xref: /freebsd/usr.bin/systat/systat.1 (revision afe61c15161c324a7af299a9b8457aba5afc92db)
1.\" Copyright (c) 1985, 1990, 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. All advertising materials mentioning features or use of this software
13.\"    must display the following acknowledgement:
14.\"	This product includes software developed by the University of
15.\"	California, Berkeley and its contributors.
16.\" 4. Neither the name of the University nor the names of its contributors
17.\"    may be used to endorse or promote products derived from this software
18.\"    without specific prior written permission.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30.\" SUCH DAMAGE.
31.\"
32.\"	@(#)systat.1	8.2 (Berkeley) 12/30/93
33.\"
34.Dd December 30, 1993
35.Dt SYSTAT 1
36.Os BSD 4.3
37.Sh NAME
38.Nm systat
39.Nd display system statistics on a crt
40.Sh SYNOPSIS
41.Nm systat
42.Op Fl display
43.Op Ar refresh-interval
44.Sh DESCRIPTION
45.Nm Systat
46displays various system statistics in a screen oriented fashion
47using the curses screen display library,
48.Xr curses 3 .
49.Pp
50While
51.Nm systat
52is running the screen is usually divided into two windows (an exception
53is the vmstat display which uses the entire screen).  The
54upper window depicts the current system load average.  The
55information displayed in the lower window may vary, depending on
56user commands.  The last line on the screen is reserved for user
57input and error messages.
58.Pp
59By default
60.Nm systat
61displays the processes getting the largest percentage of the processor
62in the lower window.  Other displays show swap space usage, disk
63.Tn I/O
64statistics (a la
65.Xr iostat  1  ) ,
66virtual memory statistics (a la
67.Xr vmstat  1  ) ,
68network ``mbuf'' utilization, and network connections (a la
69.Xr netstat  1  ) .
70.Pp
71Input is interpreted at two different levels.
72A ``global'' command interpreter processes all keyboard input.
73If this command interpreter fails to recognize a command, the
74input line is passed to a per-display command interpreter.  This
75allows each display to have certain display-specific commands.
76.Pp
77Command line options:
78.Bl -tag -width "refresh_interval"
79.It Fl Ns Ar display
80The
81.Fl
82flag expects
83.Ar display
84to be one of:
85.Ic pigs ,
86.Ic iostat ,
87.Ic swap ,
88.Ic mbufs ,
89.Ic vmstat
90or
91.Ic netstat .
92These displays can also be requested interactively (without the
93.Dq Fl )
94and are described in
95full detail below.
96.It Ar refresh-interval
97The
98.Ar refresh-value
99specifies the screen refresh time interval in seconds.
100.El
101.Pp
102Certain characters cause immediate action by
103.Nm systat  .
104These are
105.Bl -tag -width Fl
106.It Ic \&^L
107Refresh the screen.
108.It Ic \&^G
109Print the name of the current ``display'' being shown in
110the lower window and the refresh interval.
111.It Ic \&^Z
112Stop
113.Nm systat  .
114.It Ic \&:
115Move the cursor to the command line and interpret the input
116line typed as a command.  While entering a command the
117current character erase, word erase, and line kill characters
118may be used.
119.El
120.Pp
121The following commands are interpreted by the ``global''
122command interpreter.
123.Bl -tag -width Fl
124.It Ic help
125Print the names of the available displays on the command line.
126.It Ic load
127Print the load average over the past 1, 5, and 15 minutes
128on the command line.
129.It Ic stop
130Stop refreshing the screen.
131.It Xo
132.Op Ic start
133.Op Ar number
134.Xc
135Start (continue) refreshing the screen.  If a second, numeric,
136argument is provided it is interpreted as a refresh interval
137(in seconds).
138Supplying only a number will set the refresh interval to this
139value.
140.It Ic quit
141Exit
142.Nm systat  .
143(This may be abbreviated to
144.Ic q  . )
145.El
146.Pp
147The available displays are:
148.Bl -tag -width Ic
149.It Ic pigs
150Display, in the lower window, those processes resident in main
151memory and getting the
152largest portion of the processor (the default display).
153When less than 100% of the
154processor is scheduled to user processes, the remaining time
155is accounted to the ``idle'' process.
156.It Ic iostat
157Display, in the lower window, statistics about processor use
158and disk throughput.  Statistics on processor use appear as
159bar graphs of the amount of time executing in user mode (``user''),
160in user mode running low priority processes (``nice''), in
161system mode (``system''), and idle (``idle'').  Statistics
162on disk throughput show, for each drive, kilobytes of data transferred,
163number of disk transactions performed, and average seek time
164(in milliseconds).  This information may be displayed as
165bar graphs or as rows of numbers which scroll downward.  Bar
166graphs are shown by default;
167.Pp
168The following commands are specific to the
169.Ic iostat
170display; the minimum unambiguous prefix may be supplied.
