1PCP-ATOP(1) General Commands Manual PCP-ATOP(1)
2
6 pmatop, pcp-atop - Advanced System and Process Monitor
7
9 Interactive Usage:
10 pcp atop [-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y] [-C|-M|-D|-N|-A] [-afFG1xR]
12 [-L linelen] [-Plabel[,label]...] [ interval [ samples ]]
13 Writing and reading raw logfiles:
15 pcp atop -w rawfile [-a] [-S] [ interval [ samples ]]
17 pcp atop -r [ rawfile ] [-b hh:mm ] [-e hh:mm ]
18 [-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y] [-C|-M|-D|-N|-A] [-fFG1xR] [-L line‐
19 len] [-Plabel[,label]...]
20
22 The program pcp-atop is an interactive monitor to view various aspects
23 of load on a system. It shows the occupation of the most critical
24 hardware resources (from a performance point of view) on system level,
25 i.e. cpu, memory, disk and network.
26 It also shows which processes are responsible for the indicated load
27 with respect to cpu and memory load on process level. Disk load is
28 shown per process if "storage accounting" is active in the kernel.
29 Every interval (default: 10 seconds) information is shown about the
31 resource occupation on system level (cpu, memory, disks and network
32 layers), followed by a list of processes which have been active during
33 the last interval (note that all processes that were unchanged during
34 the last interval are not shown, unless the key 'a' has been pressed).
35 If the list of active processes does not entirely fit on the screen,
36 only the top of the list is shown (sorted in order of activity).
37 The intervals are repeated till the number of samples (specified as
38 command argument) is reached, or till the key 'q' is pressed in inter‐
39 active mode.
40 When invoked via the pcp(1) command, the PCPIntro(1) options -h/--host,
42 -a/--archive, -O/--origin, -s/--samples, -t/--interval, -Z/--timezone
43 and several other options become indirectly available.
44 When pcp-atop is started, it checks whether the standard output channel
46 is connected to a screen, or to a file/pipe. In the first case it pro‐
47 duces screen control codes (via the ncurses library) and behaves inter‐
48 actively; in the second case it produces flat ASCII-output.
49 In interactive mode, the output of pcp-atop scales dynamically to the
51 current dimensions of the screen/window.
52 If the window is resized horizontally, columns will be added or removed
53 automatically. For this purpose, every column has a particular weight.
54 The columns with the highest weights that fit within the current width
55 will be shown.
56 If the window is resized vertically, lines of the process/thread list
57 will be added or removed automatically.
58 Furthermore in interactive mode the output of pcp-atop can be con‐
60 trolled by pressing particular keys. However it is also possible to
61 specify such key as flag on the command line. In that case pcp-atop
62 switches to the indicated mode on beforehand; this mode can be modified
63 again interactively. Specifying such key as flag is especially useful
64 when running pcp-atop with output to a pipe or file (non-interac‐
65 tively). These flags are the same as the keys that can be pressed in
66 interactive mode (see section INTERACTIVE COMMANDS).
67 Additional flags are available to support storage of pcp-atop data in
68 PCP archive format (see section PCP DATA STORAGE).
69
71 For the resource consumption on system level, pcp-atop uses colors to
72 indicate that a critical occupation percentage has been (almost)
73 reached. A critical occupation percentage means that is likely that
74 this load causes a noticeable negative performance influence for appli‐
75 cations using this resource. The critical percentage depends on the
76 type of resource: e.g. the performance influence of a disk with a busy
77 percentage of 80% might be more noticeable for applications/user than a
78 CPU with a busy percentage of 90%.
79 Currently pcp-atop uses the following default values to calculate a
80 weighted percentage per resource:
81 Processor
83 A busy percentage of 90% or higher is considered `critical'.
84 Disk
86 A busy percentage of 70% or higher is considered `critical'.
87 Network
89 A busy percentage of 90% or higher for the load of an interface is
90 considered `critical'.
91 Memory
93 An occupation percentage of 90% is considered `critical'. Notice
94 that this occupation percentage is the accumulated memory consump‐
95 tion of the kernel (including slab) and all processes; the memory
96 for the page cache (`cache' and `buff' in the MEM-line) and the
97 reclaimable part of the slab (`slrec`) is not implied!
98 If the number of pages swapped out (`swout' in the PAG-line) is
99 larger than 10 per second, the memory resource is considered
100 `critical'. A value of at least 1 per second is considered
101 `almost critical'.
102 If the committed virtual memory exceeds the limit (`vmcom' and
103 `vmlim' in the SWP-line), the SWP-line is colored due to overcom‐
104 mitting the system.
105 Swap
107 An occupation percentage of 80% is considered `critical' because
108 swap space might be completely exhausted in the near future; it is
109 not critical from a performance point-of-view.
110 These default values can be modified in the configuration file (see
112 separate man-page of pcp-atoprc).
113 When a resource exceeds its critical occupation percentage, the con‐
115 cerning values in the screen line are colored red by default.
116 When a resource exceeded (default) 80% of its critical percentage (so
117 it is almost critical), the concerning values in the screen line are
118 colored cyan by default. This `almost critical percentage' (one value
119 for all resources) can be modified in the configuration file (see sepa‐
120 rate man-page of pcp-atoprc).
121 The default colors red and cyan can be modified in the configuration
122 file as well (see separate man-page of pcp-atoprc).
123 With the key 'x' (or flag -x), the use of colors can be suppressed.
125
127 When running pcp-atop interactively (no output redirection), keys can
128 be pressed to control the output. In general, lower case keys can be
129 used to show other information for the active processes and upper case
130 keys can be used to influence the sort order of the active
131 process/thread list.
132 g Show generic output (default).
134 Per process the following fields are shown in case of a window-
136 width of 80 positions: process-id, cpu consumption during the last
137 interval in system and user mode, the virtual and resident memory
138 growth of the process.
139 The subsequent columns depend on the used kernel:
141 When the kernel supports "storage accounting" (>= 2.6.20), the
142 data transfer for read/write on disk, the status and exit code are
143 shown for each process. When the kernel does not support "storage
144 accounting", the username, number of threads in the thread group,
145 the status and exit code are shown.
146 The last columns contain the state, the occupation percentage for
147 the chosen resource (default: cpu) and the process name.
148 When more than 80 positions are available, other information is
150 added.
151 m Show memory related output.
153 Per process the following fields are shown in case of a window-
155 width of 80 positions: process-id, minor and major memory faults,
156 size of virtual shared text, total virtual process size, total
157 resident process size, virtual and resident growth during last
158 interval, memory occupation percentage and process name.
