropriate
561privilege may change the value.
562.Bl -column "kern.pipe.maxbigpipes" "integer" "Changeable" -offset indent
563.It Sy Third level name	Type	Changeable
564.It kern.pipe.kvasiz	integer	yes
565.It kern.pipe.maxbigpipes	integer	yes
566.It kern.pipe.maxkvasz	integer	yes
567.It kern.pipe.limitkva	integer	yes
568.It kern.pipe.nbigpipes	integer	yes
569.El
570.Pp
571The variables are as follows:
572.Bl -tag -width "123456"
573.It Li kern.pipe.kvasiz ( KERN_PIPE_KVASIZ )
574Amount of kernel memory consumed by pipe buffers.
575.It Li kern.pipe.maxbigpipes ( KERN_PIPE_MAXBIGPIPES )
576Maximum number of "big" pipes.
577.It Li kern.pipe.maxkvasz ( KERN_PIPE_MAXKVASZ )
578Maximum amount of kernel memory to be used for pipes.
579.It Li kern.pipe.limitkva ( KERN_PIPE_LIMITKVA )
580Limit for direct transfers via page loan.
581.It Li kern.pipe.nbigpipes ( KERN_PIPE_NBIGPIPES )
582Number of "big" pipes.
583.El
584.It Li kern.posix1version ( KERN_POSIX1 )
585The version of ISO/IEC 9945 (POSIX 1003.1) with which the system
586attempts to comply.
587.It Li kern.posix_barriers ( KERN_POSIX_BARRIERS )
588The version of
589.St -p1003.1
590and its
591Barriers
592option to which the system attempts to conform,
593otherwise 0.
594.It Li kern.posix_reader_writer_locks ( KERN_POSIX_READER_WRITER_LOCKS )
595The version of
596.St -p1003.1
597and its
598Read-Write Locks
599option to which the system attempts to conform,
600otherwise 0.
601.It Li kern.posix_semaphores ( KERN_POSIX_SEMAPHORES )
602The version of
603.St -p1003.1
604and its
605Semaphores
606option to which the system attempts to conform,
607otherwise 0.
608.It Li kern.posix_spin_locks ( KERN_POSIX_SPIN_LOCKS )
609The version of
610.St -p1003.1
611and its
612Spin Locks
613option to which the system attempts to conform,
614otherwise 0.
615.It Li kern.posix_threads ( KERN_POSIX_THREADS )
616The version of
617.St -p1003.1
618and its
619Threads
620option to which the system attempts to conform,
621otherwise 0.
622.It Li kern.posix_timers ( KERN_POSIX_TIMERS )
623The version of
624.St -p1003.1
625and its
626Timers
627option to which the system attempts to conform,
628otherwise 0.
629.It Li kern.proc ( KERN_PROC )
630Return the entire process table, or a subset of it.
631An array of
632.Va struct kinfo_proc
633structures is returned,
634whose size depends on the current number of such objects in the system.
635The third and fourth level numeric names are as follows:
636.Bl -column "KERN_PROC_SESSION" "Fourth level is:" -offset indent
637.It Sy Third level name	Fourth level is:
638.It KERN_PROC_ALL	None
639.It KERN_PROC_GID	A group ID
640.It KERN_PROC_PID	A process ID
641.It KERN_PROC_PGRP	A process group
642.It KERN_PROC_RGID	A real group ID
643.It KERN_PROC_RUID	A real user ID
644.It KERN_PROC_SESSION	A session ID
645.It KERN_PROC_TTY	A tty device
646.It KERN_PROC_UID	A user ID
647.El
648.It Li kern.proc2 ( KERN_PROC2 )
649As for KERN_PROC, but an array of
650.Va struct kinfo_proc2
651structures are returned.
652The fifth level name is the size of the
653.Va struct kinfo_proc2
654and the sixth level name is the number of structures to return.
655.It Li kern.proc_args ( KERN_PROC_ARGS )
656Return the argv or environment strings (or the number thereof)
657of a process.
658Multiple strings are returned separated by NUL characters.
659The third level name is the process ID.
660The fourth level name is as follows:
661.Bl -column "KERN_PROG_NARGV" "The number of environ strings" -offset indent
662.It KERN_PROC_ARGV	The argv strings
663.It KERN_PROC_ENV	The environ strings
664.It KERN_PROC_NARGV	The number of argv strings
665.It KERN_PROC_NENV	The number of environ strings
666.El
667.It Li kern.profiling ( KERN_PROF )
668Return profiling information about the kernel.
669If the kernel is not compiled for profiling,
670attempts to retrieve any of the KERN_PROF values will
671fail with
672.Er EOPNOTSUPP .
673The third level names for the string and integer profiling information
674is detailed below.
675The changeable column shows whether a process with appropriate
676privilege may change the value.
677.Bl -column "kern.profiling.gmonparam" "struct gmonparam" "Changeable" -offset indent
678.It Sy Third level name	Type	Changeable
679.It kern.profiling.count	u_short[\|]	yes
680.It kern.profiling.froms	u_short[\|]	yes
681.It kern.profiling.gmonparam	struct gmonparam	no
682.It kern.profiling.state	integer	yes
683.It kern.profiling.tos	struct tostruct	yes
684.El
685.Pp
686The variables are as follows:
687.Bl -tag -width "123456"
688.It Li kern.profiling.count ( GPROF_COUNT )
689Array of statistical program counter counts.
690.It Li kern.profiling.froms ( GPROF_FROMS )
691Array indexed by program counter of call-from points.
692.It Li kern.profiling.gmonparams ( GPROF_GMONPARAM )
693Structure giving the sizes of the above arrays.
694.It Li kern.profiling.state ( GPROF_STATE )
695Profiling state.
696If set to GMON_PROF_ON, starts profiling.
697If set to GMON_PROF_OFF, stops profiling.
698.It Li kern.profiling.tos ( GPROF_TOS )
699Array of
700.Va struct tostruct
701describing destination of calls and their counts.
702.El
703.It Li kern.rawpartition ( KERN_RAWPARTITION )
704The raw partition of a disk (a == 0).
705.It Li kern.root_device ( KERN_ROOT_DEVICE )
706The name of the root device (e.g.,
707.Dq wd0 ) .
708.It Li kern.root_partition ( KERN_ROOT_PARTITION )
709The root partition on the root device (a == 0).
710.It Li kern.rtc_offset ( KERN_RTC_OFFSET )
711Return the offset of real time clock from UTC in minutes.
712.It Li kern.saved_ids ( KERN_SAVED_IDS )
713Returns 1 if saved set-group and saved set-user ID is available.
714.It Li kern.sbmax ( KERN_SBMAX )
715Maximum socket buffer size.
716.\" XXX units?
717.It Li kern.securelevel ( KERN_SECURELVL )
718See
719.Xr secmodel_securelevel 9 .
720.It Li kern.somaxkva ( KERN_SOMAXKVA )
721Maximum amount of kernel memory to be used for socket buffers.
722.\" XXX units?
723.It Li kern.synchronized_io ( KERN_SYNCHRONIZED_IO )
724Returns 1 if the POSIX 1003.1b Synchronized I/O Option is available
725on this system,
726otherwise 0.
727.It Li kern.ipc ( KERN_SYSVIPC )
728Return information about the SysV IPC parameters.
729The third level names for the ipc variables are detailed below.
730.Bl -column "kern.ipc.shm_use_phys" "integer" "Changeable" -offset indent
731.It Sy Third level name	Type	Changeable
732.It kern.ipc.sysvmsg	integer	no
733.It kern.ipc.sysvsem	integer	no
734.It kern.ipc.sysvshm	integer	no
735.It kern.ipc.sysvipc_info	struct	no
736.It kern.ipc.shmmax	integer	yes
737.It kern.ipc.shmmni	integer	yes
738.It kern.ipc.shmseg	integer	yes
739.It kern.ipc.shmmaxpgs	integer	yes
740.It kern.ipc.shm_use_phys	integer	yes
741.It kern.ipc.msgmni	integer	yes
742.It kern.ipc.msgseg	integer	yes
743.It kern.ipc.semmni	integer	yes
744.It kern.ipc.semmns	integer	yes
745.It kern.ipc.semmnu	integer	yes
746.El
747.Bl -tag -width "123456"
748.It Li kern.ipc.sysvmsg ( KERN_SYSVIPC_MSG )
749Returns 1 if System V style message queue functionality is available
750on this system,
751otherwise 0.
752.It Li kern.ipc.sysvsem ( KERN_SYSVIPC_SEM )
753Returns 1 if System V style semaphore functionality is available
754on this system,
755otherwise 0.
