1PCP2ZABBIX(1) General Commands Manual PCP2ZABBIX(1)
2
6 pcp2zabbix - pcp-to-zabbix metrics exporter
7
9 pcp2zabbix [-5CGHIjlLnrRvV?] [-4 action] [-8|-9 limit] [-a archive]
10 [-A align] [--archive-folio folio] [-b|-B space-scale] [-c config]
11 [--container container] [--daemonize] [-e derived] [-E send-interval]
12 [-g server] [-h host] [-i instances] [-J rank] [-K spec] [-N predicate]
13 [-O origin] [-p port] [-P|-0 precision] [-q|-Q count-scale] [-s sam‐
14 ples] [-S starttime] [-t interval] [-T endtime] [-x prefix] [-X host‐
15 name] [-y|-Y time-scale] metricspec [...]
16
18 pcp2zabbix is a customizable performance metrics exporter tool from PCP
19 to Zabbix. Any available performance metric, live or archived, system
20 and/or application, can be selected for exporting using either command
21 line arguments or a configuration file.
22 pcp2zabbix is a close relative of pmrep(1). Refer to pmrep(1) for the
24 metricspec description accepted on pcp2zabbix command line. See pm‐
25 rep.conf(5) for description of the pcp2zabbix.conf configuration file
26 syntax. This page describes pcp2zabbix specific options and configura‐
27 tion file differences with pmrep.conf(5). pmrep(1) also lists some us‐
28 age examples of which most are applicable with pcp2zabbix as well.
29 Only the command line options listed on this page are supported, other
31 options available for pmrep(1) are not supported.
32 Options via environment values (see pmGetOptions(3)) override the cor‐
34 responding built-in default values (if any). Configuration file op‐
35 tions override the corresponding environment variables (if any). Com‐
36 mand line options override the corresponding configuration file options
37 (if any).
38
40 pcp2zabbix uses a configuration file with syntax described in pm‐
41 rep.conf(5). The following options are common with pmrep.conf: ver‐
42 sion, source, speclocal, derived, header, globals, samples, interval,
43 type, type_prefer, ignore_incompat, names_change, instances, live_fil‐
44 ter, rank, limit_filter, limit_filter_force, invert_filter, predicate,
45 omit_flat, precision, precision_force, count_scale, count_scale_force,
46 space_scale, space_scale_force, time_scale, time_scale_force. The rest
47 of the pmrep.conf options are recognized but ignored for compatibility.
48 pcp2zabbix specific options
50 zabbix_server (string)
51 Specify the destination Zabbix server hostname or IP address. If a
52 host is monitored by a proxy, proxy hostname or IP address should
53 be used instead. Corresponding command line option is -g. De‐
54 faults to localhost.
55 zabbix_port (integer)
57 Specify the port number of server trapper running on the server.
58 Corresponding command line option is -p. Defaults to 10051.
59 zabbix_host (string)
61 Specify agent hostname as registered in Zabbix frontend. Host IP
62 address and DNS name will not work. Corresponding command line op‐
63 tion is -X. Defaults to the metrics source host.
64 zabbix_interval (string)
66 Indicates the interval to send metrics to the Zabbix server. This
67 can be longer than the reporting interval to minimize the overhead
68 when communicating with the server (as each send creates a new con‐
69 nection). Follows the time syntax described in PCPIntro(1). Cor‐
70 responding command line option is -E. Defaults to the reporting
71 interval. Zabbix tools send at most 250 metrics at a time. Ig‐
72 nored when reading metrics from PCP archives, in which case metrics
73 will be sent roughly at 250 metric batches.
74 zabbix_prefix (string)
76 Specify the prefix for metrics sent to the Zabbix server. Corre‐
77 sponding command line option is -x. Defaults to pcp..
78 zabbix_lld (boolean)
80 Specify whether to enable Zabbix low-level discovery. Correspond‐
81 ing command line option is -l. Defaults to no.
