1 <?xml version='
1.0'
?> <!--*- Mode: nxml; nxml-child-indent: 2; indent-tabs-mode: nil -*-->
2 <!DOCTYPE refentry PUBLIC
"-//OASIS//DTD DocBook XML V4.2//EN"
3 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
6 This file is part of systemd.
8 Copyright 2012 Lennart Poettering
10 systemd is free software; you can redistribute it and/or modify it
11 under the terms of the GNU Lesser General Public License as published by
12 the Free Software Foundation; either version 2.1 of the License, or
13 (at your option) any later version.
15 systemd is distributed in the hope that it will be useful, but
16 WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 Lesser General Public License for more details.
20 You should have received a copy of the GNU Lesser General Public License
21 along with systemd; If not, see <http://www.gnu.org/licenses/>.
24 <refentry id=
"systemd-cgtop"
25 xmlns:
xi=
"http://www.w3.org/2001/XInclude">
28 <title>systemd-cgtop
</title>
29 <productname>systemd
</productname>
33 <contrib>Developer
</contrib>
34 <firstname>Lennart
</firstname>
35 <surname>Poettering
</surname>
36 <email>lennart@poettering.net
</email>
42 <refentrytitle>systemd-cgtop
</refentrytitle>
43 <manvolnum>1</manvolnum>
47 <refname>systemd-cgtop
</refname>
48 <refpurpose>Show top control groups by their resource usage
</refpurpose>
53 <command>systemd-cgtop
</command>
54 <arg choice=
"opt" rep=
"repeat">OPTIONS
</arg>
59 <title>Description
</title>
61 <para><command>systemd-cgtop
</command> shows the top control
62 groups of the local Linux control group hierarchy, ordered by
63 their CPU, memory, or disk I/O load. The display is refreshed in
64 regular intervals (by default every
1s), similar in style to
65 <citerefentry project='man-pages'
><refentrytitle>top
</refentrytitle><manvolnum>1</manvolnum></citerefentry>.
</para>
67 <para>If
<command>systemd-cgtop
</command> is not connected to a
68 tty, no column headers are printed and the default is to only run
69 one iteration. The
<varname>--iterations=
</varname> argument, if
70 given, is honored. This mode is suitable for scripting.
</para>
72 <para>Resource usage is only accounted for control groups in the
73 relevant hierarchy, i.e. CPU usage is only accounted for control
74 groups in the
<literal>cpuacct
</literal> hierarchy, memory usage
75 only for those in
<literal>memory
</literal> and disk I/O usage for
76 those in
<literal>blkio
</literal>. If resource monitoring for
77 these resources is required, it is recommended to add the
78 <varname>CPUAccounting=
1</varname>,
79 <varname>MemoryAccounting=
1</varname> and
80 <varname>BlockIOAccounting=
1</varname> settings in the unit files
82 <citerefentry><refentrytitle>systemd.resource-control
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
85 <para>The CPU load value can be between
0 and
100 times the number of
86 processors the system has. For example, if the system has
8 processors,
87 the CPU load value is going to be between
0% and
800%. The number of
88 processors can be found in
<literal>/proc/cpuinfo
</literal>.
</para>
90 <para>To emphasize this: unless
91 <literal>CPUAccounting=
1</literal>,
92 <literal>MemoryAccounting=
1</literal> and
93 <literal>BlockIOAccounting=
1</literal> are enabled for the
94 services in question, no resource accounting will be available for
95 system services and the data shown by
96 <command>systemd-cgtop
</command> will be incomplete.
</para>
100 <title>Options
</title>
102 <para>The following options are understood:
</para>
106 <term><option>-p
</option></term>
107 <term><option>--order=path
</option></term>
109 <listitem><para>Order by control group
110 path name.
</para></listitem>
114 <term><option>-t
</option></term>
115 <term><option>--order=tasks
</option></term>
117 <listitem><para>Order by number of tasks/processes in the control group.
</para></listitem>
121 <term><option>-c
</option></term>
122 <term><option>--order=cpu
</option></term>
124 <listitem><para>Order by CPU load.
</para></listitem>
128 <term><option>-m
</option></term>
129 <term><option>--order=memory
</option></term>
131 <listitem><para>Order by memory usage.
</para></listitem>
135 <term><option>-i
</option></term>
136 <term><option>--order=io
</option></term>
138 <listitem><para>Order by disk I/O load.
</para></listitem>
142 <term><option>-b
</option></term>
143 <term><option>--batch
</option></term>
145 <listitem><para>Run in
"batch" mode: do not accept input and
146 run until the iteration limit set with
147 <option>--iterations=
</option> is exhausted or until killed.
148 This mode could be useful for sending output from
149 <command>systemd-cgtop
</command> to other programs or to a
150 file.
</para></listitem>
154 <term><option>-r
</option></term>
155 <term><option>--raw
</option></term>
157 <listitem><para>Format byte counts (as in memory usage and I/O metrics)
158 with raw numeric values rather than human-readable
159 numbers.
</para></listitem>
163 <term><option>--cpu=percentage
</option></term>
164 <term><option>--cpu=time
</option></term>
166 <listitem><para>Controls whether the CPU usage is shown as
167 percentage or time. By default, the CPU usage is shown as
168 percentage. This setting may also be toggled at runtime by
169 pressing the
<keycap>%
</keycap> key.
</para></listitem>
173 <term><option>-P
</option></term>
175 <listitem><para>Count only userspace processes instead of all
176 tasks. By default, all tasks are counted: each kernel thread
177 and each userspace thread individually. With this setting,
178 kernel threads are excluded from the counting and each
179 userspace process only counts as one, regardless how many
180 threads it consists of. This setting may also be toggled at
181 runtime by pressing the
<keycap>P
</keycap> key. This option
182 may not be combined with
183 <option>-k
</option>.
