1PCP-DSTAT(1) General Commands Manual PCP-DSTAT(1)
2
6 pcp-dstat - versatile tool for generating system resource statistics
7
9 pcp [pcp options] dstat [-acdfghilmnpqrstvVy?] [-C cpus] [-D disks]
10 [-L DM devices] [-M MD devices] [-P partitions] [-I interrupts] [-N
11 interfaces] [-o output-file] [-S swap-devices] [--bits] [--bw]
12 [--color] [--float] [--integer] [--nocolor] [--noheaders] [--noupdate]
13 [--list] [--pidfile pid-file] [--plugin] [--all-plugins] [delay
14 [count]]
15
17 pcp-dstat is a general performance analysis tool allowing you to view
18 multiple system resources instantly, for example you can compare disk
19 usage in combination with interrupts from a disk controller, or compare
20 the network bandwidth numbers directly with the disk throughput (in the
21 same interval).
22 It also cleverly gives you the most detailed information in columns and
24 clearly indicates in what magnitude and unit the output is being dis‐
25 played. Less confusion, fewer mistakes, more efficient.
26 The delay is the delay in seconds between each update, and the count is
28 the number of updates to display before exiting. The default delay is
29 1 second and count is unspecified (run until interrupted or end of ar‐
30 chive is reached).
31 This latest generation of Dstat, pcp-dstat, allows for analysis of his‐
33 torical performance data (in the PCP archive format created by pmlog‐
34 ger(1)), as well as distributed systems analysis of live performance
35 data from remote hosts running the pmcd(1) process.
36 Additionally, this version introduces configuration files similar to
38 pmrep.conf(5) from the pmrep(1) utility. The original Dstat notion of
39 ``plugins'' is replaced by use of named metrics in a Performance Metric
40 Name Space (PMNS(5)) supplied by Performance Metric Domain Agents
41 (PMDAs). Metrics and other formatting information is now specified as
42 plugin configuration files. This new style of plugin is either built-
43 in (time-related reporting only), or sourced from the system-wide loca‐
44 tion ($PCP_ETC_DIR/dstat) and/or sourced from an individual users set
45 of personal plugins ($HOME/pcp/dstat).
46 The list of all available plugins can be seen using the --list dstat
48 command line option.
49
51 When invoked via the pcp(1) command, the -h/--host, -a/--archive,
52 -O/--origin, -Z/--timezone and several other pcp options become indi‐
53 rectly available; refer to PCPIntro(1) for a complete description of
54 these options.
55 The additional command line options available for pcp-dstat are:
57 --list list all available plugin names
59 --plugin-name
61 enable any plugin by name
62 -a, --all
64 equals -cdngy (default plugin set)
65 -c, --cpu
67 enable CPU stats (system, user, idle, wait); for more CPU related
68 stats also see --cpu-adv and --cpu-use
69 -C 0,3,total
71 include CPU0, CPU3 and total (when using -c/--cpu); use all to
72 show all CPUs
73 -d, --disk
75 enable disk stats (read, write); for more disk related stats look
76 into the other --disk plugins
77 -D total,hda
79 include total and hda (when using -d/--disk or --disk-tps plugin)
80 --dm, --device-mapper
82 enable device mapper stats (read, write); for more device-mapper
83 related stats look into the other --dm plugins
84 --L total,root,home
86 include total, root and home (when using --dm/--device-mapper or
87 --dm-tps plugin)
88 --md, --multi-device
90 enable multi-device driver stats (read, write); for more multi-
91 device driver related stats look into the other --md plugins
92 --M total,md-0
94 include total and md-0 (when using --md/--multi-device or --md-tps
95 plugin)
96 --part, --partition
98 enable disk partition stats (read, write); for more partition
99 related stats look into the other --part plugins
100 --P total,sda2
102 include total and sda2 (when using --part/--partition or --part-
103 tps plugin)
104 -g, --page
106 enable page stats (page in, page out)
107 -i, --int
109 enable interrupt stats
110 -I 5,10
112 include interrupt 5 and 10 (when using -i/--int)
113 -l, --load
115 enable load average stats (1 min, 5 mins, 15 mins)
116 -m, --mem
118 enable memory stats (used, buffers, cache, free); for more memory
119 related stats also try --mem-adv and --swap
120 -n, --net
122 enable network stats (receive, send)
123 -N eth1,total
125 include eth1 and total (when using -n/--net)
126 --net-packets
128 show the number of packets received and transmitted
129 -p, --proc
131 enable process stats (runnable, uninterruptible, new)
132 --proc-count
134 show total number of processes
135 -r, --io
137 enable I/O request stats (read, write requests)
138 -s, --swap
140 enable swap stats (used, free)
141 -S swap1,total
143 include swap1 and total (when using -s/--swap)
