1PCP-MPSTAT(1) General Commands Manual PCP-MPSTAT(1)
2
6 pcp-mpstat - Report CPU and interrupt related statistics.
7
9 pcp [pcp options] mpstat [ -A ] [ -u ] [ -V ] [ -I { SUM | CPU | SCPU |
10 ALL } ] [ -P { cpu1,cpu2 [,...] | ON | ALL } ] [ -t interval ] [ -s
11 count ] [ -a archive ] [ -? ]
12
14 pcp-mpstat command writes to standard output activities for each avail‐
15 able processor, processor 0 being the first one. If no activity/option
16 has been selected, then the default report is the CPU utilization (-u)
17 report.
18 The interval parameter specifies the amount of time in seconds between
20 each report. The default is one second. The value of count parameter
21 determines the number of samples to be displayed. The default is con‐
22 tinous.
23
25 When invoked via the pcp(1) command, the -h/--host, -a/--archive,
26 -O/--origin, -s/--samples, -t/--interval, -Z/--timezone and several
27 other pcp options become indirectly available; refer to PCPIntro(1) for
28 a complete description of these options.
29 The additional command line options available for pcp-mpstat are:
31 -A This option is equivalent to specifying -u -I ALL -P ALL
33 -I { SUM | CPU | SCPU | ALL }
35 Report interrupts statistics.
36 With the SUM keyword, the pcp-mpstat command reports the total
37 number of interrupts per processor. The following values are
38 displayed:
39 CPU
41 Processor number. The keyword all indicates that statis‐
42 tics are calculated as averages among all processors.
43 intr/s
45 Show the total number of interrupts received per second
46 by the CPU or CPUs.
47 With the CPU keyword, the number of each individual interrupt
49 received per second by the CPU or CPUs is displayed. Interrupts
50 are those under the kernel.percpu.interrupts metric tree.
51 With the SCPU keyword, the number of each individual software
53 interrupt received per second by the CPU or CPUs is displayed.
54 Software interrupts are those under the kernel.percpu.softirqs
55 tree
56 The ALL keyword is equivalent to specifying all the keywords
58 above and therefore all the interrupts statistics are displayed.
59 -P { cpu1,cpu2[,...] | ON | ALL }
62 Indicate the processor number for which statistics are to be
63 reported. cpu1 and cpu2 are the processor numbers. A list of
64 required processor numbers can be provided. Note that processor
65 0 is the first processor.
66 The ON keyword indicates that statistics are to be reported for
68 every online processor, whereas the ALL keyword indicates that
69 statistics are to be reported for all processors.
70 -u Report CPU utilization. The following values are displayed:
72 CPU
74 Processor number. The keyword ALL indicates that statis‐
75 tics are calculated as averages among all processors.
76 %usr
78 Show the percentage of CPU utilization that occurred
79 while executing at the user level (application).
80 %nice
82 Show the percentage of CPU utilization that occurred
83 while executing at the user level with nice priority.
84 %sys
86 Show the percentage of CPU utilization that occurred
87 while executing at the system level (kernel). Note that
88 this does not include time spent servicing hardware and
89 software interrupts.
90 %iowait
92 Show the percentage of time that the CPU or CPUs were
93 idle during which the system had an outstanding disk I/O
94 request.
95 %irq
97 Show the percentage of time spent by the CPU or CPUs to
98 service hardware interrupts.
99 %soft
101 Show the percentage of time spent by the CPU or CPUs to
102 service software interrupts.
103 %steal
105 Show the percentage of time spent in involuntary wait by
106 the virtual CPU or CPUs while the hypervisor was servic‐
107 ing another virtual processor.
108 %guest
110 Show the percentage of time spent by the CPU or CPUs to
111 run a virtual processor.
112 %gnice
114 Show the percentage of time spent by the CPU or CPUs to
115 run a niced guest.
116 %idle
118 Show the percentage of time that the CPU or CPUs were
119 idle and the system did not have an outstanding disk I/O
120 request.
121 Note: On SMP machines a processor that does not have any activ‐
123 ity at all is a disabled (offline) processor.
124 -s N , --samples = N
127 Set the number of samples to be displayed. Since the first sam‐
128 ple is used for the rate conversion of some of the metrics, the
129 total number of samples reported are one less that N. Default
130 is continous.
131 -t DELTA , --interval = DELTA
134 Set the interval between two samples. The default is one sec‐
135 ond.
136 -a FILE , --archive = FILE
139 Causes pcp-mpstat to use the specified archive than connecting
140 to PMCD. The argument to -a is a comma-separated list of names,
141 each of which may be the base name of an archive or the name of
142 a directory containing one or more archives.
143 -V , --version
146 Print version number then exit.
147 -? , --help
150 Print usage message then exit.
151
153 pcp-mpstat -t 2 -s 5
154 Display four reports of global statistics among all processors
155 at two second intervals.
156 pcp mpstat -P ALL -t 2 -s 5
158 Display four reports of statistics for all processors at two
159 second intervals.
160
162 pcp-mpstat is inspired by the mpstat(1) command and aims to be command
163 line and output compatible with it.
164
166 TZ and LC_TIME environment variables can be used to override the
167 default date display format for pcp-mpstat.
168
170 Environment variables with the prefix PCP_ are used to parameterize the
171 file and directory names used by PCP. On each installation, the file
172 /etc/pcp.conf contains the local values for these variables. The
173 $PCP_CONF variable may be used to specify an alternative configuration
174 file, as described in pcp.conf(5).
175 For environment variables affecting PCP tools, see pmGetOptions(3).
177
179 PCPIntro(1), pcp(1), mpstat(1), pmParseInterval(3), pmTraversePMNS(3)
180 and environ(7).
181Performance Co-Pilot PCP PCP-MPSTAT(1)