xref: /freebsd/usr.bin/systat/systat.1 (revision 22054f88914b51113f77f6eccc11353a891f9f3e)

.\" Copyright (c) 1985, 1990, 1993
.\"	The Regents of the University of California.  All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 3. Neither the name of the University nor the names of its contributors
.\"    may be used to endorse or promote products derived from this software
.\"    without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\"	@(#)systat.1	8.2 (Berkeley) 12/30/93
.\" $FreeBSD$
.\"
.Dd February 10, 2021
.Dt SYSTAT 1
.Os
.Sh NAME
.Nm systat
.Nd display system statistics
.Sh SYNOPSIS
.Nm
.Op Fl Ar display
.Op Ar display-commands
.Op Ar refresh-interval
.Sh DESCRIPTION
The
.Nm
utility displays various system statistics in a screen oriented fashion
using the curses screen display library,
.Xr ncurses 3 .
.Pp
While
.Nm
is running the screen is usually divided into two windows (an exception
is the vmstat display which uses the entire screen).
The
upper window depicts the current system load average.
The
information displayed in the lower window may vary, depending on
user commands.
The last line on the screen is reserved for user
input and error messages.
.Pp
By default
.Nm
displays the processes getting the largest percentage of the processor
in the lower window.
Other displays show swap space usage, disk I/O statistics (a la
.Xr iostat 8 ) ,
virtual memory statistics (a la
.Xr vmstat 8 ) ,
TCP/IP statistics,
and network connections (a la
.Xr netstat 1 ) .
.Pp
Input is interpreted at two different levels.
A ``global'' command interpreter processes all keyboard input.
If this command interpreter fails to recognize a command, the
input line is passed to a per-display command interpreter.
This
allows each display to have certain display-specific commands.
.Pp
Command line options:
.Bl -tag -width "refresh_interval"
.It Fl Ns Ar display
The
.Fl
flag expects
.Ar display
to be one of:
.Ic icmp ,
.Ic icmp6 ,
.Ic ifstat ,
.Ic iolat ,
.Ic iostat ,
.Ic ip ,
.Ic ip6 ,
.Ic netstat ,
.Ic pigs ,
.Ic sctp ,
.Ic swap ,
.Ic tcp ,
.Ic vmstat ,
or
.Ic zarc ,
These displays can also be requested interactively (without the
.Dq Fl )
and are described in
full detail below.
.It Ar refresh-interval
The
.Ar refresh-value
specifies the screen refresh time interval in seconds.
Time interval can be fractional.
.It Ar display-commands
A list of commands specific to this display.
These commands can also be entered interactively and are described for
each display separately below.
If the command requires arguments, they can be specified as separate
command line arguments.
A command line argument
.Fl -
will finish display commands.
For example:
.Pp
.Dl Nm Fl ifstat Fl match Ar bge0,em1 Fl pps
.Pp
This will display statistics of packets per second for network interfaces
named as bge0 and em1.
.Pp
.Dl Nm Fl iostat Fl numbers Fl - Ar 2.1
.Pp
This will display all IO statistics in a numeric format and the information
will be refreshed each 2.1 seconds.
.El
.Pp
Certain characters cause immediate action by
.Nm .
These are
.Bl -tag -width Fl
.It Ic \&^L
Refresh the screen.
.It Ic \&^G
Print the name of the current ``display'' being shown in
the lower window and the refresh interval.
.It Ic \&:
Move the cursor to the command line and interpret the input
line typed as a command.
While entering a command the
current character erase, word erase, and line kill characters
may be used.
.El
.Pp
The following commands are interpreted by the ``global''
command interpreter.
.Bl -tag -width Fl
.It Ic help
Print the names of the available displays on the command line.
.It Ic load
Print the load average over the past 1, 5, and 15 minutes
on the command line.
.It Ic stop
Stop refreshing the screen.
.It Xo
.Op Ic start
.Op Ar number
.Xc
Start (continue) refreshing the screen.
If a second, numeric,
argument is provided it is interpreted as a refresh interval
(in seconds).
Supplying only a number will set the refresh interval to this
value.
.It Ic quit
Exit
.Nm .
(This may be abbreviated to
.Ic q . )
.El
.Pp
The available displays are:
.Bl -tag -width Ic
.It Ic pigs
Display, in the lower window, those processes resident in main
memory and getting the
largest portion of the processor (the default display).