82
84 The available command line options are:
85 -0 precision, --precision-force=precision
87 Like -P but this option will override per-metric specifications.
88 -4 action, --names-change=action
90 Specify which action to take on receiving a metric names change
91 event during sampling. These events occur when a PMDA discovers
92 new metrics sometime after starting up, and informs running client
93 tools like pcp2zabbix. Valid values for action are update (re‐
94 fresh metrics being sampled), ignore (do nothing - the default be‐
95 haviour) and abort (exit the program if such an event occurs).
96 -5, --ignore-unknown
98 Silently ignore any metric name that cannot be resolved. At least
99 one metric must be found for the tool to start.
100 -8 limit, --limit-filter=limit
102 Limit results to instances with values above/below limit. A posi‐
103 tive integer will include instances with values at or above the
104 limit in reporting. A negative integer will include instances
105 with values at or below the limit in reporting. A value of zero
106 performs no limit filtering. This option will not override possi‐
107 ble per-metric specifications. See also -J and -N.
108 -9 limit, --limit-filter-force=limit
110 Like -8 but this option will override per-metric specifications.
111 -a archive, --archive=archive
113 Performance metric values are retrieved from the set of Perfor‐
114 mance Co-Pilot (PCP) archive log files identified by the archive
115 argument, which is a comma-separated list of names, each of which
116 may be the base name of an archive or the name of a directory con‐
117 taining one or more archives.
118 -A align, --align=align
120 Force the initial sample to be aligned on the boundary of a natu‐
121 ral time unit align. Refer to PCPIntro(1) for a complete descrip‐
122 tion of the syntax for align.
123 --archive-folio=folio
125 Read metric source archives from the PCP archive folio created by
126 tools like pmchart(1) or, less often, manually with mkaf(1).
127 -b scale, --space-scale=scale
129 Unit/scale for space (byte) metrics, possible values include
130 bytes, Kbytes, KB, Mbytes, MB, and so forth. This option will not
131 override possible per-metric specifications. See also pmParseU‐
132 nitsStr(3).
133 -B scale, --space-scale-force=scale
135 Like -b but this option will override per-metric specifications.
136 -c config, --config=config
138 Specify the config file or directory to use. In case config is a
139 directory all files in it ending .conf will be included. The de‐
140 fault is the first found of: ./pcp2zabbix.conf, $HOME/.pcp2zab‐
141 bix.conf, $HOME/pcp/pcp2zabbix.conf, and $PCP_SYSCONF_DIR/pcp2zab‐
142 bix.conf. For details, see the above section and pmrep.conf(5).
143 --container=container
145 Fetch performance metrics from the specified container, either lo‐
146 cal or remote (see -h).
147 -C, --check
149 Exit before reporting any values, but after parsing the configura‐
150 tion and metrics and printing possible headers.
151 --daemonize
153 Daemonize on startup.
154 -e derived, --derived=derived
156 Specify derived performance metrics. If derived starts with a
157 slash (``/'') or with a dot (``.'') it will be interpreted as a
158 PCP derived metrics configuration file, otherwise it will be in‐
159 terpreted as comma- or semicolon-separated derived metric expres‐
160 sions. For complete description of derived metrics and PCP de‐
161 rived metrics configuration files see pmLoadDerivedConfig(3) and
162 pmRegisterDerived(3). Alternatively, using pmrep.conf(5) configu‐
163 ration syntax allows defining derived metrics as part of metric‐
164 sets.
165 -E interval, --zabbix-interval=interval
167 Specify the interval to send metrics to the server. See zab‐
168 bix_interval description above for details.
169 -g server, --zabbix-server=server
171 Destination Zabbix server.
172 -G, --no-globals
174 Do not include global metrics in reporting (see pmrep.conf(5)).
175 -h host, --host=host
177 Fetch performance metrics from pmcd(1) on host, rather than from
178 the default localhost.
179 -H, --no-header
181 Do not print any headers.
182 -i instances, --instances=instances
184 Retrieve and report only the specified metric instances. By de‐
185 fault all instances, present and future, are reported.
