1 .\" Copyright (c) 1985, 1990, 1993
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. Neither the name of the University nor the names of its contributors
13 .\" may be used to endorse or promote products derived from this software
14 .\" without specific prior written permission.
16 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
17 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
20 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28 .\" @(#)systat.1 8.2 (Berkeley) 12/30/93
36 .Nd display system statistics
40 .Op Ar display-commands
41 .Op Ar refresh-interval
45 utility displays various system statistics in a screen oriented fashion
46 using the curses screen display library,
51 is running the screen is usually divided into two windows (an exception
52 is the vmstat display which uses the entire screen).
54 upper window depicts the current system load average.
56 information displayed in the lower window may vary, depending on
58 The last line on the screen is reserved for user
59 input and error messages.
63 displays the processes getting the largest percentage of the processor
65 Other displays show swap space usage, disk I/O statistics (a la
67 virtual memory statistics (a la
70 and network connections (a la
73 Input is interpreted at two different levels.
74 A ``global'' command interpreter processes all keyboard input.
75 If this command interpreter fails to recognize a command, the
76 input line is passed to a per-display command interpreter.
78 allows each display to have certain display-specific commands.
81 .Bl -tag -width "refresh_interval"
102 These displays can also be requested interactively (without the
106 .It Ar refresh-interval
109 specifies the screen refresh time interval in seconds.
110 Time interval can be fractional.
111 .It Ar display-commands
112 A list of commands specific to this display.
113 These commands can also be entered interactively and are described for
114 each display separately below.
115 If the command requires arguments, they can be specified as separate
116 command line arguments.
117 A command line argument
119 will finish display commands.
122 .Dl Nm Fl ifstat Fl match Ar bge0,em1 Fl pps
124 This will display statistics of packets per second for network interfaces
125 named as bge0 and em1.
127 .Dl Nm Fl iostat Fl numbers Fl - Ar 2.1
129 This will display all IO statistics in a numeric format and the information
130 will be refreshed each 2.1 seconds.
133 Certain characters cause immediate action by
140 Print the name of the current ``display'' being shown in
141 the lower window and the refresh interval.
143 Move the cursor to the command line and interpret the input
144 line typed as a command.
145 While entering a command the
146 current character erase, word erase, and line kill characters
150 The following commands are interpreted by the ``global''
154 Print the names of the available displays on the command line.
156 Print the load average over the past 1, 5, and 15 minutes
159 Stop refreshing the screen.
164 Start (continue) refreshing the screen.
165 If a second, numeric,
166 argument is provided it is interpreted as a refresh interval
168 Supplying only a number will set the refresh interval to this
173 (This may be abbreviated to
177 The available displays are:
180 Display, in the lower window, those processes resident in main
181 memory and getting the
182 largest portion of the processor (the default display).
183 When less than 100% of the
184 processor is scheduled to user processes, the remaining time
185 is accounted to the ``idle'' process.
187 Display, in the lower window, statistics about messages received and
188 transmitted by the Internet Control Message Protocol
190 The left half of the screen displays information about received
191 packets, and the right half displays information regarding transmitted
196 display understands two commands:
202 command is used to select one of four display modes, given as its argument:
203 .Bl -tag -width absoluteXX -compact
205 show the rate of change of each value in packets (the default)
208 show the rate of change of each value in packets per refresh interval
210 show the total change of each value since the display was last reset
212 show the absolute value of each statistic
217 command resets the baseline for
222 command with no argument will display the current mode in the command
225 This display is like the
228 but displays statistics for IPv6 ICMP.
230 Otherwise identical to the
232 display, except that it displays IP and UDP statistics.
237 except that it displays IPv6 statistics.
238 It does not display UDP statistics.
242 but with SCTP statistics.
246 but with TCP statistics.
248 Display, in the lower window, statistics about processor use
250 Statistics on processor use appear as
251 bar graphs of the amount of time executing in user mode (``user''),
252 in user mode running low priority processes (``nice''), in
253 system mode (``system''), in interrupt mode (``interrupt''),
256 on disk throughput show, for each drive, megabytes per second,
257 average number of disk transactions per second, and
258 average kilobytes of data per transaction.
259 This information may be
260 displayed as bar graphs or as rows of numbers which scroll downward.
262 graphs are shown by default.
264 The following commands are specific to the
266 display; the minimum unambiguous prefix may be supplied.
268 .Bl -tag -width Fl -compact
270 Show the disk I/O statistics in numeric form.
272 displayed in numeric columns which scroll downward.
274 Show the disk I/O statistics in bar graph form (default).
276 Toggle the display of kilobytes per transaction.
278 not display kilobytes per transaction).
281 Show information about swap space usage on all the
282 swap areas compiled into the kernel.
283 The first column is the device name of the partition.
284 The next column is the total space available in the partition.
287 column indicates the total blocks used so far;
288 the graph shows the percentage of space in use on each partition.
289 If there are more than one swap partition in use,
290 a total line is also shown.