When less than 100% of the
processor is scheduled to user processes, the remaining time
is accounted to the ``idle'' process.
.It Ic icmp
Display, in the lower window, statistics about messages received and
transmitted by the Internet Control Message Protocol
.Pq Dq ICMP .
The left half of the screen displays information about received
packets, and the right half displays information regarding transmitted
packets.
.Pp
The
.Ic icmp
display understands two commands:
.Ic mode
and
.Ic reset .
The
.Ic mode
command is used to select one of four display modes, given as its argument:
.Bl -tag -width absoluteXX -compact
.It Ic rate :
show the rate of change of each value in packets (the default)
per second
.It Ic delta :
show the rate of change of each value in packets per refresh interval
.It Ic since :
show the total change of each value since the display was last reset
.It Ic absolute :
show the absolute value of each statistic
.El
.Pp
The
.Ic reset
command resets the baseline for
.Ic since
mode.
The
.Ic mode
command with no argument will display the current mode in the command
line.
.It Ic icmp6
This display is like the
.Ic icmp
display,
but displays statistics for IPv6 ICMP.
.It Ic ip
Otherwise identical to the
.Ic icmp
display, except that it displays IP and UDP statistics.
.It Ic ip6
Like the
.Ic ip
display,
except that it displays IPv6 statistics.
It does not display UDP statistics.
.It Ic sctp
Like
.Ic icmp ,
but with SCTP statistics.
.It Ic tcp
Like
.Ic icmp ,
but with TCP statistics.
.It Ic iolat
Display statistics describing the hardware latencies of I/O operations as
computed by the
.Va CAM_IOSCHED_DYNAMIC
option.
This option must be in the kernel config file of the running kernel for this
display to work.
All devices are displayed as there is currently no way to filter them.
The statistics displayed for the I/O latencies are the percentiles with
sufficient data during the polling interval to compute.
If a value cannot be estimated ``-'' is displayed.
The P50 (also known as the median), P90, P99 and P99.9 values are computed if
more than 2, 10, 100 or 1000 operations occurred during the polling interval.
The latency is the hardware latency values, and does not include any software
queueing time.
The latencies are estimated based on histogram data computed by the CAM I/O
scheduler and represent estimates of the actual value that are only good to
two or three significant digits.
The display of latency changes based on the scale of the latency to reflect
the precision of the estimates and to fit on the available screen space.
All latencies are reported in milliseconds.
When color is enabled
.Bl -bullet
.It Values below the medium latency threshold are displayed in green.
.It Values between the minimum latency and high latency thresholds are displayed
in magenta.
.It Values above the high latency thresholds are displayed in red.
.Pp
When color is disabled, the default foreground and background colors are always
used.
.Pp
The following commands are specific to the
.Ic iolat
display; the minimum unambiguous prefix may be supplied.
.Pp
.Bl -tag -width Fl -compact
.It Cm color
Toggle the use of color in the display.
The default is on.
.It Cm hi=XXX
Set the high latency threshold to XXX milliseconds.
.It Cm med=XXX
Set the medium latency threshold to XXX milliseconds.
.It Cm read
Toggle the display of statistics about read operations.
The default is on.
.It Cm write
Toggle the display of statistics about write operations.
The default is on.
.It Cm trim
Toggle the display of statistics about trim operations.
The default is on.
.El
.It Ic iostat
Display, in the lower window, statistics about processor use
and disk throughput.
Statistics on processor use appear as
bar graphs of the amount of time executing in user mode (``user''),
in user mode running low priority processes (``nice''), in
system mode (``system''), in interrupt mode (``interrupt''),
and idle (``idle'').
Statistics
on disk throughput show, for each drive, megabytes per second,
average number of disk transactions per second, and
average kilobytes of data per transaction.
This information may be
displayed as bar graphs or as rows of numbers which scroll downward.
Bar
graphs are shown by default.
.Pp
The following commands are specific to the
.Ic iostat
display; the minimum unambiguous prefix may be supplied.
.Pp
.Bl -tag -width Fl -compact
.It Cm numbers
Show the disk I/O statistics in numeric form.
Values are
displayed in numeric columns which scroll downward.
.It Cm bars
Show the disk I/O statistics in bar graph form (default).