159 When more than 80 positions are available, other information is
161 added.
162 d Show disk-related output.
164 When "storage accounting" is active in the kernel, the following
166 fields are shown: process-id, amount of data read from disk,
167 amount of data written to disk, amount of data that was written
168 but has been withdrawn again (WCANCL), disk occupation percentage
169 and process name.
170 s Show scheduling characteristics.
172 Per process the following fields are shown in case of a window-
174 width of 80 positions: process-id, number of threads in state
175 'running' (R), number of threads in state 'interruptible sleeping'
176 (S), number of threads in state 'uninterruptible sleeping' (D),
177 scheduling policy (normal timesharing, realtime round-robin, real‐
178 time fifo), nice value, priority, realtime priority, current pro‐
179 cessor, status, exit code, state, the occupation percentage for
180 the chosen resource and the process name.
181 When more than 80 positions are available, other information is
183 added.
184 v Show various process characteristics.
186 Per process the following fields are shown in case of a window-
188 width of 80 positions: process-id, user name and group, start date
189 and time, status (e.g. exit code if the process has finished),
190 state, the occupation percentage for the chosen resource and the
191 process name.
192 When more than 80 positions are available, other information is
194 added.
195 c Show the command line of the process.
197 Per process the following fields are shown: process-id, the occu‐
199 pation percentage for the chosen resource and the command line
200 including arguments.
201 o Show the user-defined line of the process.
203 In the configuration file the keyword ownprocline can be specified
205 with the description of a user-defined output-line.
206 Refer to the man-page of pcp-atoprc for a detailed description.
207 y Show the individual threads within a process (toggle).
209 Single-threaded processes are still shown as one line.
211 For multi-threaded processes, one line represents the process
212 while additional lines show the activity per individual thread (in
213 a different color). Depending on the option 'a' (all or active
214 toggle), all threads are shown or only the threads that were
215 active during the last interval.
216 Whether this key is active or not can be seen in the header line.
217 u Show the process activity accumulated per user.
219 Per user the following fields are shown: number of processes
221 active or terminated during last interval (or in total if combined
222 with command `a'), accumulated cpu consumption during last inter‐
223 val in system and user mode, the current virtual and resident mem‐
224 ory space consumed by active processes (or all processes of the
225 user if combined with command `a').
226 When "storage accounting" is active in the kernel, the accumulated
227 read and write throughput on disk is shown. When the kernel mod‐
228 ule `netatop' has been installed, the number of received and sent
229 network packets are shown.
230 The last columns contain the accumulated occupation percentage for
231 the chosen resource (default: cpu) and the user name.
232 p Show the process activity accumulated per program (i.e. process
234 name).
235 Per program the following fields are shown: number of processes
237 active or terminated during last interval (or in total if combined
238 with command `a'), accumulated cpu consumption during last inter‐
239 val in system and user mode, the current virtual and resident mem‐
240 ory space consumed by active processes (or all processes of the
241 user if combined with command `a').
242 When "storage accounting" is active in the kernel, the accumulated
243 read and write throughput on disk is shown. When the kernel mod‐
244 ule `netatop' has been installed, the number of received and sent
245 network packets are shown.
246 The last columns contain the accumulated occupation percentage for
247 the chosen resource (default: cpu) and the program name.
248 C Sort the current list in the order of cpu consumption (default).
250 The one-but-last column changes to ``CPU''.
251 M Sort the current list in the order of resident memory consumption.
253 The one-but-last column changes to ``MEM''.
254 D Sort the current list in the order of disk accesses issued. The
256 one-but-last column changes to ``DSK''.
257 N Sort the current list in the order of network bandwidth (received
259 and transmitted). The one-but-last column changes to ``NET''.
260 A Sort the current list automatically in the order of the most busy
262 system resource during this interval. The one-but-last column
263 shows either ``ACPU'', ``AMEM'', ``ADSK'' or ``ANET'' (the preced‐
264 ing 'A' indicates automatic sorting-order). The most busy
265 resource is determined by comparing the weighted busy-percentages
266 of the system resources, as described earlier in the section COL‐
267 ORS.
268 This option remains valid until another sorting-order is explic‐
269 itly selected again.
270 A sorting-order for disk is only possible when "storage account‐
271 ing" is active. A sorting-order for network is only possible when
272 the kernel module `netatop' is loaded.
273 Miscellaneous interactive commands:
275 ? Request for help information (also the key 'h' can be pressed).
277 V Request for version information (version number and date).
279 R Gather and calculate the proportional set size of processes (tog‐
281 gle). Gathering of all values that are needed to calculate the
282 PSIZE of a process is a relatively time-consuming task, so this
283 key should only be active when analyzing the resident memory con‐
284 sumption of processes.
285 x Suppress colors to highlight critical resources (toggle).
287 Whether this key is active or not can be seen in the header line.
288 z The pause key can be used to freeze the current situation in order
290 to investigate the output on the screen. While pcp-atop is paused,
291 the keys described above can be pressed to show other information
292 about the current list of processes. Whenever the pause key is
293 pressed again, pcp-atop will continue with a next sample.
294 i Modify the interval timer (default: 10 seconds). If an interval
296 timer of 0 is entered, the interval timer is switched off. In that
297 case a new sample can only be triggered manually by pressing the
298 key 't'.
299 t Trigger a new sample manually. This key can be pressed if the cur‐
301 rent sample should be finished before the timer has exceeded, or
302 if no timer is set at all (interval timer defined as 0). In the
303 latter case pcp-atop can be used as a stopwatch to measure the
304 load being caused by a particular application transaction, without
305 knowing on beforehand how many seconds this transaction will last.
306 When viewing the contents of a raw file, this key can be used to
308 show the next sample from the file.
309 T When viewing the contents of a raw file, this key can be used to
311 show the previous sample from the file.
312 b When viewing the contents of a raw file, this key can be used to
314 branch to a certain timestamp within the file (either forward or
315 backward).
316 r Reset all counters to zero to see the system and process activity
318 since boot again.
319 When viewing the contents of a raw file, this key can be used to
321 rewind to the beginning of the file again.
322 U Specify a search string for specific user names as a regular
324 expression. From now on, only (active) processes will be shown
325 from a user which matches the regular expression. The system sta‐
326 tistics are still system wide. If the Enter-key is pressed with‐
327 out specifying a name, (active) processes of all users will be
328 shown again.
