1PMLC(1) General Commands Manual PMLC(1)
2
6 pmlc - configure active Performance Co-Pilot pmlogger(s) interactively
7
9 pmlc [-e] [-h host] [-i] [-n pmnsfile] [-P] [-p port] [-Z timezone]
10 [-z] [pid]
11
13 pmlc may be used to change those metrics and instances which a pmlog‐
14 ger(1) writes to a Performance Co-Pilot archive (see PCPIntro(1)), the
15 frequency with which the metrics are collected and whether the logging
16 is mandatory, advisory, on or off. It also reports the current logging
17 status of metrics and instances. pmlc may be used to control pmlogger
18 instances on remote hosts as well as those on the local host.
19 Normally pmlc operates on the distributed Performance Metrics Name
21 Space (PMNS), however if the -n option is specified an alternative
22 local PMNS is loaded from the file pmnsfile.
23 If the -P option is specified, pmlc will attempt to start with a con‐
25 nection to the primary pmlogger on the local host. If the -p option is
26 specified, then pmlc will attempt to start with a connection to the
27 pmlogger on this TCP/IP port. Alternatively, if pid is specified, a
28 connection to the pmlogger instance with that process id will be
29 attempted on startup. The -h option may only be used if -P, -p port or
30 a pid is also specified. In that case pmlc will initially connect to
31 the specified (remote) pmlogger instance on host rather than the local
32 host. If the connection to the specified pmlogger instance cannot be
33 established, pmlc will start with no connection. These options typi‐
34 cally allow the same file of pmlc commands to be directed to multiple
35 pmlogger instances by varying the command line arguments. Note that
36 -P, -p port, pid and -h are used only when making an initial connection
37 to a pmlogger instance. They are not used as defaults if subsequent
38 connections are made interactively (see the connect command below).
39 By default, pmlc reports the time of day according to the local time‐
41 zone on the system where pmlc is run. The -Z option changes the time‐
42 zone to timezone in the format of the environment variable TZ as
43 described in environ(7). The -z option changes the timezone to the
44 timezone of the pmlogger instance from which information is being
45 obtained. Only one of -z or -Z may be specified.
46 If standard input is from a tty, pmlc is interactive, with prompts.
48 The -i flag may be used to force interactive behavior, and is typically
49 used in conjunction with -e to echo all command input on standard out‐
50 put.
51 The following commands may be used:
53 show [ loggers ] [ @host ]
55 Displays the process identities of all pmlogger instances running
56 on the local host (or host, if specified). The primary pmlogger
57 pid is parenthesized because it can be referred to as "primary" as
58 well as by its pid.
59 connect pid [ @host ]
61 connect primary [ @host ]
62 Connects pmlc to the specified pmlogger process. Any existing con‐
63 nection to a pmlogger instance is closed first. Each pmlogger
64 instance will accept at most one connection at a time, so if the
65 connection is successfully established, your pmlc will be the only
66 one controlling the pmlogger instance it is connected to.
67 new volume
69 This command works only while a connection to a pmlogger instance
70 is established. It tells the pmlogger to close the current volume
71 of the log and open a new volume. Closed volumes may be archived,
72 e.g. as part of a regular log management procedure to control the
73 size of the physical log files.
74 status
76 This command works only while a connection to a pmlogger instance
77 is established. It prints information about the state of the
78 pmlogger instance and its associated log.
79 timezone local | logger | "timezone"
81 This command sets the time zone used when times are printed. local
82 means use the time zone of the machine that pmlc is running on.
83 logger means use the time zone of the machine where the pmlogger
84 instance is running. Alternatively an explicit timezone enclosed
85 in quotes may be supplied (refer to TZ in environ(7) for details).
86 The default time zone is local unless one of the -z or -Z options
87 has been supplied on the command line.
88 flush
90 This command works only while a connection to a pmlogger instance
91 is established, and requests the pmlogger instance to flush to disk
92 all buffers associated with the current archive. For old-timers,
93 sync is a synonym for flush. In current versions of pmlogger(1)
94 all writes are unbuffered and aligned with the logical records in
95 the external files, so this command achieves nothing, but is
96 retained for backwards compatibility.