291 Areas known to the kernel, but not in use are shown as not available.
293 Take over the entire display and show a (rather crowded) compendium
294 of statistics related to virtual memory usage, process scheduling,
295 device interrupts, system name translation caching, disk I/O etc.
297 The upper left quadrant of the screen shows the number
298 of users logged in and the load average over the last one, five,
299 and fifteen minute intervals.
300 Below this line are statistics on memory utilization.
301 The first row of the table reports memory usage only among
302 active processes, that is processes that have run in the previous
304 The second row reports on memory usage of all processes.
305 The first column reports on the number of kilobytes in physical pages
306 claimed by processes.
307 The second column reports the number of kilobytes in physical pages that
308 are devoted to read only text pages.
309 The third and fourth columns report the same two figures for
310 virtual pages, that is the number of kilobytes in pages that would be
311 needed if all processes had all of their pages.
312 Finally the last column shows the number of kilobytes in physical pages
315 Below the memory display is a list of the
316 average number of threads (over the last refresh interval)
317 that are runnable (`r'), in page wait (`p'),
318 in disk wait other than paging (`d'),
319 sleeping (`s'), and swapped out but desiring to run (`w').
320 The row also shows the average number of context switches
321 (`Csw'), traps (`Trp'; includes page faults), system calls (`Sys'),
322 interrupts (`Int'), network software interrupts (`Sof'), and page
325 Below the process queue length listing is a numerical listing and
326 a bar graph showing the amount of
327 system (shown as `='), interrupt (shown as `+'), user (shown as `>'),
328 nice (shown as `-'), and idle time (shown as ` ').
330 Below the process display are statistics on name translations.
331 It lists the number of names translated in the previous interval,
332 the number and percentage of the translations that were
333 handled by the system wide name translation cache, and
334 the number and percentage of the translations that were
335 handled by the per process name translation cache.
337 To the right of the name translations display are lines showing
338 the number of dirty buffers in the buffer cache (`dtbuf'),
339 desired maximum size of vnode cache (`desvn'),
340 number of vnodes actually allocated (`numvn'),
342 number of allocated vnodes that are free (`frevn').
344 At the bottom left is the disk usage display.
345 It reports the number of
346 kilobytes per transaction, transactions per second, megabytes
347 per second and the percentage of the time the disk was busy averaged
348 over the refresh period of the display (by default, five seconds).
349 The system keeps statistics on most every storage device.
351 to seven devices are displayed.
352 The devices displayed by default are the
353 first devices in the kernel's device list.
358 for details on the devstat system.
360 Under the date in the upper right hand quadrant are statistics
361 on paging and swapping activity.
362 The first two columns report the average number of pages
363 brought in and out per second over the last refresh interval
364 due to page faults and the paging daemon.
365 The third and fourth columns report the average number of pages
366 brought in and out per second over the last refresh interval
367 due to swap requests initiated by the scheduler.
368 The first row of the display shows the average
369 number of disk transfers per second over the last refresh interval;
370 the second row of the display shows the average
371 number of pages transferred per second over the last refresh interval.
373 Below the paging statistics is a column of lines regarding the virtual
375 The first few lines describe,
376 in units (except as noted below)
377 of pages per second averaged over the sampling interval,
378 pages copied on write (`cow'),
379 pages zero filled on demand (`zfod'),
380 pages optimally zero filled on demand (`ozfod'),
381 the ratio of the (average) ozfod / zfod as a percentage (`%ozfod'),
382 pages freed by the page daemon (`daefr'),
383 pages freed by exiting processes (`prcfr'),
384 total pages freed (`totfr'),
385 pages reactivated from the free list (`react'),
386 the average number of
387 times per second that the page daemon was awakened (`pdwak'),
388 pages analyzed by the page daemon (`pdpgs'),
390 in-transit blocking page faults (`intrn').
391 Note that the units are special for `%ozfod' and `pdwak'.
392 The next few lines describe,
393 as amounts of memory in kilobytes,
394 pages wired down (`wire'),
395 active pages (`act'),
396 inactive pages (`inact'),
397 dirty pages queued for laundering (`laund'),
400 Note that the values displayed are the current transient ones;
401 they are not averages.
403 At the bottom of this column is a line showing the
404 amount of virtual memory, in kilobytes, mapped into the buffer cache (`buf').
405 This statistic is not useful.
406 It exists only as a placeholder for the corresponding useful statistic
407 (the amount of real memory used to cache disks).
408 The most important component of the latter (the amount of real memory
409 used by the vm system to cache disks) is not available,
410 but can be guessed from the `inact' amount under some system loads.
412 Running down the right hand side of the display is a breakdown
413 of the interrupts being handled by the system.
414 At the top of the list is the total interrupts per second
415 over the time interval.
416 The rest of the column breaks down the total on a device
418 Only devices that have interrupted at least once since boot time are shown.
420 The following commands are specific to the
422 display; the minimum unambiguous prefix may be supplied.