329 Whether this key is active or not can be seen in the header line.
330 I Specify a list with one or more PIDs to be selected. From now on,
332 only processes will be shown with a PID which matches one of the
333 given list. The system statistics are still system wide. If the
334 Enter-key is pressed without specifying a PID, all (active) pro‐
335 cesses will be shown again.
336 Whether this key is active or not can be seen in the header line.
337 P Specify a search string for specific process names as a regular
339 expression. From now on, only processes will be shown with a name
340 which matches the regular expression. The system statistics are
341 still system wide. If the Enter-key is pressed without specifying
342 a name, all (active) processes will be shown again.
343 Whether this key is active or not can be seen in the header line.
344 / Specify a specific command line search string as a regular expres‐
346 sion. From now on, only processes will be shown with a command
347 line which matches the regular expression. The system statistics
348 are still system wide. If the Enter-key is pressed without speci‐
349 fying a string, all (active) processes will be shown again.
350 Whether this key is active or not can be seen in the header line.
351 S Specify search strings for specific logical volume names, specific
353 disk names and specific network interface names. All search
354 strings are interpreted as a regular expressions. From now on,
355 only those system resources are shown that match the concerning
356 regular expression. If the Enter-key is pressed without specify‐
357 ing a search string, all (active) system resources of that type
358 will be shown again.
359 Whether this key is active or not can be seen in the header line.
360 a The `all/active' key can be used to toggle between only show‐
362 ing/accumulating the processes that were active during the last
363 interval (default) or showing/accumulating all processes.
364 Whether this key is active or not can be seen in the header line.
365 G By default, pcp-atop shows/accumulates the processes that are
367 alive and the processes that are exited during the last interval.
368 With this key (toggle), showing/accumulating the processes that
369 are exited can be suppressed.
370 Whether this key is active or not can be seen in the header line.
371 f Show a fixed (maximum) number of header lines for system resources
373 (toggle). By default only the lines are shown about system
374 resources (CPUs, paging, logical volumes, disks, network inter‐
375 faces) that really have been active during the last interval.
376 With this key you can force pcp-atop to show lines of inactive
377 resources as well.
378 Whether this key is active or not can be seen in the header line.
379 F Suppress sorting of system resources (toggle). By default system
381 resources (CPUs, logical volumes, disks, network interfaces) are
382 sorted on utilization.
383 Whether this key is active or not can be seen in the header line.
384 1 Show relevant counters as an average per second (in the format
386 `..../s') instead of as a total during the interval (toggle).
387 Whether this key is active or not can be seen in the header line.
388 l Limit the number of system level lines for the counters per-cpu,
390 the active disks and the network interfaces. By default lines are
391 shown of all CPUs, disks and network interfaces which have been
392 active during the last interval. Limiting these lines can be use‐
393 ful on systems with huge number CPUs, disks or interfaces in order
394 to be able to run pcp-atop on a screen/window with e.g. only 24
395 lines.
396 For all mentioned resources the maximum number of lines can be
397 specified interactively. When using the flag -l the maximum number
398 of per-cpu lines is set to 0, the maximum number of disk lines to
399 5 and the maximum number of interface lines to 3. These values
400 can be modified again in interactive mode.
401 k Send a signal to an active process (a.k.a. kill a process).
403 q Quit the program.
405 PgDn Show the next page of the process/thread list.
407 With the arrow-down key the list can be scrolled downwards with
408 single lines.
409 ^F Show the next page of the process/thread list (forward).
411 With the arrow-down key the list can be scrolled downwards with
412 single lines.
413 PgUp Show the previous page of the process/thread list.
415 With the arrow-up key the list can be scrolled upwards with single
416 lines.
417 ^B Show the previous page of the process/thread list (backward).
419 With the arrow-up key the list can be scrolled upwards with single
420 lines.
421 ^L Redraw the screen.
423
425 In order to store system and process level statistics for long-term
426 analysis (e.g. to check the system load and the active processes run‐
427 ning yesterday between 3:00 and 4:00 PM), pcp-atop can store the system
428 and process level statistics in the PCP archive format, as an archive
429 folio (see mkaf(1)).
430 By default only processes which have been active during the interval
431 are stored in the raw file. When the flag -a is specified, all pro‐
432 cesses will be stored.
433 The interval (default: 10 seconds) and number of samples (default:
434 infinite) can be passed as last arguments. Instead of the number of
435 samples, the flag -S can be used to indicate that pcp-atop should fin‐
436 ish anyhow before midnight.
437 A PCP archive can be read and visualized again with the flag -r fol‐
439 lowed by the filename. If no filename is specified, the file
440 $PCP_LOG_DIR/pmlogger/HOST/YYYYMMDD is opened for input (where YYYYMMDD
441 are digits representing the current date, and HOST is the hostname of
442 the machine being logged). If a filename is specified in the format
443 YYYYMMDD (representing any valid date), the file $PCP_LOG_DIR/pmlog‐
444 ger/HOST/YYYYMMDD is opened. If a filename with the symbolic name y is
445 specified, yesterday's daily logfile is opened (this can be repeated so
446 'yyyy' indicates the logfile of four days ago).
447 The samples from the file can be viewed interactively by using the key
448 't' to show the next sample, the key 'T' to show the previous sample,
449 the key 'b' to branch to a particular time or the key 'r' to rewind to
450 the begin of the file.
451 When output is redirected to a file or pipe, pcp-atop prints all sam‐
452 ples in plain ASCII. The default line length is 80 characters in that
453 case; with the flag -L followed by an alternate line length, more (or
454 less) columns will be shown.
455 With the flag -b (begin time) and/or -e (end time) followed by a time
456 argument of the form HH:MM, a certain time period within the raw file
457 can be selected.
458
460 The first sample shows the system level activity since boot (the
461 elapsed time in the header shows the time since boot). Note that par‐
462 ticular counters could have reached their maximum value (several times)
463 and started by zero again, so do not rely on these figures.
464 For every sample pcp-atop first shows the lines related to system level
466 activity. If a particular system resource has not been used during the
467 interval, the entire line related to this resource is suppressed. So
468 the number of system level lines may vary for each sample.
469 After that a list is shown of processes which have been active during
470 the last interval. This list is by default sorted on cpu consumption,
471 but this order can be changed by the keys which are previously
472 described.