186 Refer to pmrep(1) for complete description of this option.
188 -I, --ignore-incompat
190 Ignore incompatible metrics. By default incompatible metrics
191 (that is, their type is unsupported or they cannot be scaled as
192 requested) will cause pcp2zabbix to terminate with an error mes‐
193 sage. With this option all incompatible metrics are silently
194 omitted from reporting. This may be especially useful when re‐
195 questing non-leaf nodes of the PMNS tree for reporting.
196 -l, --zabbix-lld
198 Add low-level discovery (LLD) metrics. This allows PCP metric in‐
199 stance domains to be sent to Zabbix, so that prototype items/trig‐
200 gers/graphs may be automatically constructed by Zabbix. For each
201 PCP metric with an instance domain, the key
202 pcp.discovery[pcp.met.ric] is sent periodically. This symbol
203 should be added to a discovery rule, and will define the
204 {#PCP.MET.RIC} LLD macro. This macro can in turn be used to de‐
205 fine prototype items/triggers/graphs. The names are adjusted if
206 -x was given to override the pcp. prefix.
207 -j, --live-filter
209 Perform instance live filtering. This allows capturing all named
210 instances even if processes are restarted at some point (unlike
211 without live filtering). Performing live filtering over a huge
212 number of instances will add some internal overhead so a bit of
213 user caution is advised. See also -n.
214 -J rank, --rank=rank
216 Limit results to highest/lowest ranked instances of set-valued
217 metrics. A positive integer will include highest valued instances
218 in reporting. A negative integer will include lowest valued in‐
219 stances in reporting. A value of zero performs no ranking. Rank‐
220 ing does not imply sorting, see -6. See also -8.
221 -K spec, --spec-local=spec
223 When fetching metrics from a local context (see -L), the -K option
224 may be used to control the DSO PMDAs that should be made accessi‐
225 ble. The spec argument conforms to the syntax described in pm‐
226 SpecLocalPMDA(3). More than one -K option may be used.
227 -L, --local-PMDA
229 Use a local context to collect metrics from DSO PMDAs on the local
230 host without PMCD. See also -K.
231 -n, --invert-filter
233 Perform ranking before live filtering. By default instance live
234 filtering (when requested, see -j) happens before instance ranking
235 (when requested, see -J). With this option the logic is inverted
236 and ranking happens before live filtering.
237 -N predicate, --predicate=predicate
239 Specify a comma-separated list of predicate filter reference met‐
240 rics. By default ranking (see -J) happens for each metric indi‐
241 vidually. With predicates, ranking is done only for the specified
242 predicate metrics. When reporting, rest of the metrics sharing
243 the same instance domain (see PCPIntro(1)) as the predicate will
244 include only the highest/lowest ranking instances of the corre‐
245 sponding predicate. Ranking does not imply sorting, see -6.
246 So for example, using proc.memory.rss (resident memory size of
248 process) as the predicate metric together with proc.io.total_bytes
249 and mem.util.used as metrics to be reported, only the processes
250 using most/least (as per -J) memory will be included when report‐
251 ing total bytes written by processes. Since mem.util.used is a
252 single-valued metric (thus not sharing the same instance domain as
253 the process related metrics), it will be reported as usual.
254 -O origin, --origin=origin
256 When reporting archived metrics, start reporting at origin within
257 the time window (see -S and -T). Refer to PCPIntro(1) for a com‐
258 plete description of the syntax for origin.
259 -p port, --zabbix-port=port
261 Zabbix server port.
262 -P precision, --precision=precision
264 Use precision for numeric non-integer output values. The default
265 is to use 3 decimal places (when applicable). This option will
266 not override possible per-metric specifications.
267 -q scale, --count-scale=scale
269 Unit/scale for count metrics, possible values include count x
270 10^-1, count, count x 10, count x 10^2, and so forth from 10^-8 to
271 10^7. (These values are currently space-sensitive.) This option
272 will not override possible per-metric specifications. See also
273 pmParseUnitsStr(3).