424 .Bl -tag -width Ar -compact
426 Display cumulative statistics since the system was booted.
428 Display statistics as a running total from the point this
431 Display statistics averaged over the refresh interval (the default).
433 Reset running statistics to zero.
436 display arc cache usage and hit/miss statistics.
438 Display, in the lower window, network connections.
440 network servers awaiting requests are not displayed.
442 is displayed in the format ``host.port'', with each shown symbolically,
444 It is possible to have addresses displayed numerically,
445 limit the display to a set of ports, hosts, and/or protocols
446 (the minimum unambiguous prefix may be supplied):
448 .Bl -tag -width Ar -compact
450 Toggle the displaying of server processes awaiting requests (this
451 is the equivalent of the
456 Display network addresses numerically.
458 Display network addresses symbolically.
459 .It Cm proto Ar protocol
460 Display only network connections using the indicated
462 Supported protocols are ``tcp'', ``udp'', and ``all''.
463 .It Cm ignore Op Ar items
464 Do not display information about connections associated with
465 the specified hosts or ports.
466 Hosts and ports may be specified
467 by name (``vangogh'', ``ftp''), or numerically.
469 use the Internet dot notation (``128.32.0.9'').
471 may be specified with a single command by separating them with
473 .It Cm display Op Ar items
474 Display information about the connections associated with the
475 specified hosts or ports.
479 may be names or numbers.
480 .It Cm show Op Ar ports\&|hosts
481 Show, on the command line, the currently selected protocols,
483 Hosts and ports which are being ignored
484 are prefixed with a `!'.
489 is supplied as an argument to
491 then only the requested information will be displayed.
493 Reset the port, host, and protocol matching mechanisms to the default
494 (any protocol, port, or host).
497 Display the network traffic going through active interfaces on the
499 Idle interfaces will not be displayed until they receive some
502 For each interface being displayed, the current, peak and total
503 statistics are displayed for incoming and outgoing traffic.
507 display will automatically scale the units being used so that they are
508 in a human-readable format.
509 The scaling units used for the current and
511 traffic columns can be altered by the
514 .Bl -tag -width ".Cm scale Op Ar units"
515 .It Cm scale Op Ar units
516 Modify the scale used to display the current and peak traffic over all
518 The following units are recognised: kbit, kbyte, mbit,
519 mbyte, gbit, gbyte and auto.
521 Show statistics in packets per second instead of bytes/bits per second.
524 switches this mode off.
525 .It Cm match Op Ar patterns
526 Display only interfaces that match pattern provided as an argument.
527 Patterns should be in shell syntax separated by whitespaces or commas.
528 If this command is called without arguments then all interfaces are displayed.
533 This will display em0 and bge1 interfaces.
535 .Dl match em*, bge*, lo0
537 This will display all
541 interfaces and the loopback interface.
545 Commands to switch between displays may be abbreviated to the
546 minimum unambiguous prefix; for example, ``io'' for ``iostat''.
547 Certain information may be discarded when the screen size is
548 insufficient for display.
549 For example, on a machine with 10
552 bar graph displays only 3 drives on a 24 line terminal.
554 a bar graph would overflow the allotted screen space it is
555 truncated and the actual value is printed ``over top'' of the bar.
557 The following commands are common to each display which shows
558 information about disk drives.
559 These commands are used to
560 select a set of drives to report on, should your system have
561 more drives configured than can normally be displayed on the
564 .Bl -tag -width Ar -compact
565 .It Cm ignore Op Ar drives
566 Do not display information about the drives indicated.
568 drives may be specified, separated by spaces.
569 .It Cm display Op Ar drives
570 Display information about the drives indicated.
572 may be specified, separated by spaces.
573 .It Cm only Op Ar drives
574 Display only the specified drives.
575 Multiple drives may be specified,
578 Display a list of available devices.
580 .Ar type , Ns Ar if , Ns Ar pass
583 Display devices matching the given pattern.
585 expressions are the same as those used in
588 Instead of specifying multiple
590 arguments which are then ORed together, the user instead specifies multiple
591 matching expressions joined by the pipe
595 separated arguments within each matching expression are ANDed together, and
596 then the pipe separated matching expressions are ORed together.
598 device matching the combined expression will be displayed, if there is room
602 .Dl match da,scsi | cd,ide
604 This will display all SCSI Direct Access devices and all IDE CDROM devices.
606 .Dl match da | sa | cd,pass
608 This will display all Direct Access devices, all Sequential Access devices,
609 and all passthrough devices that provide access to CDROM drives.
612 .Bl -tag -width /boot/kernel/kernel -compact
613 .It Pa /boot/kernel/kernel
616 For information in main memory.
648 the notion of having different display modes for the
649 ICMP, IP, TCP, and UDP statistics was stolen from the
653 in Silicon Graphics' IRIX system.
655 Certain displays presume a minimum of 80 characters per line.
656 Ifstat does not detect new interfaces.
659 display looks out of place because it is (it was added in as
660 a separate display rather than created as a new program).