473 If values have to be shown by pcp-atop which do not fit in the column
475 width, another format is used. If e.g. a cpu-consumption of 233216 mil‐
476 liseconds should be shown in a column width of 4 positions, it is shown
477 as `233s' (in seconds). For large memory figures, another unit is cho‐
478 sen if the value does not fit (Mb instead of Kb, Gb instead of Mb, Tb
479 instead of Gb, ...). For other values, a kind of exponent notation is
480 used (value 123456789 shown in a column of 5 positions gives 123e6).
481
483 The system level information consists of the following output lines:
484 PRC Process and thread level totals.
486 This line contains the total cpu time consumed in system mode
487 (`sys') and in user mode (`user'), the total number of processes
488 present at this moment (`#proc'), the total number of threads
489 present at this moment in state `running' (`#trun'), `sleeping
490 interruptible' (`#tslpi') and `sleeping uninterruptible'
491 (`#tslpu'), the number of zombie processes (`#zombie'), the number
492 of clone system calls (`clones'), and the number of processes that
493 ended during the interval (`#exit') when process accounting is
494 used. Instead of `#exit` the last column may indicate that process
495 accounting could not be activated (`no procacct`).
496 If the screen-width does not allow all of these counters, only a
497 relevant subset is shown.
498 CPU CPU utilization.
500 At least one line is shown for the total occupation of all CPUs
501 together.
502 In case of a multi-processor system, an additional line is shown
503 for every individual processor (with `cpu' in lower case), sorted
504 on activity. Inactive CPUs will not be shown by default. The
505 lines showing the per-cpu occupation contain the cpu number in the
506 last field.
507 Every line contains the percentage of cpu time spent in kernel
509 mode by all active processes (`sys'), the percentage of cpu time
510 consumed in user mode (`user') for all active processes (including
511 processes running with a nice value larger than zero), the per‐
512 centage of cpu time spent for interrupt handling (`irq') including
513 softirq, the percentage of unused cpu time while no processes were
514 waiting for disk-I/O (`idle'), and the percentage of unused cpu
515 time while at least one process was waiting for disk-I/O (`wait').
516 In case of per-cpu occupation, the last column shows the cpu num‐
517 ber and the wait percentage (`w') for that cpu. The number of
518 lines showing the per-cpu occupation can be limited.
519 For virtual machines the steal-percentage is shown (`steal'),
521 reflecting the percentage of cpu time stolen by other virtual
522 machines running on the same hardware.
523 For physical machines hosting one or more virtual machines, the
524 guest-percentage is shown (`guest'), reflecting the percentage of
525 cpu time used by the virtual machines. Notice that this percentage
526 overlaps the user-percentage.
527 In case of frequency-scaling, all previously mentioned CPU-per‐
529 centages are relative to the used scaling of the CPU during the
530 interval. If a CPU has been active for e.g. 50% in user mode dur‐
531 ing the interval while the frequency-scaling of that CPU was 40%,
532 only 20% of the full capacity of the CPU has been used in user
533 mode.
534 If the screen-width does not allow all of these counters, only a
536 relevant subset is shown.
537 CPL CPU load information.
539 This line contains the load average figures reflecting the number
540 of threads that are available to run on a CPU (i.e. part of the
541 runqueue) or that are waiting for disk I/O. These figures are
542 averaged over 1 (`avg1'), 5 (`avg5') and 15 (`avg15') minutes.
543 Furthermore the number of context switches (`csw'), the number of
544 serviced interrupts (`intr') and the number of available CPUs are
545 shown.
546 If the screen-width does not allow all of these counters, only a
548 relevant subset is shown.
549 MEM Memory occupation.
551 This line contains the total amount of physical memory (`tot'),
552 the amount of memory which is currently free (`free'), the amount
553 of memory in use as page cache including the total resident shared
554 memory (`cache'), the amount of memory within the page cache that
555 has to be flushed to disk (`dirty'), the amount of memory used for
556 filesystem meta data (`buff'), the amount of memory being used for
557 kernel mallocs (`slab'), the amount of slab memory that is
558 reclaimable (`slrec'), the resident size of shared memory includ‐
559 ing tmpfs (`shmem`), the resident size of shared memory (`shrss`)
560 the amount of shared memory that is currently swapped (`shswp`),
561 the amount of memory that is currently claimed by vmware's balloon
562 driver (`vmbal`), the amount of memory that is claimed for huge
563 pages (`hptot`), and the amount of huge page memory that is really
564 in use (`hpuse`).
565 If the screen-width does not allow all of these counters, only a
567 relevant subset is shown.
568 SWP Swap occupation and overcommit info.
570 This line contains the total amount of swap space on disk (`tot')
571 and the amount of free swap space (`free').
572 Furthermore the committed virtual memory space (`vmcom') and the
573 maximum limit of the committed space (`vmlim', which is by default
574 swap size plus 50% of memory size) is shown. The committed space
575 is the reserved virtual space for all allocations of private mem‐
576 ory space for processes. The kernel only verifies whether the com‐
577 mitted space exceeds the limit if strict overcommit handling is
578 configured (vm.overcommit_memory is 2).
579 PAG Paging frequency.
581 This line contains the number of scanned pages (`scan') due to the
582 fact that free memory drops below a particular threshold and the
583 number times that the kernel tries to reclaim pages due to an
584 urgent need (`stall').
585 Also the number of memory pages the system read from swap space
586 (`swin') and the number of memory pages the system wrote to swap
587 space (`swout') are shown.
588 LVM/MDD/DSK
590 Logical volume/multiple device/disk utilization.
591 Per active unit one line is produced, sorted on unit activity.
592 Such line shows the name (e.g. VolGroup00-lvtmp for a logical vol‐
593 ume or sda for a hard disk), the busy percentage i.e. the portion
594 of time that the unit was busy handling requests (`busy'), the
595 number of read requests issued (`read'), the number of write
596 requests issued (`write'), the number of KiBytes per read
597 (`KiB/r'), the number of KiBytes per write (`KiB/w'), the number
598 of MiBytes per second throughput for reads (`MBr/s'), the number
599 of MiBytes per second throughput for writes (`MBw/s'), the average
600 queue depth (`avq') and the average number of milliseconds needed
601 by a request (`avio') for seek, latency and data transfer.
602 If the screen-width does not allow all of these counters, only a
603 relevant subset is shown.
604 The number of lines showing the units can be limited per class
606 (LVM, MDD or DSK) with the 'l' key or statically (see separate
607 man-page of pcp-atoprc(5)). By specifying the value 0 for a par‐
608 ticular class, no lines will be shown any more for that class.