171.Pp
172.Bl -tag -width Fl -compact
173.It Cm numbers
174Show the disk
175.Tn I/O
176statistics in numeric form.  Values are
177displayed in numeric columns which scroll downward.
178.It Cm bars
179Show the disk
180.Tn I/O
181statistics in bar graph form (default).
182.It Cm msps
183Toggle the display of average seek time (the default is to
184not display seek times).
185.El
186.It Ic swap
187Show information about swap space usage on all the
188swap areas compiled into the kernel.
189The first column is the device name of the partition.
190The next column is the total space available in the partition.
191The
192.Ar Used
193column indicates the total blocks used so far;
194the graph shows the percentage of space in use on each partition.
195If there are more than one swap partition in use,
196a total line is also shown.
197Areas known to the kernel, but not in use are shown as not available.
198.It Ic mbufs
199Display, in the lower window, the number of mbufs allocated
200for particular uses, i.e. data, socket structures, etc.
201.It Ic vmstat
202Take over the entire display and show a (rather crowded) compendium
203of statistics related to virtual memory usage, process scheduling,
204device interrupts, system name translation cacheing, disk
205.Tn I/O
206etc.
207.Pp
208The upper left quadrant of the screen shows the number
209of users logged in and the load average over the last one, five,
210and fifteen minute intervals.
211Below this line are statistics on memory utilization.
212The first row of the table reports memory usage only among
213active processes, that is processes that have run in the previous
214twenty seconds.
215The second row reports on memory usage of all processes.
216The first column reports on the number of physical pages
217claimed by processes.
218The second column reports the number of physical pages that
219are devoted to read only text pages.
220The third and fourth columns report the same two figures for
221virtual pages, that is the number of pages that would be
222needed if all processes had all of their pages.
223Finally the last column shows the number of physical pages
224on the free list.
225.Pp
226Below the memory display is the disk usage display.
227It reports the number of seeks, transfers, and number
228of kilobyte blocks transferred per second averaged over the
229refresh period of the display (by default, five seconds).
230For some disks it also reports the average milliseconds per seek.
231Note that the system only keeps statistics on at most four disks.
232.Pp
233Below the disk display is a list of the
234average number of processes (over the last refresh interval)
235that are runnable (`r'), in page wait (`p'),
236in disk wait other than paging (`d'),
237sleeping (`s'), and swapped out but desiring to run (`w').
238Below the queue length listing is a numerical listing and
239a bar graph showing the amount of
240system (shown as `='), user (shown as `>'),
241nice (shown as `-'), and idle time (shown as ` ').
242.Pp
243At the bottom left are statistics on name translations.
244It lists the number of names translated in the previous interval,
245the number and percentage of the translations that were
246handled by the system wide name translation cache, and
247the number and percentage of the translations that were
248handled by the per process name translation cache.
249.Pp
250Under the date in the upper right hand quadrant are statistics
251on paging and swapping activity.
252The first two columns report the average number of pages
253brought in and out per second over the last refresh interval
254due to page faults and the paging daemon.
255The third and fourth columns report the average number of pages
256brought in and out per second over the last refresh interval
257due to swap requests initiated by the scheduler.
258The first row of the display shows the average
259number of disk transfers per second over the last refresh interval;
260the second row of the display shows the average
261number of pages transferred per second over the last refresh interval.
262.Pp
263Below the paging statistics is a line listing the average number of
264total reclaims ('Rec'),
265intransit blocking page faults (`It'),
266swap text pages found in free list (`F/S'),
267file system text pages found in free list (`F/F'),
268reclaims from free list
269pages freed by the clock daemon (`Fre'),
270and sequential process pages freed (`SFr')
271per second over the refresh interval.
272.Pp
273Below this line are statistics on the average number of
274zero filled pages (`zf') and demand filled text pages (`xf')
275per second over the refresh period.
276The first row indicates the number of requests that were
277resolved, the second row shows the number that were set up,
278and the last row shows the percentage of setup requests that were
279actually used.
280Note that this percentage is usually less than 100%,
281however it may exceed 100% if a large number of requests
282are actually used long after they were set up during a
283period when no new pages are being set up.
284Thus this figure is most interesting when observed over
285a long time period, such as from boot time
286(see below on getting such a display).