97 help
99 Displays a summary of the available commands.
100 h and ? are synonyms for help.
101 quit
103 Exits from pmlc.
104 The remaining commands query and change the logging state of metrics
106 and instances. They will work only if pmlc has a connection to a
107 pmlogger instance. Metrics may be specified as fully qualified names
108 (e.g. hinv.ncpu) or subtrees of the PMNS (e.g. hinv) which are expanded
109 to include all metrics in the subtree (e.g. hinv.ncpu, hinv.cpuclock,
110 etc.). Lists of metrics may be specified by enclosing them in braces
111 with spaces or a comma between metrics (e.g. {hinv.ncpu hinv.ndisk}).
112 Subtrees of metrics may be included in such lists.
113 Each individual metric specification may be further qualified with a
115 space or comma separated list of instances in square brackets (e.g.
116 kernel.all.load["1 minute", "5 minute"]). External instance names or
117 numeric internal instance identifiers or both may be used in the same
118 list (e.g. sample.colour.[red,1,"blue"]). If an instance qualification
119 is applied to a subtree of the PMNS all of the metrics in the subtree
120 must have the same instance domain. Instance qualifications may not be
121 applied to entire lists of metrics but may appear inside such lists.
122 If no instances are specified for a metric, all instances are used.
124 All instances means all instances available at the time the pmlogger
125 instance in question fetches the metrics for logging. If an instance
126 domain changes over time this is not always the same as the set of
127 instances displayed by pmlc, which can only display the currently
128 available instances. To prevent unintentional errors, only the
129 instances that are currently available to pmlc may appear in instance
130 specifications.
131 query metriclist
133 The current logging state of each metric (and instances, where
134 applicable) in metriclist is displayed. This includes the logging
135 state (e.g. on, maybe, off) and the logging interval for each met‐
136 ric (and instance) requested. The following abbreviations pertain‐
137 ing to metrics (and instances) may appear in the output: adv, advi‐
138 sory; mand, mandatory; nl, not in the log; na, in the log but not
139 currently available from its Performance Metrics Domain Agent
140 (PMDA). Where appropriate, an instance name will appear last on a
141 line preceded by its numeric internal instance identifier.
142 [ log ] mandatory on interval metriclist
144 This form of the log command turns on logging for the metrics (and
145 any instances) in metriclist. interval specifies how often the
146 specified metrics/instances should be logged. once indicates that
147 the metrics/instances should appear at most once in the log. More
148 often one would use the optional keyword every followed by a posi‐
149 tive number and one of millisecond (or msec), second (or sec),
150 minute (or min), hour or their plurals.
151 Note that the keyword default which may be used for the default
152 interval in a pmlogger(1) configuration file cannot be used in
153 pmlc.
154 Internal limitations require the interval to be less than (approxi‐
155 mately) 74 hours. An interval value of zero is a synonym for once.
156 [ log ] mandatory off metriclist
158 This tells the pmlogger instance not to log any of the met‐
159 rics/instances in metriclist.
160 [ log ] mandatory maybe metriclist
162 This tells the pmlogger instance to honor any subsequent advisory
163 logging requests for the metrics/instances in metriclist. If the
164 current logging state of the metrics/instances is mandatory (either
165 on or off) the new state will be set to maybe (effectively advisory
166 off). If the current state of the metrics/instances is already
167 advisory (either on or off) the state(s) for the metrics/instances
168 will remain as they are.
169 [ log ] advisory on interval metriclist
171 [ log ] advisory off metriclist
172 Advisory logging is only applicable if the last logging state spec‐
173 ified for a metric/instance was "mandatory maybe" (which permits
174 subsequent advisory logging control) or if the logging state is
175 already advisory. These two statements turn advisory logging on or
176 off (respectively) for the specified metrics/instances.
177 The interpretation for interval is as above for the mandatory case.
178 There is no continuation character required for commands that span
180 lines.