609 NFM Network Filesystem (NFS) mount at the client side.
611 For each NFS-mounted filesystem, a line is shown that contains the
612 mounted server directory, the name of the server (`srv'), the
613 total number of bytes physically read from the server (`read') and
614 the total number of bytes physically written to the server
615 (`write'). Data transfer is subdivided in the number of bytes
616 read via normal read() system calls (`nread'), the number of bytes
617 written via normal read() system calls (`nwrit'), the number of
618 bytes read via direct I/O (`dread'), the number of bytes written
619 via direct I/O (`dwrit'), the number of bytes read via memory
620 mapped I/O pages (`mread'), and the number of bytes written via
621 memory mapped I/O pages (`mwrit').
622 NFC Network Filesystem (NFS) client side counters.
624 This line contains the number of RPC calls issues by local pro‐
625 cesses (`rpc'), the number of read RPC calls (`read`) and write
626 RPC calls (`rpwrite') issued to the NFS server, the number of RPC
627 calls being retransmitted (`retxmit') and the number of authoriza‐
628 tion refreshes (`autref').
629 NFS Network Filesystem (NFS) server side counters.
631 This line contains the number of RPC calls received from NFS
632 clients (`rpc'), the number of read RPC calls received (`cread`),
633 the number of write RPC calls received (`cwrit'), the number of
634 network requests handled via TCP (`nettcp'), the number of network
635 requests handled via UDP (`netudp'), the number of Megabytes/sec‐
636 ond returned to read requests by clients (`MBcr/s`), the number of
637 Megabytes/second passed in write requests by clients (`MBcw/s`),
638 the number of reply cache hits (`rchits'), the number of reply
639 cache misses (`rcmiss') and the number of uncached requests
640 (`rcnoca'). Furthermore some error counters indicating the number
641 of requests with a bad format (`badfmt') or a bad authorization
642 (`badaut'), and a counter indicating the number of bad clients
643 (`badcln'). and the number of authorization refreshes (`autref').
644 NET Network utilization (TCP/IP).
646 One line is shown for activity of the transport layer (TCP and
647 UDP), one line for the IP layer and one line per active interface.
648 For the transport layer, counters are shown concerning the number
649 of received TCP segments including those received in error
650 (`tcpi'), the number of transmitted TCP segments excluding those
651 containing only retransmitted octets (`tcpo'), the number of UDP
652 datagrams received (`udpi'), the number of UDP datagrams transmit‐
653 ted (`udpo'), the number of active TCP opens (`tcpao'), the number
654 of passive TCP opens (`tcppo'), the number of TCP output retrans‐
655 missions (`tcprs'), the number of TCP input errors (`tcpie'), the
656 number of TCP output resets (`tcpor'), the number of UDP no ports
657 (`udpnp'), and the number of UDP input errors (`udpie').
658 If the screen-width does not allow all of these counters, only a
659 relevant subset is shown.
660 These counters are related to IPv4 and IPv6 combined.
661 For the IP layer, counters are shown concerning the number of IP
663 datagrams received from interfaces, including those received in
664 error (`ipi'), the number of IP datagrams that local higher-layer
665 protocols offered for transmission (`ipo'), the number of received
666 IP datagrams which were forwarded to other interfaces (`ipfrw'),
667 the number of IP datagrams which were delivered to local higher-
668 layer protocols (`deliv'), the number of received ICMP datagrams
669 (`icmpi'), and the number of transmitted ICMP datagrams (`icmpo').
670 If the screen-width does not allow all of these counters, only a
671 relevant subset is shown.
672 These counters are related to IPv4 and IPv6 combined.
673 For every active network interface one line is shown, sorted on
675 the interface activity. Such line shows the name of the interface
676 and its busy percentage in the first column. The busy percentage
677 for half duplex is determined by comparing the interface speed
678 with the number of bits transmitted and received per second; for
679 full duplex the interface speed is compared with the highest of
680 either the transmitted or the received bits. When the interface
681 speed can not be determined (e.g. for the loopback interface),
682 `---' is shown instead of the percentage.
683 Furthermore the number of received packets (`pcki'), the number of
684 transmitted packets (`pcko'), the line speed of the interface
685 (`sp'), the effective amount of bits received per second (`si'),
686 the effective amount of bits transmitted per second (`so'), the
687 number of collisions (`coll'), the number of received multicast
688 packets (`mlti'), the number of errors while receiving a packet
689 (`erri'), the number of errors while transmitting a packet
690 (`erro'), the number of received packets dropped (`drpi'), and the
691 number of transmitted packets dropped (`drpo').
692 If the screen-width does not allow all of these counters, only a
693 relevant subset is shown.
694 The number of lines showing the network interfaces can be limited.
695
697 Following the system level information, the processes are shown from
698 which the resource utilization has changed during the last interval.
699 These processes might have used cpu time or issued disk or network
700 requests. However a process is also shown if part of it has been paged
701 out due to lack of memory (while the process itself was in sleep
702 state).
703 Per process the following fields may be shown (in alphabetical order),
705 depending on the current output mode as described in the section INTER‐
706 ACTIVE COMMANDS and depending on the current width of your window:
707 AVGRSZ The average size of one read-action on disk.
709 AVGWSZ The average size of one write-action on disk.
711 CMD The name of the process. This name can be surrounded by
713 "less/greater than" signs (`<name>') which means that the
714 process has finished during the last interval.
715 Behind the abbreviation `CMD' in the header line, the current
716 page number and the total number of pages of the
717 process/thread list are shown.
718 COMMAND-LINE
720 The full command line of the process (including arguments). If
721 the length of the command line exceeds the length of the
722 screen line, the arrow keys -> and <- can be used for horizon‐
723 tal scroll.
724 Behind the verb `COMMAND-LINE' in the header line, the current
725 page number and the total number of pages of the
726 process/thread list are shown.
727 CPU The occupation percentage of this process related to the
729 available capacity for this resource on system level.
730 CPUNR The identification of the CPU the (main) thread is running on
732 or has recently been running on.
733 DSK The occupation percentage of this process related to the total
735 load that is produced by all processes (i.e. total disk
736 accesses by all processes during the last interval).
737 This information is shown when per process "storage account‐
738 ing" is active in the kernel.
739 EGID Effective group-id under which this process executes.
741 ENDATE Date that the process has been finished. If the process is
743 still running, this field shows `active'.
744 ENTIME Time that the process has been finished. If the process is
746 still running, this field shows `active'.