.It Cm kbpt
Toggle the display of kilobytes per transaction.
(the default is to
not display kilobytes per transaction).
.El
.It Ic swap
Show information about swap space usage on all the
swap areas compiled into the kernel and processes that are swapped out
as well as a summary of disk activity.
.Pp
The swap areas are displayed first with their name, sizes and
usage percentage.
The
.Ar Used
column indicates the total blocks used so far;
the graph shows the percentage of space in use on each partition.
If there are more than one swap partition in use,
a total line is also shown.
Areas known to the kernel, but not in use are shown as not available.
.Pp
Below the swap space statistics,
processes are listed in order of higher swap area usage.
Pid, username, a part of command line, the total use of swap space
in bytes, the size of process, as well as per-process swap usage percentage and
per-system swap space percentage are shown per process.
.Pp
At the bottom left is the disk usage display.
It reports the number of
kilobytes per transaction, transactions per second, megabytes
per second and the percentage of the time the disk was busy averaged
over the refresh period of the display (by default, five seconds).
The system keeps statistics on most every storage device.
In general, up
to seven devices are displayed.
The devices displayed by default are the
first devices in the kernel's device list.
See
.Xr devstat 3
and
.Xr devstat 9
for details on the devstat system.
.It Ic vmstat
Take over the entire display and show a (rather crowded) compendium
of statistics related to virtual memory usage, process scheduling,
device interrupts, system name translation caching, disk I/O etc.
.Pp
The upper left quadrant of the screen shows the number
of users logged in and the load average over the last one, five,
and fifteen minute intervals.
Below this line are statistics on memory utilization.
The first row of the table reports memory usage only among
active processes, that is processes that have run in the previous
twenty seconds.
The second row reports on memory usage of all processes.
The first column reports on the number of kilobytes in physical pages
claimed by processes.
The second column reports the number of kilobytes in physical pages that
are devoted to read only text pages.
The third and fourth columns report the same two figures for
virtual pages, that is the number of kilobytes in pages that would be
needed if all processes had all of their pages.
Finally the last column shows the number of kilobytes in physical pages
on the free list.
.Pp
Below the memory display is a list of the
average number of threads (over the last refresh interval)
that are runnable (`r'), in page wait (`p'),
in disk wait other than paging (`d'),
sleeping (`s'), and swapped out but desiring to run (`w').
The row also shows the average number of context switches
(`Csw'), traps (`Trp'; includes page faults), system calls (`Sys'),
interrupts (`Int'), network software interrupts (`Sof'), and page
faults (`Flt').
.Pp
Below the process queue length listing is a numerical listing and
a bar graph showing the amount of
system (shown as `='), interrupt (shown as `+'), user (shown as `>'),
nice (shown as `-'), and idle time (shown as ` ').
.Pp
Below the process display are statistics on name translations.
It lists the number of names translated in the previous interval,
the number and percentage of the translations that were
handled by the system wide name translation cache, and
the number and percentage of the translations that were
handled by the per process name translation cache.
.Pp
To the right of the name translations display are lines showing
the number of dirty buffers in the buffer cache (`dtbuf'),
desired maximum size of vnode cache (`desvn'),
number of vnodes actually allocated (`numvn'),
and
number of allocated vnodes that are free (`frevn').
.Pp
At the bottom left is the disk usage display.
It reports the number of
kilobytes per transaction, transactions per second, megabytes
per second and the percentage of the time the disk was busy averaged
over the refresh period of the display (by default, five seconds).
The system keeps statistics on most every storage device.
In general, up
to seven devices are displayed.
The devices displayed by default are the
first devices in the kernel's device list.
See
.Xr devstat 3
and
.Xr devstat 9
for details on the devstat system.
.Pp
Under the date in the upper right hand quadrant are statistics
on paging and swapping activity.
The first two columns report the average number of pages
brought in and out per second over the last refresh interval
due to page faults and the paging daemon.
The third and fourth columns report the average number of pages
brought in and out per second over the last refresh interval
due to swap requests initiated by the scheduler.
The first row of the display shows the average
number of disk transfers per second over the last refresh interval;
the second row of the display shows the average
number of pages transferred per second over the last refresh interval.
.Pp
Below the paging statistics is a column of lines regarding the virtual
memory system.