287.Pp
288Below the page fill statistics is a column that
289lists the average number of context switches (`Csw'),
290traps (`Trp'; includes page faults), system calls (`Sys'), interrupts (`Int'),
291characters output to DZ ports using
292.No pseudo Ns -DMA
293(`Pdm'),
294network software interrupts (`Sof'),
295page faults (`Flt'), pages scanned by the page daemon (`Scn'),
296and revolutions of the page daemon's hand (`Rev')
297per second over the refresh interval.
298.Pp
299Running down the right hand side of the display is a breakdown
300of the interrupts being handled by the system.
301At the top of the list is the total interrupts per second
302over the time interval.
303The rest of the column breaks down the total on a device
304by device basis.
305Only devices that have interrupted at least once since boot time are shown.
306.Pp
307The following commands are specific to the
308.Ic vmstat
309display; the minimum unambiguous prefix may be supplied.
310.Pp
311.Bl -tag -width Ar -compact
312.It Cm boot
313Display cumulative statistics since the system was booted.
314.It Cm run
315Display statistics as a running total from the point this
316command is given.
317.It Cm time
318Display statistics averaged over the refresh interval (the default).
319.It Cm zero
320Reset running statistics to zero.
321.El
322.It Ic netstat
323Display, in the lower window, network connections.  By default,
324network servers awaiting requests are not displayed.  Each address
325is displayed in the format ``host.port'', with each shown symbolically,
326when possible.  It is possible to have addresses displayed numerically,
327limit the display to a set of ports, hosts, and/or protocols
328(the minimum unambiguous prefix may be supplied):
329.Pp
330.Bl -tag -width Ar -compact
331.It Cm all
332Toggle the displaying of server processes awaiting requests (this
333is the equivalent of the
334.Fl a
335flag to
336.Ar netstat  1  ) .
337.It Cm numbers
338Display network addresses numerically.
339.It Cm names
340Display network addresses symbolically.
341.It Ar protocol
342Display only network connections using the indicated protocol
343(currently either ``tcp'' or ``udp'').
344.It Cm ignore Op Ar items
345Do not display information about connections associated with
346the specified hosts or ports.  Hosts and ports may be specified
347by name (``vangogh'', ``ftp''), or numerically.  Host addresses
348use the Internet dot notation (``128.32.0.9'').  Multiple items
349may be specified with a single command by separating them with
350spaces.
351.It Cm display Op Ar items
352Display information about the connections associated with the
353specified hosts or ports.  As for
354.Ar ignore  ,
355.Op Ar items
356may be names or numbers.
357.It Cm show Op Ar ports\&|hosts
358Show, on the command line, the currently selected protocols,
359hosts, and ports.  Hosts and ports which are being ignored
360are prefixed with a `!'.  If
361.Ar ports
362or
363.Ar hosts
364is supplied as an argument to
365.Cm show  ,
366then only the requested information will be displayed.
367.It Cm reset
368Reset the port, host, and protocol matching mechanisms to the default
369(any protocol, port, or host).
370.El
371.El
372.Pp
373Commands to switch between displays may be abbreviated to the
374minimum unambiguous prefix; for example, ``io'' for ``iostat''.
375Certain information may be discarded when the screen size is
376insufficient for display.  For example, on a machine with 10
377drives the
378.Ic iostat
379bar graph displays only 3 drives on a 24 line terminal.  When
380a bar graph would overflow the allotted screen space it is
381truncated and the actual value is printed ``over top'' of the bar.
382.Pp
383The following commands are common to each display which shows
384information about disk drives.  These commands are used to
385select a set of drives to report on, should your system have
386more drives configured than can normally be displayed on the
387screen.
388.Pp
389.Bl -tag -width Tx -compact
390.It Cm ignore Op Ar drives
391Do not display information about the drives indicated.  Multiple
392drives may be specified, separated by spaces.
393.It Cm display Op Ar drives
394Display information about the drives indicated.  Multiple drives
395may be specified, separated by spaces.
396.El
397.Sh FILES
398.Bl -tag -width /etc/networks -compact
399.It Pa /vmunix
400For the namelist.
401.It Pa /dev/kmem
402For information in main memory.
403.It Pa /dev/drum
404For information about swapped out processes.
405.It Pa /etc/hosts
406For host names.
407.It Pa /etc/networks
408For network names.
409.It Pa /etc/services
410For port names.
411.El
412.Sh HISTORY
413The
414.Nm systat
415program appeared in
416.Bx 4.3 .
417.Sh BUGS
418Takes 2-10 percent of the cpu.
419Certain displays presume a minimum of 80 characters per line.
420The
421.Ic vmstat
422display looks out of place because it is (it was added in as
423a separate display rather than created as a new program).
424