747 ENVID Virtual environment identified (OpenVZ only).
749 EUID Effective user-id under which this process executes.
751 EXC The exit code of a terminated process (second position of col‐
753 umn `ST' is E) or the fatal signal number (second position of
754 column `ST' is S or C).
755 FSGID Filesystem group-id under which this process executes.
757 FSUID Filesystem user-id under which this process executes.
759 MAJFLT The number of page faults issued by this process that have
761 been solved by creating/loading the requested memory page.
762 MEM The occupation percentage of this process related to the
764 available capacity for this resource on system level.
765 MINFLT The number of page faults issued by this process that have
767 been solved by reclaiming the requested memory page from the
768 free list of pages.
769 NET The occupation percentage of this process related to the total
771 load that is produced by all processes (i.e. consumed network
772 bandwidth of all processes during the last interval).
773 This information will only be shown when kernel module
774 `netatop' is loaded.
775 NICE The more or less static priority that can be given to a
777 process on a scale from -20 (high priority) to +19 (low prior‐
778 ity).
779 NPROCS The number of active and terminated processes accumulated for
781 this user or program.
782 PID Process-id.
784 POLI The policies 'norm' (normal, which is SCHED_OTHER), 'btch'
786 (batch) and 'idle' refer to timesharing processes. The poli‐
787 cies 'fifo' (SCHED_FIFO) and 'rr' (round robin, which is
788 SCHED_RR) refer to realtime processes.
789 PPID Parent process-id.
791 PRI The process' priority ranges from 0 (highest priority) to 139
793 (lowest priority). Priority 0 to 99 are used for realtime pro‐
794 cesses (fixed priority independent of their behavior) and pri‐
795 ority 100 to 139 for timesharing processes (variable priority
796 depending on their recent CPU consumption and the nice value).
797 PSIZE The proportional memory size of this process (or user).
799 Every process shares resident memory with other processes.
800 E.g. when a particular program is started several times, the
801 code pages (text) are only loaded once in memory and shared by
802 all incarnations. Also the code of shared libraries is shared
803 by all processes using that shared library, as well as shared
804 memory and memory-mapped files. For the PSIZE calculation of
805 a process, the resident memory of a process that is shared
806 with other processes is divided by the number of sharers.
807 This means, that every process is accounted for a proportional
808 part of that memory. Accumulating the PSIZE values of all pro‐
809 cesses in the system gives a reliable impression of the total
810 resident memory consumed by all processes.
811 Since gathering of all values that are needed to calculate the
812 PSIZE is a relatively time-consuming task, the 'R' key (or
813 '-R' flag) should be active. Gathering these values also
814 requires superuser privileges (otherwise '?K' is shown in the
815 output).
816 RDDSK When the kernel maintains standard io statistics (>= 2.6.20):
818 The read data transfer issued physically on disk (so reading
819 from the disk cache is not accounted for).
820 Unfortunately, the kernel aggregates the data tranfer of a
821 process to the data transfer of its parent process when termi‐
822 nating, so you might see transfers for (parent) processes like
823 cron, bash or init, that are not really issued by them.
824 RGID The real group-id under which the process executes.
826 RGROW The amount of resident memory that the process has grown dur‐
828 ing the last interval. A resident growth can be caused by
829 touching memory pages which were not physically created/loaded
830 before (load-on-demand). Note that a resident growth can also
831 be negative e.g. when part of the process is paged out due to
832 lack of memory or when the process frees dynamically allocated
833 memory. For a process which started during the last interval,
834 the resident growth reflects the total resident size of the
835 process at that moment.
836 RSIZE The total resident memory usage consumed by this process (or
838 user). Notice that the RSIZE of a process includes all resi‐
839 dent memory used by that process, even if certain memory parts
840 are shared with other processes (see also the explanation of
841 PSIZE).
842 RTPR Realtime priority according the POSIX standard. Value can be
844 0 for a timesharing process (policy 'norm', 'btch' or 'idle')
845 or ranges from 1 (lowest) till 99 (highest) for a realtime
846 process (policy 'rr' or 'fifo').
847 RUID The real user-id under which the process executes.
849 S The current state of the (main) thread: `R' for running (cur‐
851 rently processing or in the runqueue), `S' for sleeping inter‐
852 ruptible (wait for an event to occur), `D' for sleeping non-
853 interruptible, `Z' for zombie (waiting to be synchronized with
854 its parent process), `T' for stopped (suspended or traced),
855 `W' for swapping, and `E' (exit) for processes which have fin‐
856 ished during the last interval.
857 SGID The saved group-id of the process.
859 ST The status of a process.
861 The first position indicates if the process has been started
862 during the last interval (the value N means 'new process').
863 The second position indicates if the process has been finished
865 during the last interval.
866 The value E means 'exit' on the process' own initiative; the
867 exit code is displayed in the column `EXC'.
868 The value S means that the process has been terminated unvol‐
869 untarily by a signal; the signal number is displayed in the in
870 the column `EXC'.
871 The value C means that the process has been terminated unvol‐
872 untarily by a signal, producing a core dump in its current
873 directory; the signal number is displayed in the column `EXC'.
874 STDATE The start date of the process.
876 STTIME The start time of the process.
878 SUID The saved user-id of the process.
880 SWAPSZ The swap space consumed by this process (or user).
882 SYSCPU CPU time consumption of this process in system mode (kernel
884 mode), usually due to system call handling.
885 THR Total number of threads within this process. All related
887 threads are contained in a thread group, represented by pcp-
888 atop as one line or as a separate line when the 'y' key (or -y
889 flag) is active.
890 On Linux 2.4 systems it is hardly possible to determine which
892 threads (i.e. processes) are related to the same thread group.
893 Every thread is represented by pcp-atop as a separate line.
894 TID Thread-id. All threads within a process run with the same PID
896 but with a different TID. This value is shown for individual
897 threads in multi-threaded processes (when using the key 'y').
898 TRUN Number of threads within this process that are in the state
900 'running' (R).
901 TSLPI Number of threads within this process that are in the state
903 'interruptible sleeping' (S).
904 TSLPU Number of threads within this process that are in the state
906 'uninterruptible sleeping' (D).
907 USRCPU CPU time consumption of this process in user mode, due to pro‐
909 cessing the own program text.
910 VDATA The virtual memory size of the private data used by this
912 process (including heap and shared library data).
913 VGROW The amount of virtual memory that the process has grown during
915 the last interval. A virtual growth can be caused by e.g.