274 -Q scale, --count-scale-force=scale
276 Like -q but this option will override per-metric specifications.
277 -r, --raw
279 Output raw metric values, do not convert cumulative counters to
280 rates. This option will override possible per-metric specifica‐
281 tions.
282 -R, --raw-prefer
284 Like -r but this option will not override per-metric specifica‐
285 tions.
286 -s samples, --samples=samples
288 The argument samples defines the number of samples to be retrieved
289 and reported. If samples is 0 or -s is not specified, pcp2zabbix
290 will sample and report continuously (in real time mode) or until
291 the end of the set of PCP archives (in archive mode). See also
292 -T.
293 -S starttime, --start=starttime
295 When reporting archived metrics, the report will be restricted to
296 those records logged at or after starttime. Refer to PCPIntro(1)
297 for a complete description of the syntax for starttime.
298 -t interval, --interval=interval
300 Set the reporting interval to something other than the default 1
301 second. The interval argument follows the syntax described in
302 PCPIntro(1), and in the simplest form may be an unsigned integer
303 (the implied units in this case are seconds). See also the -T op‐
304 tion.
305 -T endtime, --finish=endtime
307 When reporting archived metrics, the report will be restricted to
308 those records logged before or at endtime. Refer to PCPIntro(1)
309 for a complete description of the syntax for endtime.
310 When used to define the runtime before pcp2zabbix will exit, if no
312 samples is given (see -s) then the number of reported samples de‐
313 pends on interval (see -t). If samples is given then interval
314 will be adjusted to allow reporting of samples during runtime. In
315 case all of -T, -s, and -t are given, endtime determines the actu‐
316 al time pcp2zabbix will run.
317 -v, --omit-flat
319 Report only set-valued metrics with instances (e.g. disk.dev.read)
320 and omit single-valued ``flat'' metrics without instances (e.g.
321 kernel.all.sysfork). See -i and -I.
322 -V, --version
324 Display version number and exit.
325 -x prefix, --zabbix-prefix=prefix
327 Metrics prefix for measurements sent to the server.
328 -X hostname, --zabbix-host=hostname
330 Agent hostname as registered in Zabbix frontend.
331 -y scale, --time-scale=scale
333 Unit/scale for time metrics, possible values include nanosec, ns,
334 microsec, us, millisec, ms, and so forth up to hour, hr. This op‐
335 tion will not override possible per-metric specifications. See
336 also pmParseUnitsStr(3).
337 -Y scale, --time-scale-force=scale
339 Like -y but this option will override per-metric specifications.
340 -?, --help
342 Display usage message and exit.
343
345 pcp2zabbix.conf
346 pcp2zabbix configuration file (see -c)
347 $PCP_SYSCONF_DIR/pmrep/*.conf
349 system provided default pmrep configuration files
350
352 Environment variables with the prefix PCP_ are used to parameterize the
353 file and directory names used by PCP. On each installation, the file
354 /etc/pcp.conf contains the local values for these variables. The
355 $PCP_CONF variable may be used to specify an alternative configuration
356 file, as described in pcp.conf(5).
357 For environment variables affecting PCP tools, see pmGetOptions(3).
359
361 PCPIntro(1), mkaf(1), pcp(1), pcp2elasticsearch(1), pcp2graphite(1),
362 pcp2influxdb(1), pcp2json(1), pcp2spark(1), pcp2xlsx(1), pcp2xml(1),
363 pmcd(1), pminfo(1), pmrep(1), pmGetOptions(3), pmLoadDerivedConfig(3),
364 pmParseUnitsStr(3), pmRegisterDerived(3), pmSpecLocalPMDA(3), zbx‐
365 pcp(3), LOGARCHIVE(5), pcp.conf(5), pmrep.conf(5) and PMNS(5).
366Performance Co-Pilot PCP PCP2ZABBIX(1)