1PMVAL(1) General Commands Manual PMVAL(1)
2
3
4
6 pmval, pmevent - arbitrary performance metrics value dumper
7
9 pmval [-dgLrvVz?] [-a archive] [-A align] [-f N] [-h host] [-i
10 instances] [-K spec] [-n pmnsfile] [-O offset] [-p port] [-s samples]
11 [-S starttime] [-t interval] [-T endtime] [-U archive] [-w width] [-x
12 pattern] [-Z timezone] [--container=name] [--derived=file] metricname
13
15 pmval prints current or archived values for the nominated performance
16 metric. The metric of interest is named in the metricname argument,
17 subject to instance qualification with the -i flag as described below.
18
19 Unless directed to another host by the -h option, or to a set of ar‐
20 chives by the -a or -U options, pmval will contact the Performance Met‐
21 rics Collector Daemon (PMCD) on the local host to obtain the required
22 information.
23
24 The metricname argument may also be given in the metric specification
25 syntax, as described in PCPIntro(1), where the source, metric and
26 instance may all be included in the metricname, e.g. thathost:ker‐
27 nel.all.load["1 minute"]. When this format is used, none of the -h or
28 -a or -U options may be specified.
29
30 When using the metric specification syntax, the ``hostname'' @ is
31 treated specially and causes pmval to use a local context to collect
32 metrics from PMDAs on the local host without PMCD. Only some metrics
33 are available in this mode.
34
35 When processing a set of archives, pmval may relinquish its own timing
36 control, and operate as a ``slave'' of a pmtime(1) process that uses a
37 GUI dialog to provide timing control. In this case, either the -g
38 option should be used to start pmval as the sole slave of a new
39 pmtime(1) instance, or -p should be used to attach pmval to an existing
40 pmtime(1) instance via the IPC channel identified by the port argument.
41
42 The -S, -T, -O and -A options may be used to define a time window to
43 restrict the samples retrieved, set an initial origin within the time
44 window, or specify a ``natural'' alignment of the sample times; refer
45 to PCPIntro(1) for a complete description of these options.
46
47 The output from pmval is directed to standard output. The following
48 symbols may occasionally appear, in place of a metric value, in pmval
49 output: A question mark symbol (?) indicates that a value is no longer
50 available for that metric instance. An exclamation mark (!) indicates
51 that a 64-bit counter wrapped during the sample.
52
54 The available command line options are:
55
56 -a archive, --archive=archive
57 Performance metric values are retrieved from the set of Perfor‐
58 mance Co-Pilot (PCP) archive log files identified by the archive
59 argument, which is a comma-separated list of names, each of which
60 may be the base name of an archive or the name of a directory con‐
61 taining one or more archives. See also -U.
62
63 -A align, --align=align
64 Force the initial sample to be aligned on the boundary of a natu‐
65 ral time unit align. Refer to PCPIntro(1) for a complete descrip‐
66 tion of the syntax for align.
67
68 -d, --delay
69 When replaying from an archive, this option requests that the pre‐
70 vailing real-time delay be applied between samples (see -t) to
71 effect a pause, rather than the default behaviour of replaying at
72 full speed.
73
74 -f precision, --precision=precision
75 Numbers are reported in ``fixed point'' notation, rather than the
76 default scientific notation, using precision digits for precision.
77 Each number will be up to the column width determined by the
78 default heuristics, else the -w option if specified, and include
79 precision digits after the decimal point. So, the options -f 3 -w
80 8 would produce numbers of the form 9999.999. A value of zero for
81 precision omits the decimal point and any fractional digits.
82
83 -g, --guimode
84 Start pmval as the slave of a new pmtime(1) process for replay of
85 archived performance data using the pmtime(1) graphical user
86 interface.
87
88 -h host, --host=host
89 Fetch performance metrics from pmcd(1) on host, rather than from
90 the default localhost.
91
92 -i instances, --instances=instances
93 Specify a list of one or more names of instances for the nominated
94 performance metric - just these instances will be retrieved and
95 reported (the default is to report all instances). The list must
96 be a single argument, with elements of the list separated by com‐
97 mas and/or white space.
98
99 The instance name may be quoted with single (') or double (")
100 quotes for those cases where the instance name contains white
101 space or commas.
102
103 Multiple -i options are allowed as an alternative way of specify‐
104 ing more than one instance of interest.
105
106 As an example, the following are all equivalent:
107
108 $ pmval -i "'1 minute','5 minute'" kernel.all.load
109 $ pmval -i '"1 minute","5 minute"' kernel.all.load
110 $ pmval -i "'1 minute' '5 minute'" kernel.all.load
111 $ pmval -i "'1 minute'" -i "'5 minute'" kernel.all.load
112 $ pmval 'localhost:kernel.all.load["1 minute","5 minute"]'
113
114
115 -K spec, --spec-local=spec
116 When fetching metrics from a local context (see -L), the -K option
117 may be used to control the DSO PMDAs that should be made accessi‐
118 ble. The spec argument conforms to the syntax described in
119 pmSpecLocalPMDA(3). More than one -K option may be used.
120
121 -L, --local-PMDA
122 Use a local context to collect metrics from DSO PMDAs on the local
123 host without PMCD. See also -K.
124
125 -n pmnsfile, --namespace=pmnsfile
126 Load an alternative Performance Metrics Name Space (PMNS(5)) from
127 the file pmnsfile.
128
129 -O origin, --origin=origin
130 When reporting archived metrics, start reporting at origin within
131 the time window (see -S and -T). Refer to PCPIntro(1) for a com‐
132 plete description of the syntax for origin.