916 issueing a malloc() or attaching a shared memory segment. Note
917 that a virtual growth can also be negative by e.g. issueing a
918 free() or detaching a shared memory segment. For a process
919 which started during the last interval, the virtual growth
920 reflects the total virtual size of the process at that moment.
921 VSIZE The total virtual memory usage consumed by this process (or
923 user).
924 VSLIBS The virtual memory size of the (shared) text of all shared
926 libraries used by this process.
927 VSTACK The virtual memory size of the (private) stack used by this
929 process
930 VSTEXT The virtual memory size of the (shared) text of the executable
932 program.
933 WRDSK When the kernel maintains standard io statistics (>= 2.6.20):
935 The write data transfer issued physically on disk (so writing
936 to the disk cache is not accounted for). This counter is
937 maintained for the application process that writes its data to
938 the cache (assuming that this data is physically transferred
939 to disk later on). Notice that disk I/O needed for swapping is
940 not taken into account.
941 Unfortunately, the kernel aggregates the data tranfer of a
942 process to the data transfer of its parent process when termi‐
943 nating, so you might see transfers for (parent) processes like
944 cron, bash or init, that are not really issued by them.
945 WCANCL When the kernel maintains standard io statistics (>= 2.6.20):
947 The write data transfer previously accounted for this process
948 or another process that has been cancelled. Suppose that a
949 process writes new data to a file and that data is removed
950 again before the cache buffers have been flushed to disk.
951 Then the original process shows the written data as WRDSK,
952 while the process that removes/truncates the file shows the
953 unflushed removed data as WCANCL.
954
956 With the flag -P followed by a list of one or more labels (comma-sepa‐
957 rated), parseable output is produced for each sample. The labels that
958 can be specified for system-level statistics correspond to the labels
959 (first verb of each line) that can be found in the interactive output:
960 "CPU", "cpu" "CPL" "MEM", "SWP", "PAG", "LVM", "MDD", "DSK", "NFM",
961 "NFC", "NFS" and "NET".
962 For process-level statistics special labels are introduced: "PRG" (gen‐
963 eral), "PRC" (cpu), "PRM" (memory), "PRD" (disk, only if "storage
964 accounting" is active) and "PRN" (network, only if the kernel module
965 'netatop' has been installed).
966 With the label "ALL", all system and process level statistics are
967 shown.
968 For every interval all requested lines are shown whereafter pcp-atop
970 shows a line just containing the label "SEP" as a separator before the
971 lines for the next sample are generated.
972 When a sample contains the values since boot, pcp-atop shows a line
973 just containing the label "RESET" before the lines for this sample are
974 generated.
975 The first part of each output-line consists of the following six
977 fields: label (the name of the label), host (the name of this machine),
978 epoch (the time of this interval as number of seconds since 1-1-1970),
979 date (date of this interval in format YYYY/MM/DD), time (time of this
980 interval in format HH:MM:SS), and interval (number of seconds elapsed
981 for this interval).
982 The subsequent fields of each output-line depend on the label:
984 CPU Subsequent fields: total number of clock-ticks per second for
986 this machine, number of processors, consumption for all CPUs
987 in system mode (clock-ticks), consumption for all CPUs in user
988 mode (clock-ticks), consumption for all CPUs in user mode for
989 niced processes (clock-ticks), consumption for all CPUs in
990 idle mode (clock-ticks), consumption for all CPUs in wait mode
991 (clock-ticks), consumption for all CPUs in irq mode (clock-
992 ticks), consumption for all CPUs in softirq mode (clock-
993 ticks), consumption for all CPUs in steal mode (clock-ticks),
994 consumption for all CPUs in guest mode (clock-ticks) overlap‐
995 ping user mode, frequency of all CPUs and frequency percentage
996 of all CPUs.
997 cpu Subsequent fields: total number of clock-ticks per second for
999 this machine, processor-number, consumption for this CPU in
1000 system mode (clock-ticks), consumption for this CPU in user
1001 mode (clock-ticks), consumption for this CPU in user mode for
1002 niced processes (clock-ticks), consumption for this CPU in
1003 idle mode (clock-ticks), consumption for this CPU in wait mode
1004 (clock-ticks), consumption for this CPU in irq mode (clock-
1005 ticks), consumption for this CPU in softirq mode (clock-
1006 ticks), consumption for this CPU in steal mode (clock-ticks),
1007 consumption for this CPU in guest mode (clock-ticks) overlap‐
1008 ping user mode, frequency of this CPU and frequency percentage
1009 of this CPU.
1010 CPL Subsequent fields: number of processors, load average for last
1012 minute, load average for last five minutes, load average for
1013 last fifteen minutes, number of context-switches, and number
1014 of device interrupts.
1015 MEM Subsequent fields: page size for this machine (in bytes), size
1017 of physical memory (pages), size of free memory (pages), size
1018 of page cache (pages), size of buffer cache (pages), size of
1019 slab (pages), dirty pages in cache (pages), reclaimable part
1020 of slab (pages), size of vmware's balloon pages (pages), total
1021 size of shared memory (pages), size of resident shared memory
1022 (pages), size of swapped shared memory (pages), huge page size
1023 (in bytes), total size of huge pages (huge pages), and size of
1024 free huge pages (huge pages).
1025 SWP Subsequent fields: page size for this machine (in bytes), size
1027 of swap (pages), size of free swap (pages), 0 (future use),
1028 size of committed space (pages), and limit for committed space
1029 (pages).
1030 PAG Subsequent fields: page size for this machine (in bytes), num‐
1032 ber of page scans, number of allocstalls, 0 (future use), num‐
1033 ber of swapins, and number of swapouts.
1034 LVM/MDD/DSK
1036 For every logical volume/multiple device/hard disk one line is
1037 shown.
1038 Subsequent fields: name, number of milliseconds spent for I/O,
1039 number of reads issued, number of sectors transferred for
1040 reads, number of writes issued, and number of sectors trans‐
1041 ferred for write.
1042 NFM Subsequent fields: mounted NFS filesystem, total number of
1044 bytes read, total number of bytes written, number of bytes
1045 read by normal system calls, number of bytes written by normal
1046 system calls, number of bytes read by direct I/O, number of
1047 bytes written by direct I/O, number of pages read by memory-
1048 mapped I/O, and number of pages written by memory-mapped I/O.
1049 NFC Subsequent fields: number of transmitted RPCs, number of
1051 transmitted read RPCs, number of transmitted write RPCs, num‐
1052 ber of RPC retransmissions, and number of authorization
1053 refreshes.