756.It Li kern.ipc.sysvshm ( KERN_SYSVIPC_SHM )
757Returns 1 if System V style share memory functionality is available
758on this system,
759otherwise 0.
760.It Li kern.ipc.sysvipc_info ( KERN_SYSVIPC_INFO )
761Return System V style IPC configuration and run-time information.
762The fourth level name selects the System V style IPC facility.
763.Bl -column "KERN_SYSVIPC_MSG_INFO" "struct shm_sysctl_info" -offset indent
764.It Sy Fourth level name	Type
765.It KERN_SYSVIPC_MSG_INFO	struct msg_sysctl_info
766.It KERN_SYSVIPC_SEM_INFO	struct sem_sysctl_info
767.It KERN_SYSVIPC_SHM_INFO	struct shm_sysctl_info
768.El
769.Pp
770.Bl -tag -width "123456"
771.It Li KERN_SYSVIPC_MSG_INFO
772Return information on the System V style message facility.
773The
774.Sy msg_sysctl_info
775structure is defined in
776.Aq Pa sys/msg.h .
777.It Li KERN_SYSVIPC_SEM_INFO
778Return information on the System V style semaphore facility.
779The
780.Sy sem_sysctl_info
781structure is defined in
782.Aq Pa sys/sem.h .
783.It Li KERN_SYSVIPC_SHM_INFO
784Return information on the System V style shared memory facility.
785The
786.Sy shm_sysctl_info
787structure is defined in
788.Aq Pa sys/shm.h .
789.El
790.It Li kern.ipc.shmmax ( KERN_SYSVIPC_SHMMAX )
791Max shared memory segment size in bytes.
792.It Li kern.ipc.shmmni ( KERN_SYSVIPC_SHMMNI )
793Max number of shared memory identifiers.
794.It Li kern.ipc.shmseg ( KERN_SYSVIPC_SHMSEG )
795Max shared memory segments per process.
796.It Li kern.ipc.shmmaxpgs ( KERN_SYSVIPC_SHMMAXPGS )
797Max amount of shared memory in pages.
798.It Li kern.ipc.shm_use_phys ( KERN_SYSVIPC_SHMUSEPHYS )
799Locking of shared memory in physical memory.
800If 0, memory can be swapped
801out, otherwise it will be locked in physical memory.
802.It Li kern.ipc.msgmni
803Max number of message queue identifiers.
804.It Li kern.ipc.msgseg
805Max number of number of message segments.
806.It Li kern.ipc.semmni
807Max number of number of semaphore identifiers.
808.It Li kern.ipc.semmns
809Max number of number of semaphores in system.
810.It Li kern.ipc.semmnu
811Max number of undo structures in system.
812.El
813.It Li kern.timecounter ( dynamic )
814Display and control the timecounter source of the system.
815.Bl -column "kern.timecounter.timestepwarnings" "integer" "Changeable" -offset indent
816.It Sy Third level name	Type	Changeable
817.It kern.timecounter.choice	string	no
818.It kern.timecounter.hardware	string	yes
819.It kern.timecounter.timestepwarnings	integer	yes
820.El
821.Pp
822The variables are as follows:
823.Bl -tag -width "123456"
824.It Li kern.timecounter.choice ( dynamic )
825The list of available timecounters with their quality and frequency.
826.It Li kern.timecounter.hardware ( dynamic )
827The currently selected timecounter source.
828.It Li kern.timecounter.timestepwarnings ( dynamic )
829If non-zero display a message each time the time is stepped.
830.El
831.It Li kern.timex ( KERN_TIMEX )
832Not available.
833.It Li kern.tkstat ( KERN_TKSTAT )
834Return information about the number of characters sent and received
835on ttys.
836The third level names for the tty statistic variables are detailed below.
837The changeable column shows whether a process
838with appropriate privilege may change the value.
839.Bl -column "kern.tkstat.cancc" "quad" "Changeable" -offset indent
840.It Sy Third level name	Type	Changeable
841.It kern.tkstat.cancc	quad	no
842.It kern.tkstat.nin	quad	no
843.It kern.tkstat.nout	quad	no
844.It kern.tkstat.rawcc	quad	no
845.El
846.Pp
847The variables are as follows:
848.Bl -tag -width "123456"
849.It Li kern.tkstat.cancc ( KERN_TKSTAT_CANCC )
850The number of canonical input characters.
851.It Li kern.tkstat.nin ( KERN_TKSTAT_NIN )
852The total number of input characters.
853.It Li kern.tkstat.nout ( KERN_TKSTAT_NOUT )
854The total number of output characters.
855.It Li kern.tkstat.rawcc ( KERN_TKSTAT_RAWCC )
856The number of raw input characters.
857.El
858.It Li kern.urandom ( KERN_URND )
859Random integer value.
860.It Li kern.veriexec
861Tunings for Verixec.
862.Bl -tag -width "123456"
863.It Li kern.veriexec.algorithms
864Returns a string with the supported algorithms in Veriexec.
865.It Li kern.veriexec.count
866Sub-nodes are added to this node as new mounts are monitored by Veriexec.
867Each mount will be under its own
868.No tableN
869node.
870Under each node there will be three variables, indicating the mount
871point, the file-system type, and the number of entries.
872.It Li kern.veriexec.strict
873Controls the strict level of Veriexec.
874See
875.Xr security 8
876for more information on each level's implications.
877.It Li kern.veriexec.verbose
878Controls the verbosity level of Veriexec.
879If 0, only the minimal
880indication required will be given about what's happening - fingerprint
881mismatches, removal of entries from the tables, modification of a
882fingerprinted file.
883If 1, more messages will be printed (ie., when a file with a valid
884fingerprint is accessed).
885Verbose level 2 is debug mode.
886.El
887.It Li kern.version ( KERN_VERSION )
888The system version string.
889.It Li kern.vnode ( KERN_VNODE )
890Return the entire vnode table.
891Note, the vnode table is not necessarily a consistent snapshot of
892the system.
893The returned data consists of an array whose size depends on the
894current number of such objects in the system.
895Each element of the array contains the kernel address of a vnode
896.Va struct vnode *
897followed by the vnode itself
898.Va struct vnode .
899.It Li kern.coredump.setid
900Settings related to set-id processes coredumps.
901By default, set-id processes do not dump core in situations where
902other processes would.
903The settings in this node allows an administrator to change this
904behavior.
905.Pp
906.Bl -tag -width "123456"
907.It Li kern.coredump.setid.dump
908If non-zero, set-id processes will dump core.
909.It Li kern.coredump.setid.group
910The group-id for the set-id processes' coredump.
911.It Li kern.coredump.setid.mode
912The mode for the set-id processes' coredump.
913See
914.Xr chmod 1 .
915.It Li kern.coredump.setid.owner
916The user-id that will be used as the owner of the set-id processes'
917coredump.
918.It Li kern.coredump.setid.path
919The path to which set-id processes' coredumps will be saved to.
920Same syntax as kern.defcorename.
921.El
922.\" XXX kern.lwp
923.El
924.Sh The machdep.* subtree
925The set of variables defined is architecture dependent.
926Most architectures define at least the following variables.
927.Bl -column "Second level name" "Type" "Changeable" -offset indent
928.It Sy Second level name	Type	Changeable
929.It Li CPU_CONSDEV	dev_t	no
930.El
931.Sh The net.* subtree
932The string and integer information available for the
933.Li net
934level is detailed below.
935The changeable column shows whether a process with appropriate
936privilege may change the value.
937The second and third levels are typically the protocol family and
938protocol number, though this is not always the case.
939.Bl -column "Second level name" "IPsec key management values" "Changeable" -offset indent
940.It Sy Second level name	Type	Changeable
941.It net.route	routing messages	no
942.It net.inet	IPv4 values	yes
943.It net.inet6	IPv6 values	yes
944.It net.key	IPsec key management values	yes
945.El
946.Pp
947.Bl -tag -width "123456"
948.It Li net.route ( PF_ROUTE )
949.\" XXX really?
950Return the entire routing table or a subset of it.
951The data is returned as a sequence of routing messages (see
952.Xr route 4
953for the header file, format and meaning).
954The length of each message is contained in the message header.
955.Pp
956The third level name is a protocol number, which is currently always 0.
957The fourth level name is an address family, which may be set to 0 to
958select all address families.