133
134 -p port, --guiport=port
135 Attach pmval to an existing pmtime(1) time control process
136 instance via the IPC channel identified by the port argument.
137 This option is normally only used by other tools, e.g.
138 pmchart(1), when they launch pmval with synchronized time control.
139
140 -r, --raw
141 Print raw values for cumulative counter metrics. Normally cumula‐
142 tive counter metrics are converted to rates. For example, disk
143 transfers are reported as number of disk transfers per second dur‐
144 ing the preceding sample interval, rather than the raw value of
145 number of disk transfers since the machine was booted. If you
146 specify this option, the raw metric values are printed.
147
148 -s samples, --samples=samples
149 The samples argument defines the number of samples to be retrieved
150 and reported. If samples is 0 or -s is not specified, pmval will
151 sample and report continuously (in real time mode) or until the
152 end of the set of PCP archives (in archive mode).
153
154 -S starttime, --start=starttime
155 When reporting archived metrics, the report will be restricted to
156 those records logged at or after starttime. Refer to PCPIntro(1)
157 for a complete description of the syntax for starttime.
158
159 -t interval, --interval=interval
160 Set the reporting interval to something other than the default 1
161 second. The interval argument follows the syntax described in
162 PCPIntro(1), and in the simplest form may be an unsigned integer
163 (the implied units in this case are seconds).
164
165 -T endtime, --finish=endtime
166 When reporting archived metrics, the report will be restricted to
167 those records logged before or at endtime. Refer to PCPIntro(1)
168 for a complete description of the syntax for endtime.
169
170 -U archive, --nointerp=archive
171 Performance metric values are retrieved from the Performance Co-
172 Pilot (PCP) archive. The argument is a comma-separated list of
173 names, each of which may be the base name of an archive or the
174 name of a directory containing one or more archives. However,
175 unlike -a every recorded value in the archive for the selected
176 metric and instances is reported (so no interpolation mode, and
177 the sample interval (-t option) is ignored. See also -a.
178
179 At most one of the options -a and -U may be specified.
180
181 -v, --verbose
182 Enable verbose mode.
183
184 -V, --version
185 Display version number and exit.
186
187 -w width, --width=width
188 Set the width of each column of output to be width columns. If
189 not specified columns are wide enough to accommodate the largest
190 value of the type being printed.
191
192 -x pattern, --filter=pattern
193 The given pattern is sent to the performance metric domain agent
194 for the requested metricname before any values are requested.
195 This serves two purposes. Firstly, it provides a mechanism for
196 server-side event filtering that is customisable for individual
197 event streams. In addition, some performance metrics domain
198 agents also use the PMCD store mechanism to provide a basic secu‐
199 rity model (e.g. for sensitive log files, only a client host with
200 pmStore(3) access would be able to access the event stream).
201
202 As pattern may be processed by regcomp(3) it should be a non-empty
203 string. Use . (dot) for a “match all” pattern.
204
205 -z, --hostzone
206 Use the local timezone of the host that is the source of the per‐
207 formance metrics, as identified by either the -h or the -a or the
208 -U options. The default is to use the timezone of the local host.
209
210 -Z timezone, --timezone=timezone
211 Use timezone for the date and time. Timezone is in the format of
212 the environment variable TZ as described in environ(7).
213
214 -?, --help
215 Display usage message and exit.
216
217 --container=container
218 Specify an individual container to be queried.
219
220 --derived=file
221 Load derived metric definitions from file.
222
224 By default, pmval attempts to display non-integer numeric values in a
225 way that does not distort the inherent precision (rarely more than 4
226 significant digits), and tries to maintain a tabular format in the out‐
227 put. These goals are sometimes in conflict.
228
229 In the absence of the -f option (described above), the following table
230 describes the formats used for different ranges of numeric values for
231 any metric that is of type PM_TYPE_FLOAT or PM_TYPE_DOUBLE, or any met‐
232 ric that has the semantics of a counter (for which pmval reports the
233 rate converted value):
234
235 ┌──────────┬──────────────────────┐
236 │ Format │ Value Range │
237 ├──────────┼──────────────────────┤
238 │ ! │ No values available │
239 │9.999E-99 │ < 0.1 │
240 │ 0.0 │ 0 │
241 │ 9.9999 │ > 0 and <= 0.9999 │
242 │ 9.999 │ > 0.9999 and < 9.999 │
243 │ 99.99 │ > 9.999 and < 99.99 │
244 │ 999.9 │ > 99.99 and < 999.9 │
245 │9999. │ > 999.9 and < 9999 │
246 │9.999E+99 │ > 9999 │
247 └──────────┴──────────────────────┘
249 All are generated on standard error and are intended to be self-
250 explanatory.
251
253 Environment variables with the prefix PCP_ are used to parameterize the
254 file and directory names used by PCP. On each installation, the file
255 /etc/pcp.conf contains the local values for these variables. The
256 $PCP_CONF variable may be used to specify an alternative configuration
257 file, as described in pcp.conf(5).
258
259 For environment variables affecting PCP tools, see pmGetOptions(3).
260
262 PCPIntro(1), pmcd(1), pmchart(1), pmdumplog(1), pmdumptext(1),
263 pminfo(1), pmlogger(1), pmrep(1), pmtime(1), PMAPI(3), pmStore(3),
264 pmSpecLocalPMDA(3), pcp.conf(5), pcp.env(5) and PMNS(5).
265
266
267
268Performance Co-Pilot PCP PMVAL(1)