The first few lines describe,
in units (except as noted below)
of pages per second averaged over the sampling interval,
pages copied on write (`cow'),
pages zero filled on demand (`zfod'),
pages optimally zero filled on demand (`ozfod'),
the ratio of the (average) ozfod / zfod as a percentage (`%ozfod'),
pages freed by the page daemon (`daefr'),
pages freed by exiting processes (`prcfr'),
total pages freed (`totfr'),
pages reactivated from the free list (`react'),
the average number of
times per second that the page daemon was awakened (`pdwak'),
pages analyzed by the page daemon (`pdpgs'),
and
in-transit blocking page faults (`intrn').
Note that the units are special for `%ozfod' and `pdwak'.
The next few lines describe,
as amounts of memory in kilobytes,
pages wired down (`wire'),
active pages (`act'),
inactive pages (`inact'),
dirty pages queued for laundering (`laund'),
and
free pages (`free').
Note that the values displayed are the current transient ones;
they are not averages.
.Pp
At the bottom of this column is a line showing the
amount of virtual memory, in kilobytes, mapped into the buffer cache (`buf').
This statistic is not useful.
It exists only as a placeholder for the corresponding useful statistic
(the amount of real memory used to cache disks).
The most important component of the latter (the amount of real memory
used by the vm system to cache disks) is not available,
but can be guessed from the `inact' amount under some system loads.
.Pp
Running down the right hand side of the display is a breakdown
of the interrupts being handled by the system.
At the top of the list is the total interrupts per second
over the time interval.
The rest of the column breaks down the total on a device
by device basis.
Only devices that have interrupted at least once since boot time are shown.
.Pp
The following commands are specific to the
.Ic vmstat
display; the minimum unambiguous prefix may be supplied.
.Pp
.Bl -tag -width Ar -compact
.It Cm boot
Display cumulative statistics since the system was booted.
.It Cm run
Display statistics as a running total from the point this
command is given.
.It Cm time
Display statistics averaged over the refresh interval (the default).
.It Cm zero
Reset running statistics to zero.
.El
.It Ic zarc
display arc cache usage and hit/miss statistics.
.It Ic netstat
Display, in the lower window, network connections.
By default,
network servers awaiting requests are not displayed.
Each address
is displayed in the format ``host.port'', with each shown symbolically,
when possible.
It is possible to have addresses displayed numerically,
limit the display to a set of ports, hosts, and/or protocols
(the minimum unambiguous prefix may be supplied):
.Pp
.Bl -tag -width Ar -compact
.It Cm all
Toggle the displaying of server processes awaiting requests (this
is the equivalent of the
.Fl a
flag to
.Xr netstat 1 ) .
.It Cm numbers
Display network addresses numerically.
.It Cm names
Display network addresses symbolically.
.It Cm proto Ar protocol
Display only network connections using the indicated
.Ar protocol .
Supported protocols are ``tcp'', ``udp'', and ``all''.
.It Cm ignore Op Ar items
Do not display information about connections associated with
the specified hosts or ports.
Hosts and ports may be specified
by name (``vangogh'', ``ftp''), or numerically.
Host addresses
use the Internet dot notation (``128.32.0.9'').
Multiple items
may be specified with a single command by separating them with
spaces.
.It Cm display Op Ar items
Display information about the connections associated with the
specified hosts or ports.
As for
.Ar ignore ,
.Op Ar items
may be names or numbers.
.It Cm show Op Ar ports\&|hosts
Show, on the command line, the currently selected protocols,
hosts, and ports.
Hosts and ports which are being ignored
are prefixed with a `!'.
If
.Ar ports
or
.Ar hosts
is supplied as an argument to
.Cm show ,
then only the requested information will be displayed.
.It Cm reset
Reset the port, host, and protocol matching mechanisms to the default
(any protocol, port, or host).
.El
.It Ic ifstat
Display the network traffic going through active interfaces on the
system.
Idle interfaces will not be displayed until they receive some
traffic.
.Pp
For each interface being displayed, the current, peak and total
statistics are displayed for incoming and outgoing traffic.
By default,
the
.Ic ifstat
display will automatically scale the units being used so that they are
in a human-readable format.
The scaling units used for the current and
peak
traffic columns can be altered by the
.Ic scale
command.