959The fifth and sixth level names are as follows:
960.Bl -column "Fifth level name" "Sixth level is:" -offset indent
961.It Sy Fifth level name	Sixth level is:
962.It NET_RT_FLAGS	rtflags
963.It NET_RT_DUMP	None
964.It NET_RT_IFLIST	None
965.El
966.It Li net.inet ( PF_INET )
967Get or set various global information about the IPv4
968.Pq Internet Protocol version 4 .
969The third level name is the protocol.
970The fourth level name is the variable name.
971The currently defined protocols and names are:
972.Bl -column "Protocol name" "sack.globalmaxholes" "integer" "Changeable" -offset 4n
973.It Sy Protocol name	Variable name	Type	Changeable
974.It arp	down	integer	yes
975.It arp	keep	integer	yes
976.It arp	prune	integer	yes
977.It arp	refresh	integer	yes
978.It carp	allow	integer	yes
979.It carp	preempt	integer	yes
980.It carp	log	integer	yes
981.It carp	arpbalance	integer	yes
982.It icmp	errppslimit	integer	yes
983.It icmp	maskrepl	integer	yes
984.It icmp	rediraccept	integer	yes
985.It icmp	redirtimeout	integer	yes
986.It ip	allowsrcrt	integer	yes
987.It ip	anonportmax	integer	yes
988.It ip	anonportmin	integer	yes
989.It ip	checkinterface	integer	yes
990.It ip	directed-broadcast	integer	yes
991.It ip	do_loopback_cksum	integer	yes
992.It ip	forwarding	integer	yes
993.It ip	forwsrcrt	integer	yes
994.It ip	gifttl	integer	yes
995.It ip	grettl	integer	yes
996.It ip	hashsize	integer	yes
997.It ip	hostzerobroadcast	integer	yes
998.It ip	lowportmin	integer	yes
999.It ip	lowportmax	integer	yes
1000.It ip	maxflows	integer	yes
1001.It ip	maxfragpackets	integer	yes
1002.It ip	mtudisc	integer	yes
1003.It ip	mtudisctimeout	integer	yes
1004.It ip	random_id	integer	yes
1005.It ip	redirect	integer	yes
1006.It ip	subnetsarelocal	integer	yes
1007.It ip	ttl	integer	yes
1008.It tcp	rfc1323	integer	yes
1009.It tcp	sendspace	integer	yes
1010.It tcp	recvspace	integer	yes
1011.It tcp	mssdflt	integer	yes
1012.It tcp	syn_cache_limit	integer	yes
1013.It tcp	syn_bucket_limit	integer	yes
1014.It tcp	syn_cache_interval	integer	yes
1015.It tcp	init_win	integer	yes
1016.It tcp	init_win_local	integer	yes
1017.It tcp	mss_ifmtu	integer	yes
1018.It tcp	win_scale	integer	yes
1019.It tcp	timestamps	integer	yes
1020.It tcp	compat_42	integer	yes
1021.It tcp	cwm	integer	yes
1022.It tcp	cwm_burstsize	integer	yes
1023.It tcp	ack_on_push	integer	yes
1024.It tcp	keepidle	integer	yes
1025.It tcp	keepintvl	integer	yes
1026.It tcp	keepcnt	integer	yes
1027.It tcp	slowhz	integer	no
1028.It tcp	keepinit	integer	yes
1029.It tcp	log_refused	integer	yes
1030.It tcp	rstppslimit	integer	yes
1031.It tcp	ident	struct	no
1032.It tcp	drop	struct	no
1033.It tcp	sack.enable	integer	yes
1034.It tcp	sack.globalholes	integer	no
1035.It tcp	sack.globalmaxholes	integer	yes
1036.It tcp	sack.maxholes	integer	yes
1037.It tcp	ecn.enable	integer	yes
1038.It tcp	ecn.maxretries	integer	yes
1039.It tcp	congctl.selected	string	yes
1040.It tcp	congctl.available	string	yes
1041.It tcp	abc.enable	integer	yes
1042.It tcp	abc.aggressive	integer	yes
1043.It udp	checksum	integer	yes
1044.It udp	do_loopback_cksum	integer	yes
1045.It udp	recvspace	integer	yes
1046.It udp	sendspace	integer	yes
1047.El
1048.Pp
1049The variables are as follows:
1050.Bl -tag -width "123456"
1051.It Li arp.down
1052Failed ARP entry lifetime.
1053.It Li arp.keep
1054Valid ARP entry lifetime.
1055.It Li arp.prune
1056ARP cache pruning interval.
1057.It Li arp.refresh
1058ARP entry refresh interval.
1059.It Li carp.allow
1060If set to 0, incoming
1061.Xr carp 4
1062packets will not be processed.
1063If set to any other value, processing will occur.
1064Enabled by default.
1065.It Li carp.arpbalance
1066If set to any value other than 0, the ARP balancing functionality of
1067.Xr carp 4
1068is enabled.
1069When ARP requests are received for an IP address which is part of any virtual
1070host, carp will hash the source IP in the ARP request to select one of the
1071virtual hosts from the set of all the virtual hosts which have that IP address.
1072The master of that host will respond with the correct virtual MAC address.
1073Disabled by default.
1074.It Li carp.log
1075If set to any value other than 0,
1076.Xr carp 4
1077will log errors.
1078Disabled by default.
1079.It Li carp.preempt
1080If set to 0,
1081.Xr carp 4
1082will not attempt to become master if it is receiving advertisements from
1083another active master.
1084If set to any other value, carp will become master of the virtual host if it
1085believes it can send advertisements more frequently than the current master.
1086Disabled by default.
1087.It Li ip.allowsrcrt
1088If set to 1, the host accepts source routed packets.
1089.It Li ip.anonportmax
1090The highest port number to use for TCP and UDP ephemeral port allocation.
1091This cannot be set to less than 1024 or greater than 65535, and must
1092be greater than
1093.Li ip.anonportmin .
1094.It Li ip.anonportmin
1095The lowest port number to use for TCP and UDP ephemeral port allocation.
1096This cannot be set to less than 1024 or greater than 65535.
1097.It Li ip.checkinterface
1098If set to non-zero, the host will reject packets addressed to it
1099that arrive on an interface not bound to that address.
1100Currently, this must be disabled if ipnat is used to translate the
1101destination address to another local interface, or if addresses
1102are added to the loopback interface instead of the interface where
1103the packets for those packets are received.
1104.It Li ip.directed-broadcast
1105If set to 1, enables directed broadcast behavior for the host.
1106.It Li ip.do_loopback_cksum
1107Perform IP checksum on loopback.
1108.It Li ip.forwarding
1109If set to 1, enables IP forwarding for the host,
1110meaning that the host is acting as a router.
1111.It Li ip.forwsrcrt
1112If set to 1, enables forwarding of source-routed packets for the host.
1113This value may only be changed if the kernel security level is less than 1.
1114.It Li ip.gifttl
1115The maximum time-to-live (hop count) value for an IPv4 packet generated by
1116.Xr gif 4
1117tunnel interface.
1118.It Li ip.grettl
1119The maximum time-to-live (hop count) value for an IPv4 packet generated by
1120.Xr gre 4
1121tunnel interface.
1122.It Li ip.hashsize
1123The size of IPv4 Fast Forward hash table.
1124This value must be a power of 2 (64, 256...).
1125A larger hash table size results in fewer collisions.
1126Also see
1127.Li ip.maxflows .
1128.It Li ip.hostzerobroadcast
1129All zeroes address is broadcast address.
1130.It Li ip.lowportmax
1131The highest port number to use for TCP and UDP reserved port allocation.
1132This cannot be set to less than 0 or greater than 1024, and must
1133be greater than
1134.Li ip.lowportmin .
1135.It Li ip.lowportmin
1136The lowest port number to use for TCP and UDP reserved port allocation.
1137This cannot be set to less than 0 or greater than 1024, and must
1138be smaller than
1139.Li ip.lowportmax .
1140.It Li ip.maxflows
1141IPv4 Fast Forwarding is enabled by default.
1142If set to 0, IPv4 Fast Forwarding is disabled.
1143.Li ip.maxflows
1144controls the maximum amount of flows which can be created.
1145The default value is 256.
1146.It Li ip.maxfragpackets
1147The maximum number of fragmented packets the node will accept.
11480 means that the node will not accept any fragmented packets.
1149\-1 means that the node will accept as many fragmented packets as it receives.
1150The flag is provided basically for avoiding possible DoS attacks.
1151.It Li ip.mtudisc
1152If set to 1, enables Path MTU Discovery (RFC 1191).
1153When Path MTU Discovery is enabled, the transmitted TCP segment
1154size will be determined by the advertised maximum segment size
1155(MSS) from the remote end, as constrained by the path MTU.