</para></listitem>
187 <term><option>-k
</option></term>
189 <listitem><para>Count only userspace processes and kernel
190 threads instead of all tasks. By default, all tasks are
191 counted: each kernel thread and each userspace thread
192 individually. With this setting, kernel threads are included in
193 the counting and each userspace process only counts as on one,
194 regardless how many threads it consists of. This setting may
195 also be toggled at runtime by pressing the
<keycap>k
</keycap>
196 key. This option may not be combined with
197 <option>-P
</option>.
</para></listitem>
201 <term><option>--recursive=
</option></term>
203 <listitem><para>Controls whether the number of processes shown
204 for a control group shall include all processes that are
205 contained in any of the child control groups as well. Takes a
206 boolean argument, which defaults to
<literal>yes
</literal>. If
207 enabled, the processes in child control groups are included, if
208 disabled, only the processes in the control group itself are
209 counted. This setting may also be toggled at runtime by
210 pressing the
<keycap>r
</keycap> key. Note that this setting
211 only applies to process counting, i.e. when the
212 <option>-P
</option> or
<option>-k
</option> options are
213 used. It has not effect if all tasks are counted, in which
214 case the counting is always recursive.
</para></listitem>
218 <term><option>-n
</option></term>
219 <term><option>--iterations=
</option></term>
221 <listitem><para>Perform only this many iterations. A value of
222 0 indicates that the program should run
223 indefinitely.
</para></listitem>
227 <term><option>-d
</option></term>
228 <term><option>--delay=
</option></term>
230 <listitem><para>Specify refresh delay in seconds (or if one of
231 <literal>ms
</literal>,
<literal>us
</literal>,
232 <literal>min
</literal> is specified as unit in this time
233 unit). This setting may also be increased and decreased at
234 runtime by pressing the
<keycap>+
</keycap> and
235 <keycap>-
</keycap> keys.
</para></listitem>
239 <term><option>--depth=
</option></term>
241 <listitem><para>Maximum control group tree traversal depth.
242 Specifies how deep
<command>systemd-cgtop
</command> shall
243 traverse the control group hierarchies. If
0 is specified,
244 only the root group is monitored. For
1, only the first level
245 of control groups is monitored, and so on. Defaults to
250 <term><option>-M
<replaceable>MACHINE
</replaceable></option></term>
251 <term><option>--machine=
<replaceable>MACHINE
</replaceable></option></term>
253 <listitem><para>Limit control groups shown to the part
254 corresponding to the container
255 <replaceable>MACHINE
</replaceable>.
</para></listitem>
258 <xi:include href=
"standard-options.xml" xpointer=
"help" />
259 <xi:include href=
"standard-options.xml" xpointer=
"version" />
267 <para><command>systemd-cgtop
</command> is an interactive tool and
268 may be controlled via user input using the following keys:
</para>
272 <term><keycap>h
</keycap></term>
274 <listitem><para>Shows a short help text.
</para></listitem>
278 <term><keycap function=
"space"/></term>
280 <listitem><para>Immediately refresh output.
</para></listitem>
284 <term><keycap>q
</keycap></term>
286 <listitem><para>Terminate the program.
</para></listitem>
290 <term><keycap>p
</keycap></term>
291 <term><keycap>t
</keycap></term>
292 <term><keycap>c
</keycap></term>
293 <term><keycap>m
</keycap></term>
294 <term><keycap>i
</keycap></term>
296 <listitem><para>Sort the control groups by path, number of
297 tasks, CPU load, memory usage, or I/O load, respectively. This
298 setting may also be controlled using the
299 <option>--order=
</option> command line
300 switch.
</para></listitem>
304 <term><keycap>%
</keycap></term>
306 <listitem><para>Toggle between showing CPU time as time or
307 percentage. This setting may also be controlled using the
308 <option>--cpu=
</option> command line switch.
</para></listitem>
312 <term><keycap>+
</keycap></term>
313 <term><keycap>-
</keycap></term>
315 <listitem><para>Increase or decrease refresh delay,
316 respectively. This setting may also be controlled using the
317 <option>--delay=
</option> command line
318 switch.
</para></listitem>
322 <term><keycap>P
</keycap></term>
324 <listitem><para>Toggle between counting all tasks, or only
325 userspace processes. This setting may also be controlled using
326 the
<option>-P
</option> command line switch (see
327 above).
</para></listitem>
331 <term><keycap>k
</keycap></term>
333 <listitem><para>Toggle between counting all tasks, or only
334 userspace processes and kernel threads. This setting may also
335 be controlled using the
<option>-k
</option> command line
336 switch (see above).
</para></listitem>
340 <term><keycap>r
</keycap></term>
342 <listitem><para>Toggle between recursively including or
343 excluding processes in child control groups in control group
344 process counts. This setting may also be controlled using the
345 <option>--recursive=
</option> command line switch. This key is
346 not available if all tasks are counted, it is only available
347 if processes are counted, as enabled with the
348 <keycap>P
</keycap> or
<keycap>k
</keycap>
349 keys.
</para></listitem>
356 <title>Exit status
</title>
358 <para>On success,
0 is returned, a non-zero failure code
363 <title>See Also
</title>
365 <citerefentry><refentrytitle>systemd
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
366 <citerefentry><refentrytitle>systemctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
367 <citerefentry><refentrytitle>systemd-cgls
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
368 <citerefentry><refentrytitle>systemd.resource-control
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
369 <citerefentry project='man-pages'
><refentrytitle>top
</refentrytitle><manvolnum>1</manvolnum></citerefentry>