1054 NFS Subsequent fields: number of handled RPCs, number of received
1056 read RPCs, number of received write RPCs, number of bytes read
1057 by clients, number of bytes written by clients, number of RPCs
1058 with bad format, number of RPCs with bad authorization, number
1059 of RPCs from bad client, total number of handled network
1060 requests, number of handled network requests via TCP, number
1061 of handled network requests via UDP, number of handled TCP
1062 connections, number of hits on reply cache, number of misses
1063 on reply cache, and number of uncached requests.
1064 NET First one line is produced for the upper layers of the TCP/IP
1066 stack.
1067 Subsequent fields: the verb "upper", number of packets
1068 received by TCP, number of packets transmitted by TCP, number
1069 of packets received by UDP, number of packets transmitted by
1070 UDP, number of packets received by IP, number of packets
1071 transmitted by IP, number of packets delivered to higher lay‐
1072 ers by IP, and number of packets forwarded by IP.
1073 Next one line is shown for every interface.
1075 Subsequent fields: name of the interface, number of packets
1076 received by the interface, number of bytes received by the
1077 interface, number of packets transmitted by the interface,
1078 number of bytes transmitted by the interface, interface speed,
1079 and duplex mode (0=half, 1=full).
1080 PRG For every process one line is shown.
1082 Subsequent fields: PID (unique ID of task), name (between
1083 brackets), state, real uid, real gid, TGID (group number of
1084 related tasks/threads), total number of threads, exit code,
1085 start time (epoch), full command line (between brackets),
1086 PPID, number of threads in state 'running' (R), number of
1087 threads in state 'interruptible sleeping' (S), number of
1088 threads in state 'uninterruptible sleeping' (D), effective
1089 uid, effective gid, saved uid, saved gid, filesystem uid,
1090 filesystem gid, elapsed time (hertz), is_process (y/n), vir‐
1091 tual pid and container id.
1092 PRC For every process one line is shown.
1094 Subsequent fields: PID, name (between brackets), state, total
1095 number of clock-ticks per second for this machine, CPU-con‐
1096 sumption in user mode (clockticks), CPU-consumption in system
1097 mode (clockticks), nice value, priority, realtime priority,
1098 scheduling policy, current CPU, sleep average, TGID (group
1099 number of related tasks/threads) and is_process (y/n).
1100 PRM For every process one line is shown.
1102 Subsequent fields: PID, name (between brackets), state, page
1103 size for this machine (in bytes), virtual memory size
1104 (Kbytes), resident memory size (Kbytes), shared text memory
1105 size (Kbytes), virtual memory growth (Kbytes), resident memory
1106 growth (Kbytes), number of minor page faults, number of major
1107 page faults, virtual library exec size (Kbytes), virtual data
1108 size (Kbytes), virtual stack size (Kbytes), swap space used
1109 (Kbytes), TGID (group number of related tasks/threads),
1110 is_process (y/n) and proportional set size (Kbytes) if in 'R'
1111 option is specified.
1112 PRD For every process one line is shown.
1114 Subsequent fields: PID, name (between brackets), state, obso‐
1115 leted kernel patch installed ('n'), standard io statistics
1116 used ('y' or 'n'), number of reads on disk, cumulative number
1117 of sectors read, number of writes on disk, cumulative number
1118 of sectors written, cancelled number of written sectors, TGID
1119 (group number of related tasks/threads) and is_process (y/n).
1120 If the standard I/O statistics (>= 2.6.20) are not used, the
1121 disk I/O counters per process are not relevant. The counters
1122 'number of reads on disk' and 'number of writes on disk' are
1123 obsoleted anyhow.
1124 PRN For every process one line is shown.
1126 Subsequent fields: PID, name (between brackets), state, kernel
1127 module 'netatop' loaded ('y' or 'n'), number of TCP-packets
1128 transmitted, cumulative size of TCP-packets transmitted, num‐
1129 ber of TCP-packets received, cumulative size of TCP-packets
1130 received, number of UDP-packets transmitted, cumulative size
1131 of UDP-packets transmitted, number of UDP-packets received,
1132 cumulative size of UDP-packets transmitted, number of raw
1133 packets transmitted (obsolete, always 0), number of raw pack‐
1134 ets received (obsolete, always 0), TGID (group number of
1135 related tasks/threads) and is_process (y/n).
1136
1138 To monitor the current system load interactively with an interval of 5
1139 seconds:
1140 pcp atop 5
1142 To monitor the system load and write it to a file (in plain ASCII) with
1144 an interval of one minute during half an hour with active processes
1145 sorted on memory consumption:
1146 pcp atop -M 60 30 > /log/pcp-atop.mem
1148 Store information about the system and process activity in a PCP ar‐
1150 chive folio with an interval of ten minutes during an hour:
1151 pcp atop -w /tmp/pcp-atop 600 6
1153 View the contents of this file interactively:
1155 pcp atop -r /tmp/pcp-atop
1157 View the processor and disk utilization of this file in parseable for‐
1159 mat:
1160 pcp atop -PCPU,DSK -r /tmp/pcp-atop.raw
1162 View the contents of today's standard logfile interactively:
1164 pcp atop -r
1166 View the contents of the standard logfile of the day before yesterday
1168 interactively:
1169 pcp atop -r yy
1171 View the contents of the standard logfile of 2014, June 7 from 02:00 PM
1173 onwards interactively:
1174 pcp atop -r 20140607 -b 14:00
1176
1178 /etc/atoprc
1179 Configuration file containing system-wide default values. See
1180 related man-page.
1181 ~/.atoprc
1183 Configuration file containing personal default values. See
1184 related man-page.
1185
1187 pcp-atop is based on the source code of the atop(1) command from
1188 http://atoptool.nl and aims to be command line and output compatible
1189 with it as much as possible. Some features of that atop command are
1190 not available in pcp-atop.
1191 Some features of pcp-atop (such as reporting on the Apache HTTP daemon,
1193 and NFS client mounts) are only activated if the corresonding PCP met‐
1194 rics are available. Refer to the documentation for pmdaapache(1) and
1195 pmdanfsclient(1) for further details on activating these metrics.
1196
1198 pcp(1), pcp-atopsar(1), pmdaapache(1), pmdanfsclient(1), mkaf(1),
1199 pmlogger(1), pmlogger_daily(1), PCPIntro(1) and pcp-atoprc(5).
1200Performance Co-Pilot PCP PCP-ATOP(1)