1156If MTU Discovery is disabled, the transmitted segment size will
1157never be greater than
1158.Li tcp.mssdflt
1159(the local maximum segment size).
1160.It Li ip.mtudisctimeout
1161The number of seconds in which a route added by the Path MTU
1162Discovery engine will time out.
1163When the route times out, the Path
1164MTU Discovery engine will attempt to probe a larger path MTU.
1165.It Li ip.random_id
1166Assign random ip_id values.
1167.It Li ip.redirect
1168If set to 1, ICMP redirects may be sent by the host.
1169This option is ignored unless the host is routing IP packets,
1170and should normally be enabled on all systems.
1171.It Li ip.subnetsarelocal
1172If set to 1, subnets are to be considered local addresses.
1173.It Li ip.ttl
1174The maximum time-to-live (hop count) value for an IP packet sourced by
1175the system.
1176This value applies to normal transport protocols, not to ICMP.
1177.It Li icmp.errppslimit
1178The variable specifies the maximum number of outgoing ICMP error messages,
1179per second.
1180ICMP error messages that exceeded the value are subject to rate limitation
1181and will not go out from the node.
1182Negative value disables rate limitation.
1183.It Li icmp.maskrepl
1184If set to 1, ICMP network mask requests are to be answered.
1185.It Li icmp.rediraccept
1186If set to non-zero, the host will accept ICMP redirect packets.
1187Note that routers will never accept ICMP redirect packets,
1188and the variable is meaningful on IP hosts only.
1189.It Li icmp.redirtimeout
1190The variable specifies lifetime of routing entries generated by incoming
1191ICMP redirect.
1192This defaults to 600 seconds.
1193.It Li icmp.returndatabytes
1194Number of bytes to return in an ICMP error message.
1195.It Li tcp.ack_on_push
1196If set to 1, TCP is to immediately transmit an ACK upon reception of
1197a packet with PUSH set.
1198This can avoid losing a round trip time in some rare situations,
1199but has the caveat of potentially defeating TCP's delayed ACK algorithm.
1200Use of this option is generally not recommended, but
1201the variable exists in case your configuration really needs it.
1202.It Li tcp.compat_42
1203If set to 1, enables work-arounds for bugs in the 4.2BSD TCP implementation.
1204Use of this option is not recommended, although it may be
1205required in order to communicate with extremely old TCP implementations.
1206.It Li tcp.cwm
1207If set to 1, enables use of the Hughes/Touch/Heidemann Congestion Window
1208Monitoring algorithm.
1209This algorithm prevents line-rate bursts of packets that could
1210otherwise occur when data begins flowing on an idle TCP connection.
1211These line-rate bursts can contribute to network and router congestion.
1212This can be particularly useful on World Wide Web servers
1213which support HTTP/1.1, which has lingering connections.
1214.It Li tcp.cwm_burstsize
1215The Congestion Window Monitoring allowed burst size, in terms
1216of packet count.
1217.It Li tcp.delack_ticks
1218Number of ticks to delay sending an ACK.
1219.It Li tcp.do_loopback_cksum
1220Perform TCP checksum on loopback.
1221.It Li tcp.init_win
1222A value indicating the TCP initial congestion window.
1223If this value is 0, an auto-tuning algorithm designed to use an initial
1224window of approximately 4K bytes is in use.
1225Otherwise, this value indicates a fixed number of packets.
1226.It Li tcp.init_win_local
1227Like
1228.Li tcp.init_win ,
1229but used when communicating with hosts on a local network.
1230.It Li tcp.keepcnt
1231Number of keepalive probes sent before declaring a connection dead.
1232If set to zero, there is no limit;
1233keepalives will be sent until some kind of
1234response is received from the peer.
1235.It Li tcp.keepidle
1236Time a connection must be idle before keepalives are sent (if keepalives
1237are enabled for the connection).
1238See also tcp.slowhz.
1239.It Li tcp.keepintvl
1240Time after a keepalive probe is sent until, in the absence of any response,
1241another probe is sent.
1242See also tcp.slowhz.
1243.It Li tcp.log_refused
1244If set to 1, refused TCP connections to the host will be logged.
1245.It Li tcp.keepinit
1246Timeout in seconds during connection establishment.
1247.It Li tcp.mss_ifmtu
1248If set to 1, TCP calculates the outgoing maximum segment size based on
1249the MTU of the appropriate interface.
1250If set to 0, it is calculated based on the greater of the MTU of the
1251interface, and the largest (non-loopback) interface MTU on the system.
1252.It Li tcp.mssdflt
1253The default maximum segment size both advertised to the peer
1254and to use when either the peer does not advertise a maximum segment size to
1255us during connection setup or Path MTU Discovery
1256.Li ( ip.mtudisc )
1257is disabled.
1258Do not change this value unless you really know what you are doing.
1259.It Li tcp.recvspace
1260The default TCP receive buffer size.
1261.It Li tcp.rfc1323
1262If set to 1, enables RFC 1323 extensions to TCP.
1263.It Li tcp.rstppslimit
1264The variable specifies the maximum number of outgoing TCP RST packets,
1265per second.
1266TCP RST packet that exceeded the value are subject to rate limitation
1267and will not go out from the node.
1268Negative value disables rate limitation.
1269.It Li tcp.ident
1270Return the user ID of a connected socket pair.
1271(RFC1413 Identification Protocol lookups.)
1272.It Li tcp.drop
1273Drop a TCP socket pair connection.
1274.It Li tcp.sack.enable
1275If set to 1, enables RFC 2018 Selective ACKnowledgement.
1276.It Li tcp.sack.globalholes
1277Global number of TCP SACK holes.
1278.It Li tcp.sack.globalmaxholes
1279Global maximum number of TCP SACK holes.
1280.It Li tcp.sack.maxholes
1281Maximum number of TCP SACK holes allowed per connection.
1282.It Li tcp.ecn.enable
1283If set to 1, enables RFC 3168 Explicit Congestion Notification.
1284.It Li tcp.ecn.maxretries
1285Number of times to retry sending the ECN-setup packet.
1286.It Li tcp.sendspace
1287The default TCP send buffer size.
1288.It Li tcp.slowhz
1289The units for tcp.keepidle and tcp.keepintvl; those variables are in ticks
1290of a clock that ticks tcp.slowhz times per second.
1291(That is, their values
1292must be divided by the tcp.slowhz value to get times in seconds.)
1293.It Li tcp.syn_bucket_limit
1294The maximum number of entries allowed per hash bucket in the TCP
1295compressed state engine.
1296.It Li tcp.syn_cache_limit
1297The maximum number of entries allowed in the TCP compressed state
1298engine.
1299.It Li tcp.timestamps
1300If rfc1323 is enabled, a value of 1 indicates RFC 1323 time stamp options,
1301used for measuring TCP round trip times, are enabled.
1302.It Li tcp.win_scale
1303If rfc1323 is enabled, a value of 1 indicates RFC 1323 window scale options,
1304for increasing the TCP window size, are enabled.
1305.It Li tcp.congctl.available
1306The available TCP congestion control algorithms.
1307.It Li tcp.congctl.selected
1308The currently selected TCP congestion control algorithm.
1309.It Li tcp.abc.enable
1310If set to 1, use RFC 3465 Appropriate Byte Counting (ABC).
1311If set to 0, use traditional Packet Counting.
1312.It Li tcp.abc.aggressive
1313Choose the L parameter found in RFC 3465.
1314L is the maximum cwnd increase for an ack during slow start.
1315If set to 1, use L=2*SMSS.
1316If set to 0, use L=1*SMSS.
1317It has no effect unless tcp.abc.enable is set to 1.
1318.It Li udp.checksum
1319If set to 1, UDP checksums are being computed.
1320Received non-zero UDP checksums are always checked.
1321Disabling UDP checksums is strongly discouraged.
1322.It Li udp.sendspace
1323The default UDP send buffer size.
1324.It Li udp.recvspace
1325The default UDP receive buffer size.
1326.El
1327.Pp
1328For variables net.*.ipsec, please refer to
1329.Xr ipsec 4 .
1330.It Li net.inet6 ( PF_INET6 )
1331Get or set various global information about the IPv6
1332.Pq Internet Protocol version 6 .
1333The third level name is the protocol.
1334The fourth level name is the variable name.
