1PCP-PIDSTAT(1) General Commands Manual PCP-PIDSTAT(1)
2
3
4
6 pcp-pidstat - Report statistics for Linux tasks.
7
9 pcp pidstat [-s N] [-t DELTA] [-I] [-a FILE] [-f FORMAT] [-G NAME]
10 [-U[USERNAME]] [-p PID1,PID2..] [-R/-r/-k] [-V] [-?]
11
12
14 pcp-pidstat command is used for monitoring individual tasks currently
15 being managed by the Linux kernel. Using various options it helps user
16 to see useful information related with the processes. This information
17 can include percentage CPU, memory and stack usages, scheduling and
18 priority information. By default pcp-pidstat reports live data for the
19 local host but can also report for a remote host (--host) or from a
20 previously captured PCP archive (-a). pcp-pidstat fetches the data to
21 be analysed from PMCD unless provided -a option. (See below)
22
23
25 Output control options:
26
27 -s N, --samples=N
28 Set the number of samples to be displayed. Since the first sam‐
29 ple is used for the rate conversion of some of the metrics, the
30 total number of samples reported are one less that N. Default is
31 continous.
32
33
34 -t DELTA, --interval=DELTA
35 Set the interval between two samples. Default is one second.
36
37
38 General Options:
39
40
41 -I In an SMP environment, indicates that tasks CPU usage should
42 be divided by the total number of processors.
43
44
45 -a FILE, --archive=FILE
46 Causes pcp-pidstat to use the specified archive than connecting
47 to PMCD. The argument to -a is a comma-separated list of names,
48 each of which may be the base name of an archive or the name of
49 a directory containing one or more archives. Archives can be
50 created using pmlogger(1).
51
52
53 --host Current performance metric values are retrieved from the nomi‐
54 nated host machine.
55
56
57 -B Report process states. The argument to -B is one of the follow‐
58 ing:
59
60 detail
61 Show total time processes have spent in each of the 5
62 different states
63
64 all
65 Show total time processes spent in their current state
66
67 [R,S,T,D,Z]
68 A comma separated list of states which can be any or all
69 of [R,S,T,Z,D]. For example, -B R,S will report processes cur‐
70 rently in either R or S states and not report processes cur‐
71 rently in any other states.
72
73
74 -f Use the format string for formatting the timestamp. The format
75 will be used with Python's datetime.strftime method which is
76 mostly the same as that described in strftime(3). An empty for‐
77 mat string(i.e,"") will remove the timestamps from the output.
78 The default with stdout is %H:%M:%S.
79
80
81 -R Report realtime priority and scheduling policy information. The
82 following values may be displayed:
83
84 UID
85 The real user identification number of the task being
86 monitored.
87
88 USER
89 The name of the real user owning the task being moni‐
90 tored.
91
92 PID
93 The identification number of the task being monitored.
94
95 prio
96 The realtime priority of the task being monitored.
97
98 policy
99 The scheduling policy of the task being monitored.
100
101 Command
102 The command name of the task.
103
104
105 -r Report page faults and memory utilization. The following values
106 may be displayed:
107
108 UID
109 The real user identification number of the task being
110 monitored.
111
112 USER
113 The name of the real user owning the task being moni‐
114 tored.
115
116 PID
117 The identification number of the task being monitored.
118
119 minflt/s
120 Total number of minor faults the task has made per sec‐
121 ond, those which have not required loading a memory page from
122 disk.
123
124 majflt/s
125 Total number of major faults the task has made per sec‐
126 ond, those which have required loading a memory page from disk.
127
128 VSZ
129 Virtual Size: The virtual memory usage of entire task in
130 kilobytes.
131
132 RSS
133 Resident Set Size: The non-swapped physical memory used
134 by the task in kilobytes.
135
136 %MEM
137 The tasks's currently used share of available physical
138 memory.
139
140 Command
141 The command name of the task.
142
143
144 -k Report stack utilization. The following values may be dis‐
145 played:
146
147 UID
148 The real user identification number of the task being mon‐
149 itored.
150
151 USER
152 The name of the real user owning the task being monitored.
153
154 PID
155 The identification number of the task being monitored.
156
157 StkSize
158 The amount of memory in kilobytes reserved for the task as
159 stack, but not necessarily used.
160
161 StkRef
162 The amount of memory in kilobytes used as stack, refer‐
163 enced by the task.
164
165 Command
166 The command name of the task.
167
168
169 -V Display version and exit
170
171
172 -? , --help
173 Display help and exit
174
175
176 Ouput Filter Options
177
178
179 -G NAME, --process-name=NAME
180 Display only processes whose command name includes the string
181 NAME. This string can be a regular expression.
182
183
184 -U[USERNAME], --user-name=[USERNAME]
185 Display the real user name of the tasks being monitored instead
186 of the UID. If username is specified, then only tasks belonging
187 to the specified user are displayed.
188
189
190 -p PID1,PID2.. , --pid-list=PID1,PID2..
191 Display only processes whose PID belongs to the given Pid List.
192
193
195 pcp-pidstat is inspired by the pidstat(1) command and aims to be com‐
196 mand line and output compatible with it.
197
198
200 pcp(1), pidstat(1), PCPIntro(1), pmParseInterval(3) and environ(7).
201
202
203
204Performance Co-Pilot PCP PCP-PIDSTAT(1)