.Bl -tag -width ".Cm scale Op Ar units"
.It Cm scale Op Ar units
Modify the scale used to display the current and peak traffic over all
interfaces.
The following units are recognised: kbit, kbyte, mbit,
mbyte, gbit, gbyte and auto.
.It Cm pps
Show statistics in packets per second instead of bytes/bits per second.
A subsequent call of
.Ic pps
switches this mode off.
.It Cm match Op Ar patterns
Display only interfaces that match pattern provided as an argument.
Patterns should be in shell syntax separated by whitespaces or commas.
If this command is called without arguments then all interfaces are displayed.
For example:
.Pp
.Dl match em0, bge1
.Pp
This will display em0 and bge1 interfaces.
.Pp
.Dl match em*, bge*, lo0
.Pp
This will display all
.Ic em
interfaces, all
.Ic bge
interfaces and the loopback interface.
.El
.El
.Pp
Commands to switch between displays may be abbreviated to the
minimum unambiguous prefix; for example, ``io'' for ``iostat''.
Certain information may be discarded when the screen size is
insufficient for display.
For example, on a machine with 10
drives the
.Ic iostat
bar graph displays only 3 drives on a 24 line terminal.
When
a bar graph would overflow the allotted screen space it is
truncated and the actual value is printed ``over top'' of the bar.
.Pp
The following commands are common to each display which shows
information about disk drives.
These commands are used to
select a set of drives to report on, should your system have
more drives configured than can normally be displayed on the
screen.
.Pp
.Bl -tag -width Ar -compact
.It Cm ignore Op Ar drives
Do not display information about the drives indicated.
Multiple
drives may be specified, separated by spaces.
.It Cm display Op Ar drives
Display information about the drives indicated.
Multiple drives
may be specified, separated by spaces.
.It Cm only Op Ar drives
Display only the specified drives.
Multiple drives may be specified,
separated by spaces.
.It Cm drives
Display a list of available devices.
.It Cm match Xo
.Ar type , Ns Ar if , Ns Ar pass
.Op | Ar ...
.Xc
Display devices matching the given pattern.
The basic matching
expressions are the same as those used in
.Xr iostat 8
with one difference.
Instead of specifying multiple
.Fl t
arguments which are then ORed together, the user instead specifies multiple
matching expressions joined by the pipe
.Pq Ql \&|
character.
The comma
separated arguments within each matching expression are ANDed together, and
then the pipe separated matching expressions are ORed together.
Any
device matching the combined expression will be displayed, if there is room
to display it.
For example:
.Pp
.Dl match da,scsi | cd,ide
.Pp
This will display all SCSI Direct Access devices and all IDE CDROM devices.
.Pp
.Dl match da | sa | cd,pass
.Pp
This will display all Direct Access devices, all Sequential Access devices,
and all passthrough devices that provide access to CDROM drives.
.El
.Sh FILES
.Bl -tag -width /boot/kernel/kernel -compact
.It Pa /boot/kernel/kernel
For the namelist.
.It Pa /dev/kmem
For information in main memory.
.It Pa /etc/hosts
For host names.
.It Pa /etc/networks
For network names.
.It Pa /etc/services
For port names.
.El
.Sh SEE ALSO
.Xr netstat 1 ,
.Xr kvm 3 ,
.Xr icmp 4 ,
.Xr icmp6 4 ,
.Xr ip 4 ,
.Xr ip6 4 ,
.Xr tcp 4 ,
.Xr udp 4 ,
.Xr gstat 8 ,
.Xr iostat 8 ,
.Xr vmstat 8
.Sh HISTORY
The
.Nm
program appeared in
.Bx 4.3 .
The
.Ic icmp ,
.Ic ip ,
and
.Ic tcp
displays appeared in
.Fx 3.0 ;
the notion of having different display modes for the
ICMP, IP, TCP, and UDP statistics was stolen from the
.Fl C
option to
.Xr netstat 1
in Silicon Graphics' IRIX system.
.Sh BUGS
Certain displays presume a minimum of 80 characters per line.
Ifstat does not detect new interfaces.
The
.Ic vmstat
display looks out of place because it is (it was added in as
a separate display rather than created as a new program).
The
.Ic iolat
command doesn't implement the common device commands including
filtering, as it doesn't use the
.Xr devstat 3
mechanism to obtain its statistics.