1335The currently defined protocols and names are:
1336.Bl -column "Protocol name" "do_loopback_cksum" "integer" "Changeable" -offset indent
1337.It Sy Protocol name	Variable name	Type	Changeable
1338.It icmp6	errppslimit	integer	yes
1339.It icmp6	mtudisc_hiwat	integer	yes
1340.It icmp6	mtudisc_lowat	integer	yes
1341.It icmp6	nd6_debug	integer	yes
1342.It icmp6	nd6_delay	integer	yes
1343.It icmp6	nd6_maxnudhint	integer	yes
1344.It icmp6	nd6_mmaxtries	integer	yes
1345.It icmp6	nd6_prune	integer	yes
1346.It icmp6	nd6_umaxtries	integer	yes
1347.It icmp6	nd6_useloopback	integer	yes
1348.It icmp6	nodeinfo	integer	yes
1349.It icmp6	rediraccept	integer	yes
1350.It icmp6	redirtimeout	integer	yes
1351.It ip6	accept_rtadv	integer	yes
1352.It ip6	anonportmax	integer	yes
1353.It ip6	anonportmin	integer	yes
1354.It ip6	auto_flowlabel	integer	yes
1355.It ip6	dad_count	integer	yes
1356.It ip6	defmcasthlim	integer	yes
1357.It ip6	forwarding	integer	yes
1358.It ip6	gifhlim	integer	yes
1359.It ip6	hashsize	integer	yes
1360.It ip6	hlim	integer	yes
1361.It ip6	hdrnestlimit	integer	yes
1362.It ip6	kame_version	string	no
1363.It ip6	keepfaith	integer	yes
1364.It ip6	log_interval	integer	yes
1365.It ip6	lowportmax	integer	yes
1366.It ip6	lowportmin	integer	yes
1367.It ip6	maxflows	integer	yes
1368.It ip6	maxfragpackets	integer	yes
1369.It ip6	maxfrags	integer	yes
1370.It ip6	redirect	integer	yes
1371.It ip6	rr_prune	integer	yes
1372.It ip6	use_deprecated	integer	yes
1373.It ip6	v6only	integer	yes
1374.It udp6	do_loopback_cksum	integer	yes
1375.It udp6	recvspace	integer	yes
1376.It udp6	sendspace	integer	yes
1377.El
1378.Pp
1379The variables are as follows:
1380.Bl -tag -width "123456"
1381.It Li ip6.accept_rtadv
1382If set to non-zero, the node will accept ICMPv6 router advertisement packets
1383and autoconfigures address prefixes and default routers.
1384The node must be a host
1385.Pq not a router
1386for the option to be meaningful.
1387.It Li ip6.anonportmax
1388The highest port number to use for TCP and UDP ephemeral port allocation.
1389This cannot be set to less than 1024 or greater than 65535, and must
1390be greater than
1391.Li ip6.anonportmin .
1392.It Li ip6.anonportmin
1393The lowest port number to use for TCP and UDP ephemeral port allocation.
1394This cannot be set to less than 1024 or greater than 65535.
1395.It Li ip6.auto_flowlabel
1396On connected transport protocol packets,
1397fill IPv6 flowlabel field to help intermediate routers to identify packet flows.
1398.It Li ip6.dad_count
1399The variable configures number of IPv6 DAD
1400.Pq duplicated address detection
1401probe packets.
1402The packets will be generated when IPv6 interface addresses are configured.
1403.It Li ip6.defmcasthlim
1404The default hop limit value for an IPv6 multicast packet sourced by the node.
1405This value applies to all the transport protocols on top of IPv6.
1406There are APIs to override the value, as documented in
1407.Xr ip6 4 .
1408.It Li ip6.forwarding
1409If set to 1, enables IPv6 forwarding for the node,
1410meaning that the node is acting as a router.
1411If set to 0, disables IPv6 forwarding for the node,
1412meaning that the node is acting as a host.
1413IPv6 specification defines node behavior for
1414.Dq router
1415case and
1416.Dq host
1417case quite differently, and changing this variable during operation
1418may cause serious trouble.
1419It is recommended to configure the variable at bootstrap time,
1420and bootstrap time only.
1421.It Li ip6.gifhlim
1422The maximum hop limit value for an IPv6 packet generated by
1423.Xr gif 4
1424tunnel interface.
1425.It Li ip6.hdrnestlimit
1426The number of IPv6 extension headers permitted on incoming IPv6 packets.
1427If set to 0, the node will accept as many extension headers as possible.
1428.It Li ip6.hashsize
1429The size of IPv6 Fast Forward hash table.
1430This value must be a power of 2 (64, 256...).
1431A larger hash table size results in fewer collisions.
1432Also see
1433.Li ip6.maxflows .
1434.It Li ip6.hlim
1435The default hop limit value for an IPv6 unicast packet sourced by the node.
1436This value applies to all the transport protocols on top of IPv6.
1437There are APIs to override the value, as documented in
1438.Xr ip6 4 .
1439.It Li ip6.kame_version
1440The string identifies the version of KAME IPv6 stack implemented in the kernel.
1441.It Li ip6.keepfaith
1442If set to non-zero, it enables
1443.Dq FAITH
1444TCP relay IPv6-to-IPv4 translator code in the kernel.
1445Refer
1446.Xr faith 4
1447and
1448.Xr faithd 8
1449for detail.
1450.It Li ip6.log_interval
1451The variable controls amount of logs generated by IPv6 packet
1452forwarding engine, by setting interval between log output
1453.Pq in seconds .
1454.It Li ip6.lowportmax
1455The highest port number to use for TCP and UDP reserved port allocation.
1456This cannot be set to less than 0 or greater than 1024, and must
1457be greater than
1458.Li ip6.lowportmin .
1459.It Li ip6.lowportmin
1460The lowest port number to use for TCP and UDP reserved port allocation.
1461This cannot be set to less than 0 or greater than 1024, and must
1462be smaller than
1463.Li ip6.lowportmax .
1464.It Li ip6.maxflows
1465IPv6 Fast Forwarding is enabled by default.
1466If set to 0, IPv6 Fast Forwarding is disabled.
1467.Li ip6.maxflows
1468controls the maximum amount of flows which can be created.
1469The default value is 256.
1470.It Li ip6.maxfragpackets
1471The maximum number of fragmented packets the node will accept.
14720 means that the node will not accept any fragmented packets.
1473\-1 means that the node will accept as many fragmented packets as it receives.
1474The flag is provided basically for avoiding possible DoS attacks.
1475.It Li ip6.maxfrags
1476The maximum number of fragments the node will accept.
14770 means that the node will not accept any fragments.
1478\-1 means that the node will accept as many fragments as it receives.
1479The flag is provided basically for avoiding possible DoS attacks.
1480.It Li ip6.redirect
1481If set to 1, ICMPv6 redirects may be sent by the node.
1482This option is ignored unless the node is routing IP packets,
1483and should normally be enabled on all systems.
1484.It Li ip6.rr_prune
1485The variable specifies interval between IPv6 router renumbering prefix
1486babysitting, in seconds.
1487.It Li ip6.use_deprecated
1488The variable controls use of deprecated address, specified in RFC 2462 5.5.4.
1489.It Li ip6.v6only
1490The variable specifies initial value for
1491.Dv IPV6_V6ONLY
1492socket option for
1493.Dv AF_INET6
1494socket.
1495Please refer to
1496.Xr ip6 4
1497for detail.
1498.It Li icmp6.errppslimit
1499The variable specifies the maximum number of outgoing ICMPv6 error messages,
1500per second.
1501ICMPv6 error messages that exceeded the value are subject to rate limitation
1502and will not go out from the node.
1503Negative value disables rate limitation.
1504.It Li icmp6.mtudisc_hiwat
1505.It Li icmp6.mtudisc_lowat
1506The variables define the maximum number of routing table entries,
1507created due to path MTU discovery
1508.Pq prevents denial-of-service attacks with ICMPv6 too big messages .
1509When IPv6 path MTU discovery happens, we keep path MTU information into
1510the routing table.
1511If the number of routing table entries exceed the value,
1512the kernel will not attempt to keep the path MTU information.
1513.Li icmp6.mtudisc_hiwat
1514is used when we have verified ICMPv6 too big messages.
1515.Li icmp6.mtudisc_lowat
1516is used when we have unverified ICMPv6 too big messages.
1517Verification is performed by using address/port pairs kept in connected pcbs.
1518Negative value disables the upper limit.
1519.It Li icmp6.nd6_debug
1520If set to non-zero, kernel IPv6 neighbor discovery code will generate
1521debugging messages.
1522The debug outputs are useful to diagnose IPv6 interoperability issues.
1523The flag must be set to 0 for normal operation.
1524.It Li icmp6.nd6_delay
1525The variable specifies
1526.Dv DELAY_FIRST_PROBE_TIME
1527timing constant in IPv6 neighbor discovery specification
1528.Pq RFC 2461 ,
1529in seconds.