181 The word at may be used interchangeably with @.
183 A request to log all instances of a metric will supersede any prior
185 request to log either all or specific instances of a metric (if the
186 request specifies a permissible transition in the logging state). A
187 request to log specific instances of a metric when all instances of a
188 metric are already being logged is refused by pmlogger.
189
191 pmlc may have restricted access to and control over pmlogger(1) pro‐
192 cesses.
193 If a pmlogger(1) is unable to export its control information to the
195 local pmcd(1), then that pmlogger(1) cannot cannot be connected to nor
196 controlled by pmlc. In practice, this means the pmlogger(1) process
197 has to be owned by the user ``pcp'' and/or the group ``pcp''. If
198 pmlogger(1) is running on the host ``foo'' then use ``pminfo -f -h foo
199 pmcd.pmlogger'' to verify that the pmlogger(1) of interest is known to
200 pmcd(1), alternatively pmlogger(1) instances that are not reported from
201 the pmlc show loggers @foo command are not known to pmcd(1) on the host
202 ``foo''.
203 If pmlogger(1) is launched with a configuration file that contains an
205 [access] section, then pmlc will be unable to connect to that pmlog‐
206 ger(1) unless the access controls allow some access from the host where
207 pmlc is being run. Minimally this requires the enquire access to be
208 permitted in the pmlogger(1) access control section.
209 If pmlc is able to connect to the pmlogger(1) of interest, then the
211 following table summarizes the permissions needed to perform different
212 pmlc commands:
213 ┌──────────────────┬───────────────────────────────────────┐
215 │ pmlc command │ Required pmlogger access │
216 ├──────────────────┼───────────────────────────────────────┤
217 │show loggers │ Any │
218 │connect │ Any of enquire, advisory or mandatory │
219 │status │ Any of enquire, advisory or mandatory │
220 │query ... │ Any of enquire, advisory or mandatory │
221 │log advisory ... │ advisory │
222 │log mandatory ... │ mandatory │
223 │new volume │ mandatory │
224 └──────────────────┴───────────────────────────────────────┘
226 Environment variables with the prefix PCP_ are used to parameterize the
227 file and directory names used by PCP. On each installation, the file
228 /etc/pcp.conf contains the local values for these variables. The
229 $PCP_CONF variable may be used to specify an alternative configuration
230 file, as described in pcp.conf(5).
231
233 PCPIntro(1), pmcd(1), pmdumplog(1), pmlogger(1), pcp.conf(5),
234 pcp.env(5) and environ(7).
235
237 Most error or warning messages are self-explanatory. A message of the
238 form
239 Warning: unable to change logging state for...
240 followed by a list of metrics (and possibly instances) indicates that
241 pmlogger refused the request for the metrics (and instances) that
242 appear. Any metrics (and instances) that were specified but do not
243 appear in the message have had their logging state updated successfully
244 (no news is good news). Usually this warning results from requesting
245 advisory logging when a mandatory control is already in place, or
246 requesting logging for specific instances when all instances are
247 already being logged.
248
250 If all instances of a metric are being logged and a request is made to
251 log specific instances of the metric with the same state and frequency,
252 the request may appear to succeed, even though pmlogger has refused the
253 request. This is not normally a problem, as the required information
254 will still be placed into the log by pmlogger.
255 However in the case where the metric is to be logged once, the outcome
257 is not what might be expected. When pmlogger receives a request to log
258 a metric once, it places the current value(s) of the metric into the
259 log as soon as it can, regardless of whether the metric is already in
260 the log. This may be used to force values into the log. When a
261 request to log specific instances of a metric arrives and is refused
262 because all instances of the metric are already being logged, pmlogger
263 does not place values for the instances requested into the log. It
264 returns the current logging state for each instance requested to pmlc.
265 The requested and returned states are identical, so pmlc doesn't raise
266 an error as it should.
267 To ensure that only certain instances of a metric are being logged, one
269 should always turn off logging for all instances of the metric prior to
270 turning on logging for the specific instances required.
271Performance Co-Pilot PCP PMLC(1)