144 --snooze show time spent between updates in seconds
146 -t, --time
148 enable time/date output (try --time-adv for millisecond preci‐
149 sion)
150 -T, --epoch
152 enable time counter (seconds since epoch, or millisecond preci‐
153 sion from the --epoch-adv plugin)
154 -y, --sys
156 enable system stats (interrupts, context switches)
157 --aio enable aio stats (asynchronous I/O)
159 --cpu-adv
161 enable advanced CPU stats
162 --cpu-use
164 enable only CPU usage stats
165 --disk-avgqu
167 average queue length of the requests that were issued to the
168 device
169 --disk-avgrq
171 average size (in sectors) of the requests that were issued to
172 the device
173 --disk-svctm
175 average service time (in milliseconds) for I/O requests that
176 were issued to the device
177 --disk-tps
179 number of transfers per second that were issued to the device
180 --disk-util
182 percentage of CPU time during which I/O requests were issued to
183 the device (bandwidth utilization for the device)
184 --disk-wait
186 average time (in milliseconds) for I/O requests issued to the
187 device to be served
188 --fs, --filesystem
190 enable filesystem stats (open files, inodes)
191 --freespace
193 per filesystem used and available space
194 --ipc enable ipc stats (message queue, semaphores, shared memory)
196 --lock enable file lock stats (posix, flock, read, write)
198 --mem-adv
200 enable advanced memory stats
201 --raw enable raw stats (raw sockets)
203 --socket
205 enable socket stats (total, tcp, udp, raw, ip-fragments)
206 --tcp enable tcp stats (listen, established, syn, time_wait, close)
208 --udp enable udp stats (listen, active)
210 --unix enable unix stats (datagram, stream, listen, active)
212 --utmp shows login information from utmp(5)
214 --vm enable virtual memory stats (hard pagefaults, soft pagefaults,
216 allocated, free)
217 --vm-adv
219 enable advance virtual memory stats (steal, scanK, scanD, pgoru,
220 astll)
221 --nfs3 show NFS v3 client operations
223 --nfs3-ops
225 show extended NFS v3 client operations
226 --nfsd3
228 show NFS v3 server operations
229 --nfsd3-ops
231 show extended NFS v3 server operations
232 --nfsd4-ops
234 show extended NFS v4 server operations
235 --nfsstat4
237 show NFS v4 stats
238 --rpc show remote procedure call (RPC) client calls stats
240 --rpcd show remote procedure call (RPC) server calls stats
242 -f, --full
244 expand -C, -D, -I, -N and -S discovery lists
245 -v, --vmstat
247 equals -pmgdsc -D total
248 --bits force bits for values expressed in bytes
250 --float
252 force float values on screen (mutually exclusive with --integer)
253 --integer
255 force integer values on screen (mutually exclusive with --float)
256 --bw, --blackonwhite
258 change colors for white background terminal
259 --nocolor
261 disable colors
262 --noheaders
264 disable repetitive headers
265 --noupdate
267 disable intermediate updates when delay greater than 1.
268 -o file, --output=file
270 write CSV (Comma-Separated Value) format output to a file.
271 -p file, --pidfile=file
273 write the process identifier to a given file.
274
276 Some pcp-dstat configuration files require the installation of optional
277 Performance Metric Domain Agents, above and beyond the default
278 installed set.
279 --innodb-buffer
281 show innodb buffer stats (needs the MySQL PMDA)
282 --innodb-io
284 show innodb I/O stats (needs the MySQL PMDA)
285 --innodb-ops
287 show innodb operations counters (needs the MySQL PMDA)
288 --lustre
290 show lustre I/O throughput (needs the Lustre PMDA)
291 --memcache-hits
293 show the number of hits and misses from memcache
294 --mysql5-cmds
296 show the MySQL5 command stats (needs the MySQL PMDA)
297 --mysql5-conn
299 show the MySQL5 connection stats (needs the MySQL PMDA)
300 --mysql5-innodb
302 show the MySQL5 innodb stats (needs the MySQL PMDA)
303 --mysql5-io
305 show the MySQL5 I/O stats (needs the MySQL PMDA)
306 --mysql5-keys
308 show the MySQL5 keys stats (needs the MySQL PMDA)
309 --mysql-io
311 show the MySQL I/O stats (needs the MySQL PMDA)
312 --mysql-keys
314 show the MySQL keys stats (needs the MySQL PMDA)
315 --postfix
317 show postfix queue sizes (needs the Postfix PMDA)
318 --redis
320 show Redis stats (needs the Redis PMDA)
321 --sendmail
323 show sendmail queue size (needs the Sendmail PMDA)
324
326 Anyone can create additional, custom pcp-dstat plugin configuration
327 files, for any metrics - the list of available metrics can be produced
328 by either the pminfo(1) or pmprobe(1) command.
329 The following do not yet have metrics backing them, but have been
331 included from the original Dstat utility. Please contact
332 <pcp@groups.io> if you need or implement any of these, and we'll work
333 with you to get them included.