1530.It Li icmp6.nd6_maxnudhint
1531IPv6 neighbor discovery permits upper layer protocols to supply reachability
1532hints, to avoid unnecessary neighbor discovery exchanges.
1533The variable defines the number of consecutive hints the neighbor discovery
1534layer will take.
1535For example, by setting the variable to 3, neighbor discovery layer
1536will take 3 consecutive hints in maximum.
1537After receiving 3 hints, neighbor discovery layer will perform
1538normal neighbor discovery process.
1539.It Li icmp6.nd6_mmaxtries
1540The variable specifies
1541.Dv MAX_MULTICAST_SOLICIT
1542constant in IPv6 neighbor discovery specification
1543.Pq RFC 2461 .
1544.It Li icmp6.nd6_prune
1545The variable specifies interval between IPv6 neighbor cache babysitting,
1546in seconds.
1547.It Li icmp6.nd6_umaxtries
1548The variable specifies
1549.Dv MAX_UNICAST_SOLICIT
1550constant in IPv6 neighbor discovery specification
1551.Pq RFC 2461 .
1552.It Li icmp6.nd6_useloopback
1553If set to non-zero, kernel IPv6 stack will use loopback interface for
1554local traffic.
1555.It Li icmp6.nodeinfo
1556The variable enables responses to ICMPv6 node information queries.
1557If you set the variable to 0, responses will not be generated for
1558ICMPv6 node information queries.
1559Since node information queries can have a security impact, it is
1560possible to fine tune which responses should be answered.
1561Two separate bits can be set.
1562.Bl -tag -width "12345"
1563.It 1
1564Respond to ICMPv6 FQDN queries, e.g.
1565.Li ping6 -w .
1566.It 2
1567Respond to ICMPv6 node addresses queries, e.g.
1568.Li ping6 -a .
1569.El
1570.It Li icmp6.rediraccept
1571If set to non-zero, the host will accept ICMPv6 redirect packets.
1572Note that IPv6 routers will never accept ICMPv6 redirect packets,
1573and the variable is meaningful on IPv6 hosts
1574.Pq non-router
1575only.
1576.It Li icmp6.redirtimeout
1577The variable specifies lifetime of routing entries generated by incoming
1578ICMPv6 redirect.
1579.It Li udp6.do_loopback_cksum
1580Perform UDP checksum on loopback.
1581.It Li udp6.recvspace
1582Default UDP receive buffer size.
1583.It Li udp6.sendspace
1584Default UDP send buffer size.
1585.El
1586.Pp
1587We reuse net.*.tcp for
1588.Tn TCP
1589over
1590.Tn IPv6 ,
1591and therefore we do not have variables net.*.tcp6.
1592Variables net.inet6.udp6 have identical meaning to net.inet.udp.
1593Please refer to
1594.Li PF_INET
1595section above.
1596For variables net.*.ipsec6, please refer to
1597.Xr ipsec 4 .
1598.It Li net.key ( PF_KEY )
1599Get or set various global information about the IPsec key management.
1600The third level name is the variable name.
1601The currently defined variable and names are:
1602.Bl -column "blockacq_lifetime" "integer" "Changeable" -offset indent
1603.It Sy Variable name	Type	Changeable
1604.It debug	integer	yes
1605.It spi_try	integer	yes
1606.It spi_min_value	integer	yes
1607.It spi_max_value	integer	yes
1608.It larval_lifetime	integer	yes
1609.It blockacq_count	integer	yes
1610.It blockacq_lifetime	integer	yes
1611.It esp_keymin	integer	yes
1612.It esp_auth	integer	yes
1613.It ah_keymin	integer	yes
1614.El
1615.Pp
1616The variables are as follows:
1617.Bl -tag -width "123456"
1618.It Li debug
1619Turn on debugging message from within the kernel.
1620The value is a bitmap, as defined in
1621.Pa /usr/include/netkey/key_debug.h .
1622.It Li spi_try
1623The number of times the kernel will try to obtain an unique SPI
1624when it generates it from random number generator.
1625.It Li spi_min_value
1626Minimum SPI value when generating it within the kernel.
1627.It Li spi_max_value
1628Maximum SPI value when generating it within the kernel.
1629.It Li larval_lifetime
1630Lifetime for LARVAL SAD entries, in seconds.
1631.It Li blockacq_count
1632Number of ACQUIRE PF_KEY messages to be blocked after an ACQUIRE message.
1633It avoids flood of ACQUIRE PF_KEY from being sent from the kernel to the
1634key management daemon.
1635.It Li blockacq_lifetime
1636Lifetime of ACQUIRE PF_KEY message.
1637.It Li esp_keymin
1638Minimum ESP key length, in bits.
1639The value is used when the kernel creates proposal payload
1640on ACQUIRE PF_KEY message.
1641.It Li esp_auth
1642Whether ESP authentication should be used or not.
1643Non-zero value indicates that ESP authentication should be used.
1644The value is used when the kernel creates proposal payload
1645on ACQUIRE PF_KEY message.
1646.It Li ah_keymin
1647Minimum AH key length, in bits,
1648The value is used when the kernel creates proposal payload
1649on ACQUIRE PF_KEY message.
1650.El
1651.El
1652.Sh The proc.* subtree
1653The string and integer information available for the
1654.Li proc
1655level is detailed below.
1656The changeable column shows whether a process with appropriate
1657privilege may change the value.
1658These values are per-process,
1659and as such may change from one process to another.
1660When a process is created,
1661the default values are inherited from its parent.
1662When a set-user-ID or set-group-ID binary is executed, the
1663value of PROC_PID_CORENAME is reset to the system default value.
1664The second level name is either the magic value PROC_CURPROC, which
1665points to the current process, or the PID of the target process.
1666.Bl -column "proc.pid.corename" "string" "not applicable" -offset indent
1667.It Sy Third level name	Type	Changeable
1668.It proc.pid.corename	string	yes
1669.It proc.pid.rlimit	node	not applicable
1670.It proc.pid.stopfork	int	yes
1671.It proc.pid.stopexec	int	yes
1672.It proc.pid.stopexit	int	yes
1673.El
1674.Bl -tag -width "123456"
1675.It Li proc.pid.corename ( PROC_PID_CORENAME )
1676The template used for the core dump file name (see
1677.Xr core 5
1678for details).
1679The base name must either be
1680.Nm core
1681or end with the suffix ``.core'' (the super-user may set arbitrary names).
1682By default it points to KERN_DEFCORENAME.
1683.It Li proc.pid.rlimit ( PROC_PID_LIMIT )
1684Return resources limits, as defined for the
1685.Xr getrlimit 2
1686and
1687.Xr setrlimit 2
1688system calls.
1689The fourth level name is one of:
1690.Bl -tag -width PROC_PID_LIMIT_MEMLOCKAA
1691.It Li proc.pid.rlimit.cputime ( PROC_PID_LIMIT_CPU )
1692The maximum amount of CPU time (in seconds) to be used by each process.
1693.It Li proc.pid.rlimit.filesize ( PROC_PID_LIMIT_FSIZE )
1694The largest size (in bytes) file that may be created.
1695.It Li proc.pid.rlimit.datasize ( PROC_PID_LIMIT_DATA )
1696The maximum size (in bytes) of the data segment for a process;
1697this defines how far a program may extend its break with the
1698.Xr sbrk 2
1699system call.
1700.It Li proc.pid.rlimit.stacksize ( PROC_PID_LIMIT_STACK )
1701The maximum size (in bytes) of the stack segment for a process;
1702this defines how far a program's stack segment may be extended.
1703Stack extension is performed automatically by the system.
1704.It Li proc.pid.rlimit.coredumpsize ( PROC_PID_LIMIT_CORE )
1705The largest size (in bytes)
1706.Pa core
1707file that may be created.
1708.It Li proc.pid.rlimit.memoryuse ( PROC_PID_LIMIT_RSS )
1709The maximum size (in bytes) to which a process's resident set size may
1710grow.
1711This imposes a limit on the amount of physical memory to be given to
1712a process; if memory is tight, the system will prefer to take memory
1713from processes that are exceeding their declared resident set size.
1714.It Li proc.pid.rlimit.memorylocked ( PROC_PID_LIMIT_MEMLOCK )
1715The maximum size (in bytes) which a process may lock into memory
1716using the
1717.Xr mlock 2
1718function.
1719.It Li proc.pid.rlimit.maxproc ( PROC_PID_LIMIT_NPROC )
1720The maximum number of simultaneous processes for this user id.
1721.It Li proc.pid.rlimit.descriptors ( PROC_PID_LIMIT_NOFILE )
1722The maximum number of open files for this process.
1723.It Li proc.pid.rlimit.sbsize ( PROC_PID_LIMIT_SBSIZE )
1724The maximum size (in bytes) of the socket buffers
1725set by the
1726.Xr setsockopt 2
1727.Dv SO_RCVBUF
1728and
1729.Dv SO_SNDBUF
1730options.
1731.El
1732.Pp
1733The fifth level name is one of
1734.Li soft ( PROC_PID_LIMIT_TYPE_SOFT ) or
1735.Li hard ( PROC_PID_LIMIT_TYPE_HARD ) ,
1736to select respectively the soft or hard limit.
1737Both are of type integer.
1738.It Li proc.pid.stopfork ( PROC_PID_STOPFORK )
1739If non zero, the process' children will be stopped after
1740.Xr fork 2
1741calls.
1742The children is created in the SSTOP state and is never scheduled
1743for running before being stopped.
1744This feature helps attaching a process with a debugger such as
1745.Xr gdb 1
1746before it had the opportunity to actually do anything.
1747.Pp
1748This value is inherited by the process's children, and it also
1749apply to emulation specific system calls that fork a new process, such as
1750.Fn sproc
1751or
1752.Fn clone .
1753.It Li proc.pid.stopexec ( PROC_PID_STOPEXEC )
1754If non zero, the process will be stopped on next
1755.Xr exec 3
1756call.
1757The process created by
1758.Xr exec 3
1759is created in the SSTOP state and is never scheduled for running
1760before being stopped.
1761This feature helps attaching a process with a debugger such as
1762.Xr gdb 1
1763before it had the opportunity to actually do anything.
1764.Pp
1765This value is inherited by the process's children.
1766.It Li proc.pid.stopexit ( PROC_PID_STOPEXIT )
1767If non zero, the process will be stopped on when it has cause to exit,
1768either by way of calling
1769.Xr exit 3 ,
1770.Xr _exit 2 ,
1771or by the receipt of a specific signal.
1772The process is stopped before any of its resources or vm space is
1773released allowing examination of the termination state of a process
1774before it disappears.
1775This feature can be used to examine the final conditions of the
1776process's vmspace via
1777.Xr pmap 1
1778or its resource settings with
1779.Xr sysctl 8
1780before it disappears.
1781.Pp
1782This value is also inherited by the process's children.
1783.El
1784.Sh The user.* subtree ( CTL_USER )
1785The string and integer information available for the
1786.Li user
1787level is detailed below.
1788The changeable column shows whether a process with appropriate
1789privilege may change the value.
1790.Bl -column "user.coll_weights_max" "integer" "Changeable" -offset indent
1791.It Sy Second level name	Type	Changeable
1792.It user.atexit_max	integer	no
1793.It user.bc_base_max	integer	no
1794.It user.bc_dim_max	integer	no
1795.It user.bc_scale_max	integer	no
1796.It user.bc_string_max	integer	no
1797.It user.coll_weights_max	integer	no
1798.It user.cs_path	string	no
1799.It user.expr_nest_max	integer	no
1800.It user.line_max	integer	no
1801.It user.posix2_c_bind	integer	no
1802.It user.posix2_c_dev	integer	no
1803.It user.posix2_char_term	integer	no
1804.It user.posix2_fort_dev	integer	no
1805.It user.posix2_fort_run	integer	no
1806.It user.posix2_localedef	integer	no
1807.It user.posix2_sw_dev	integer	no
1808.It user.posix2_upe	integer	no
1809.It user.posix2_version	integer	no
1810.It user.re_dup_max	integer	no
1811.It user.stream_max	integer	no
1812.It user.stream_max	integer	no
1813.It user.tzname_max	integer	no
1814.El
1815.Bl -tag -width "123456"
1816.It Li user.atexit_max ( USER_ATEXIT_MAX )
1817The maximum number of functions that may be registered with
1818.Xr atexit 3 .
1819.It Li user.bc_base_max ( USER_BC_BASE_MAX )
1820The maximum ibase/obase values in the
1821.Xr bc 1
1822utility.
1823.It Li user.bc_dim_max ( USER_BC_DIM_MAX )
1824The maximum array size in the
1825.Xr bc 1
1826utility.
1827.It Li user.bc_scale_max ( USER_BC_SCALE_MAX )
1828The maximum scale value in the
1829.Xr bc 1
1830utility.
1831.It Li user.bc_string_max ( USER_BC_STRING_MAX )
1832The maximum string length in the
1833.Xr bc 1
1834utility.
1835.It Li user.coll_weights_max ( USER_COLL_WEIGHTS_MAX )
1836The maximum number of weights that can be assigned to any entry of
1837the LC_COLLATE order keyword in the locale definition file.
1838.It Li user.cs_path ( USER_CS_PATH )
1839Return a value for the
1840.Ev PATH
1841environment variable that finds all the standard utilities.
1842.It Li user.expr_nest_max ( USER_EXPR_NEST_MAX )
1843The maximum number of expressions that can be nested within
1844parenthesis by the
1845.Xr expr 1
1846utility.
1847.It Li user.line_max ( USER_LINE_MAX )
1848The maximum length in bytes of a text-processing utility's input
1849line.
1850.It Li user.posix2_char_term ( USER_POSIX2_CHAR_TERM )
1851Return 1 if the system supports at least one terminal type capable of
1852all operations described in POSIX 1003.2, otherwise 0.
1853.It Li user.posix2_c_bind ( USER_POSIX2_C_BIND )
1854Return 1 if the system's C-language development facilities support the
1855C-Language Bindings Option, otherwise 0.
1856.It Li user.posix2_c_dev ( USER_POSIX2_C_DEV )
1857Return 1 if the system supports the C-Language Development Utilities Option,
1858otherwise 0.
1859.It Li user.posix2_fort_dev ( USER_POSIX2_FORT_DEV )
1860Return 1 if the system supports the FORTRAN Development Utilities Option,
1861otherwise 0.
1862.It Li user.posix2_fort_run ( USER_POSIX2_FORT_RUN )
1863Return 1 if the system supports the FORTRAN Runtime Utilities Option,
1864otherwise 0.
1865.It Li user.posix2_localedef ( USER_POSIX2_LOCALEDEF )
1866Return 1 if the system supports the creation of locales, otherwise 0.
1867.It Li user.posix2_sw_dev ( USER_POSIX2_SW_DEV )
1868Return 1 if the system supports the Software Development Utilities Option,
1869otherwise 0.
1870.It Li user.posix2_upe ( USER_POSIX2_UPE )
1871Return 1 if the system supports the User Portability Utilities Option,
1872otherwise 0.
1873.It Li user.posix2_version ( USER_POSIX2_VERSION )
1874The version of POSIX 1003.2 with which the system attempts to comply.
1875.It Li user.re_dup_max ( USER_RE_DUP_MAX )
1876The maximum number of repeated occurrences of a regular expression
1877permitted when using interval notation.
1878.It Li user.stream_max ( USER_STREAM_MAX )
1879The minimum maximum number of streams that a process may have open
1880at any one time.
1881.It Li user.tzname_max ( USER_TZNAME_MAX )
1882The minimum maximum number of types supported for the name of a
1883timezone.
1884.El
1885.Sh The vm.* subtree ( CTL_VM )
1886The string and integer information available for the
1887.Li vm
1888level is detailed below.
1889The changeable column shows whether a process with appropriate
1890privilege may change the value.
1891.Bl -column "Second level name" "struct uvmexp_sysctl" "Changeable" -offset indent
1892.It Sy Second level name	Type	Changeable
1893.It vm.anonmax	int	yes
1894.It vm.anonmin	int	yes
1895.It vm.bufcache	int	yes
1896.It vm.bufmem	int	no
1897.It vm.bufmem_hiwater	int	yes
1898.It vm.bufmem_lowater	int	yes
1899.It vm.execmax	int	yes
1900.It vm.execmin	int	yes
1901.It vm.filemax	int	yes
1902.It vm.filemin	int	yes
1903.It vm.loadavg	struct loadavg	no
1904.It vm.maxslp	int	no
1905.It vm.nkmempages	int	no
1906.It vm.uspace	int	no
1907.It vm.uvmexp	struct uvmexp	no
1908.It vm.uvmexp2	struct uvmexp_sysctl	no
1909.It vm.vmmeter	struct vmtotal	no
1910.El
1911.Pp
1912.Bl -tag -width "123456"
1913.It Li vm.anonmax ( VM_ANONMAX )
1914The percentage of physical memory which will be reclaimed
1915from other types of memory usage to store anonymous application data.