334 --battery
336 battery in percentage (needs an ACPI PMDA)
337 --battery-remain
339 battery remaining in hours, minutes (needs an ACPI PMDA)
340 --cpufreq
342 CPU frequency in percentage (needs an ACPI PMDA)
343 --dbus
345 number of dbus connections (needs a python-dbus PMDA)
346 --fan
348 fan speed (needs an ACPI PMDA)
349 --gpfs
351 GPFS read/write I/O (needs mmpmon and a GPFS PMDA)
352 --gpfs-ops
354 GPFS filesystem operations (needs mmpmon and a GPFS PMDA)
355 --md-status
357 show software raid (MD driver) progress and speed (needs new
358 disk.md metrics)
359 --power
361 show power usage (needs an ACPI PMDA)
362 --qmail
364 show qmail queue sizes (needs qmail)
365 --squid
367 show squid usage statistics (needs a Squid PMDA)
368 --thermal
370 system temperature sensors (needs an ACPI PMDA)
371 --vm-cpu
373 show VMware CPU stats from hypervisor (needs a VMware PMDA)
374 --vm-mem
376 show VMware memory stats from hypervisor (needs a VMware PMDA)
377 --vm-mem-adv
379 show advanced VMware memory stats from hypervisor (needs a VMware
380 PMDA)
381 --vmk-hba
383 show VMware ESX kernel vmhba stats (needs a VMware PMDA)
384 --vmk-int
386 show VMware ESX kernel interrupt stats (needs a VMware PMDA)
387 --vmk-nic
389 show VMware ESX kernel port stats (needs a VMware PMDA)
390 --vz-cpu
392 show CPU usage per OpenVZ guest (needs an OpenVZ PMDA)
393 --vz-io
395 show I/O usage per OpenVZ guest (needs an OpenVZ PMDA)
396 --vz-ubc
398 show OpenVZ user beancounters (needs an OpenVZ PMDA)
399 --wifi
401 wireless link quality and signal to noise ratio (needs Linux PMDA
402 metrics)
403 --zfs-arc
405 show ZFS arc stats (needs a Linux ZFS PMDA)
406 --zfs-l2arc
408 show ZFS l2arc stats (needs a Linux ZFS PMDA)
409 --zfs-zil
411 show ZFS zil stats (needs a Linux ZFS PMDA)
412
414 When invoking pcp-dstat with a delay greater than 1 second and without
415 the --noupdate option, it will show intermediate updates, i.e., the
416 first time a 1 second average, the second update a 2 second average,
417 etc. until the delay has been reached.
418 So in case you specified a delay of 10, the 9 intermediate updates are
420 NOT snapshots, they are averages over the time that passed since the
421 last final update. The end result is that you get a 10 second average
422 on a new line, just like with vmstat(1).
423
425 Using pcp-dstat to relate disk-throughput with network-usage (eth0),
426 total CPU-usage and system counters:
427 $ pcp dstat -dnyc -N eth0 -C total -f 5
429 Using the time plugin together with cpu, net, disk, system, load and
431 proc plugins:
432 $ pcp dstat -tcndylp
434 This is identical to:
436 $ pcp dstat --time --cpu --net --disk --sys --load --proc
438 Using pcp-dstat to report 10 samples from metrics recorded in a PCP ar‐
440 chive 20180729 from 2:30 AM:
441 $ pcp --origin '@02:30' -a 20180729 dstat --time --cpu-adv --sys 1 10
442 Examine the same metrics live from a remote host:
444 $ pcp --host www.acme.com dstat --time --cpu-adv --sys 1 10
446
449 $HOME/.pcp/dstat/
450 private per-user configuration files
451 $PCP_SYSCONF_DIR/pcp/dstat/
453 system-wide configuration files
454
456 Internal plugins behaviour can be changed through environment vari‐
457 ables.
458 DSTAT_TIMEFMT
460 strftime(3) format string for reporting time (see --time)
461
463 Environment variables with the prefix PCP_ are used to parameterize the
464 file and directory names used by PCP. On each installation, the file
465 /etc/pcp.conf contains the local values for these variables. The
466 $PCP_CONF variable may be used to specify an alternative configuration
467 file, as described in pcp.conf(5).
468 For environment variables affecting PCP tools, see pmGetOptions(3).
470
472 The Dstat utility was initially written by Dag Wieers <dag@wieers.com>.
473 The Dstat homepage is at http://dag.wieers.com/home-made/dstat/.
475 This manpage was initially written by Andrew Pollock <apol‐
477 lock@debian.org> for the Debian GNU/Linux system.
478 The pcp-dstat utility is written and maintained by the PCP developers
480 <pcp@groups.io>.
481 The PCP homepage is at https://pcp.io/.
483
485 PCPIntro(1), pcp(1), pmcd(1), pminfo(1), pmlogger(1), pmprobe(1),
486 pmrep(1), vmstat(1), pmGetOptions(3), strftime(3), PMNS(5),
487 pcp.conf(5), pmrep.conf(5) and utmp(5).
488Performance Co-Pilot PCP PCP-DSTAT(1)