1916.It Li vm.anonmin ( VM_ANONMIN )
1917The percentage of physical memory which will be always be available for
1918anonymous application data.
1919.It Li vm.bufcache ( VM_BUFCACHE )
1920The percentage of physical memory which will be available
1921for the buffer cache.
1922.It Li vm.bufmem ( VM_BUFMEM )
1923The amount of kernel memory that is being used by the buffer cache.
1924.It Li vm.bufmem_lowater ( VM_BUFMEM_LOWATER )
1925The minimum amount of kernel memory to reserve for the
1926buffer cache.
1927.It Li vm.bufmem_hiwater ( VM_BUFMEM_HIWATER )
1928The maximum amount of kernel memory to be used for the
1929buffer cache.
1930.It Li vm.execmax ( VM_EXECMAX )
1931The percentage of physical memory which will be reclaimed
1932from other types of memory usage to store cached executable data.
1933.It Li vm.execmin ( VM_EXECMIN )
1934The percentage of physical memory which will be always be available for
1935cached executable data.
1936.It Li vm.filemax ( VM_FILEMAX )
1937The percentage of physical memory which will be reclaimed
1938from other types of memory usage to store cached file data.
1939.It Li vm.filemin ( VM_FILEMIN )
1940The percentage of physical memory which will be always be available for
1941cached file data.
1942.It Li vm.loadavg ( VM_LOADAVG )
1943Return the load average history.
1944The returned data consists of a
1945.Va struct loadavg .
1946.It Li vm.maxslp ( VM_MAXSLP )
1947The value of the maxslp kernel global variable.
1948.It Li vm.vmmeter ( VM_METER )
1949Return system wide virtual memory statistics.
1950The returned data consists of a
1951.Va struct vmtotal .
1952.It Li vm.uspace ( VM_USPACE )
1953The number of bytes allocated for each kernel stack.
1954.It Li vm.uvmexp ( VM_UVMEXP )
1955Return system wide virtual memory statistics.
1956The returned data consists of a
1957.Va struct uvmexp .
1958.It Li vm.uvmexp2 ( VM_UVMEXP2 )
1959Return system wide virtual memory statistics.
1960The returned data consists of a
1961.Va struct uvmexp_sysctl .
1962.\" XXX vm.idlezero
1963.El
1964.Sh The ddb.* subtree ( CTL_DDB )
1965The integer information available for the
1966.Li ddb
1967level is detailed below.
1968The changeable column shows whether a process with appropriate
1969privilege may change the value.
1970.\" XXX sort
1971.Bl -column "Second level name" "integer" "Changeable" -offset indent
1972.It Sy Second level name	Type	Changeable
1973.It ddb.radix	integer	yes
1974.It ddb.maxoff	integer	yes
1975.It ddb.lines	integer	yes
1976.It ddb.tabstops	integer	yes
1977.It ddb.onpanic	integer	yes
1978.It ddb.fromconsole	integer	yes
1979.El
1980.Pp
1981.Bl -tag -width "123456"
1982.It Li ddb.radix ( DBCTL_RADIX )
1983The input and output radix.
1984.It Li ddb.maxoff ( DBCTL_MAXOFF )
1985The maximum symbol offset.
1986.It Li ddb.lines ( DBCTL_LINES )
1987Number of display lines.
1988.It Li ddb.tabstops ( DBCTL_TABSTOPS )
1989Tab width.
1990.It Li ddb.onpanic ( DBCTL_ONPANIC )
1991If non-zero, DDB will be entered if the kernel panics.
1992.It Li ddb.fromconsole ( DBCTL_FROMCONSOLE )
1993If not zero, DDB may be entered by sending a break on a serial
1994console or by a special key sequence on a graphics console.
1995.\" XXX tee_msgbuf maxwidth commandonenter
1996.El
1997.Pp
1998These MIB nodes are also available as variables from within the DDB.
1999See
2000.Xr ddb 4
2001for more details.
2002.Sh The security.* subtree ( CTL_SECURITY )
2003The
2004.Li security
2005level contains various security-related settings for
2006the system.
2007Available settings are detailed below.
2008.Pp
2009.Bl -tag -width "123456"
2010.It Li security.curtain
2011If non-zero, will filter return objects according to the user-id
2012requesting information about them, preventing from users any
2013access to objects they don't own.
2014.Pp
2015At the moment, it affects
2016.Xr ps 1 ,
2017.Xr netstat 1
2018(for
2019.Dv PF_INET ,
2020.Dv PF_INET6 ,
2021and
2022.Dv PF_UNIX
2023PCBs), and
2024.Xr w 1 .
2025.It Li security.models
2026.Nx
2027supports pluggable security models.
2028Every security model used, whether if loaded as a module or built with the system,
2029is required to add an entry to this node with at least one element,
2030.Dq name ,
2031indicating the name of the security model.
2032.Pp
2033In addition to the name, any settings and other information private to the
2034security model will be available under this node.
2035See
2036.Xr secmodel 9
2037for more information.
2038.It Li security.pax
2039Settings for PaX -- exploit mitigation features.
2040For more information on any of the PaX features, please see
2041.Xr paxctl 8
2042and
2043.Xr security 8 .
2044.Pp
2045.Bl -tag -width "123456"
2046.It Li security.pax.aslr.enable
2047Enable PaX ASLR (Address Space Layout Randomization).
2048.Pp
2049The value of this
2050knob must be non-zero for PaX ASLR to be enabled, even if a program is set to
2051explicit enable.
2052.It Li security.pax.aslr.global
2053Specifies the default global policy for programs without an
2054explicit enable/disable flag.
2055.Pp
2056When non-zero, all programs will get PaX ASLR, except those exempted with
2057.Xr paxctl 8  .
2058Otherwise, all programs will not get PaX ASLR, except those specifically
2059marked as such with
2060.Xr paxctl 8 .
2061.It Li security.pax.mprotect.enable
2062Enable PaX MPROTECT restrictions.
2063.Pp
2064These are
2065.Xr mprotect 2
2066restrictions to better enforce a W^X policy.
2067The value of this
2068knob must be non-zero for PaX MPROTECT to be enabled, even if a
2069program is set to explicit enable.
2070.It Li security.pax.mprotect.global
2071Specifies the default global policy for programs without an
2072explicit enable/disable flag.
2073.Pp
2074When non-zero, all programs will get the PaX MPROTECT restrictions,
2075except those exempted with
2076.Xr paxctl 8  .
2077Otherwise, all programs will not get the PaX MPROTECT restrictions,
2078except those specifically marked as such with
2079.Xr paxctl 8 .
2080.It Li security.pax.segvguard.enable
2081Enable PaX Segvguard.
2082.Pp
2083PaX Segvguard can detect and prevent certain exploitation attempts, where
2084an attacker may try for example to brute-force function return addresses
2085of respawning daemons.
2086.Pp
2087.Em Note :
2088The
2089.Nx
2090interface and implementation of the Segvguard is still experimental, and may
2091change in future releases.
2092.It Li security.pax.segvguard.global
2093Specifies the default global policy for programs without an
2094explicit enable/disable flag.
2095.Pp
2096When non-zero, all programs will get the PaX Segvguard,
2097except those exempted with
2098.Xr paxctl 8  .
2099Otherwise, no program will get the PaX Segvguard restrictions,
2100except those specifically marked as such with
2101.Xr paxctl 8 .
2102.It Li security.pax.segvguard.expiry_timeout
2103If the max number was not reached within this timeout (in seconds), the entry
2104will expire.
2105.It Li security.pax.segvguard.suspend_timeout
2106Number of seconds to suspend a user from running a faulting program when the
2107limit was exceeded.
2108.It Li security.pax.segvguard.max_crashes
2109Max number of segfaults a program can receive before suspension.
2110.El
2111.El
2112.Sh The vendor.* subtree ( CTL_VENDOR )
2113The
2114.Li vendor
2115toplevel name is reserved to be used by vendors who wish to
2116have their own private MIB tree.
2117Intended use is to store values under
2118.Dq vendor.\*[Lt]yourname\*[Gt].* .
2119.Sh SEE ALSO
2120.Xr sysctl 3 ,
2121.Xr ipsec 4 ,
2122.Xr tcp 4 ,
2123.Xr security 8 ,
2124.Xr sysctl 8
2125.Sh HISTORY
2126The
2127.Nm
2128variables first appeared in
2129.Bx 4.4 .
2130