1PCP-ATOP(1) General Commands Manual PCP-ATOP(1)
2
6 pcp-atop - Advanced System and Process Monitor
7
9 Interactive Usage:
10 pcp [pcp options] atop [-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y]
12 [-C|-M|-D|-N|-A] [-afFG1xR] [-L linelen] [-Plabel[,label]...] [interval
13 [samples]]
14 Writing and reading raw logfiles:
16 pcp atop -w rawfile [-a] [-S] [interval [samples]]
18 pcp atop -r [ rawfile ] [-b hh:mm ] [-e hh:mm ]
19 [-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y] [-C|-M|-D|-N|-A] [-fFG1xR] [-L line‐
20 len] [-Plabel[,label]...]
21
23 The program pcp-atop is an interactive monitor to view various aspects
24 of load on a system. It shows the occupation of the most critical
25 hardware resources (from a performance point of view) on system level,
26 i.e. cpu, memory, disk and network.
27 It also shows which processes are responsible for the indicated load
28 with respect to cpu and memory load on process level. Disk load is
29 shown per process if "storage accounting" is active in the kernel.
30 Every interval (default: 10 seconds) information is shown about the
32 resource occupation on system level (cpu, memory, disks and network
33 layers), followed by a list of processes which have been active during
34 the last interval (note that all processes that were unchanged during
35 the last interval are not shown, unless the key 'a' has been pressed or
36 unless sorting on memory occupation is done). If the list of active
37 processes does not entirely fit on the screen, only the top of the list
38 is shown (sorted in order of activity).
39 The intervals are repeated till the number of samples (specified as
40 command argument) is reached, or till the key 'q' is pressed in inter‐
41 active mode.
42 When invoked via the pcp(1) command, the PCPIntro(1) options -h/--host,
44 -a/--archive, -O/--origin, -s/--samples, -t/--interval, -Z/--timezone
45 and several other pcp options become indirectly available. The long
46 option form of these is directly available. Additionally, the --hot‐
47 proc option can be used to request the per-process PCP metrics be used
48 instead of the default proc metrics from pmdaproc(1).
49 When pcp-atop is started, it checks whether the standard output channel
51 is connected to a screen, or to a file/pipe. In the first case it pro‐
52 duces screen control codes (via the ncurses library) and behaves inter‐
53 actively; in the second case it produces flat ASCII-output.
54 In interactive mode, the output of pcp-atop scales dynamically to the
56 current dimensions of the screen/window.
57 If the window is resized horizontally, columns will be added or removed
58 automatically. For this purpose, every column has a particular weight.
59 The columns with the highest weights that fit within the current width
60 will be shown.
61 If the window is resized vertically, lines of the process/thread list
62 will be added or removed automatically.
63 Furthermore in interactive mode the output of pcp-atop can be con‐
65 trolled by pressing particular keys. However it is also possible to
66 specify such key as flag on the command line. In that case pcp-atop
67 switches to the indicated mode on beforehand; this mode can be modified
68 again interactively. Specifying such key as flag is especially useful
69 when running pcp-atop with output to a pipe or file (non-interac‐
70 tively). These flags are the same as the keys that can be pressed in
71 interactive mode (see section INTERACTIVE COMMANDS).
72 Additional flags are available to support storage of pcp-atop data in
73 PCP archive format (see section PCP DATA STORAGE).
74
76 For the resource consumption on system level, pcp-atop uses colors to
77 indicate that a critical occupation percentage has been (almost)
78 reached. A critical occupation percentage means that is likely that
79 this load causes a noticeable negative performance influence for appli‐
80 cations using this resource. The critical percentage depends on the
81 type of resource: e.g. the performance influence of a disk with a busy
82 percentage of 80% might be more noticeable for applications/user than a
83 CPU with a busy percentage of 90%.
84 Currently pcp-atop uses the following default values to calculate a
85 weighted percentage per resource:
86 Processor
88 A busy percentage of 90% or higher is considered `critical'.
89 Disk
91 A busy percentage of 70% or higher is considered `critical'.
92 Network
94 A busy percentage of 90% or higher for the load of an interface is
95 considered `critical'.
96 Memory
98 An occupation percentage of 90% is considered `critical'. Notice
99 that this occupation percentage is the accumulated memory consump‐
100 tion of the kernel (including slab) and all processes; the memory
101 for the page cache (`cache' and `buff' in the MEM-line) and the
102 reclaimable part of the slab (`slrec`) is not implied!
103 If the number of pages swapped out (`swout' in the PAG-line) is
104 larger than 10 per second, the memory resource is considered
105 `critical'. A value of at least 1 per second is considered
106 `almost critical'.
107 If the committed virtual memory exceeds the limit (`vmcom' and
108 `vmlim' in the SWP-line), the SWP-line is colored due to overcom‐
109 mitting the system.
110 Swap
112 An occupation percentage of 80% is considered `critical' because
113 swap space might be completely exhausted in the near future; it is
114 not critical from a performance point-of-view.
115 These default values can be modified in the configuration file (see
117 separate man-page of pcp-atoprc).
118 When a resource exceeds its critical occupation percentage, the con‐
120 cerning values in the screen line are colored red by default.
121 When a resource exceeded (default) 80% of its critical percentage (so
122 it is almost critical), the concerning values in the screen line are
123 colored cyan by default. This `almost critical percentage' (one value
124 for all resources) can be modified in the configuration file (see sepa‐
125 rate man-page of pcp-atoprc).
126 The default colors red and cyan can be modified in the configuration
127 file as well (see separate man-page of pcp-atoprc).
128 With the key 'x' (or flag -x), the use of colors can be suppressed.
130
132 When running pcp-atop interactively (no output redirection), keys can
133 be pressed to control the output. In general, lower case keys can be
134 used to show other information for the active processes and upper case
135 keys can be used to influence the sort order of the active
136 process/thread list.
137 g Show generic output (default).
139 Per process the following fields are shown in case of a window-
141 width of 80 positions: process-id, cpu consumption during the last
142 interval in system and user mode, the virtual and resident memory
143 growth of the process.
144 The subsequent columns depend on the used kernel:
146 When the kernel supports "storage accounting" (>= 2.6.20), the
147 data transfer for read/write on disk, the status and exit code are
148 shown for each process. When the kernel does not support "storage
149 accounting", the username, number of threads in the thread group,
150 the status and exit code are shown.
151 The last columns contain the state, the occupation percentage for
152 the chosen resource (default: cpu) and the process name.
153 When more than 80 positions are available, other information is
155 added.
156 m Show memory related output.
158 Per process the following fields are shown in case of a window-
160 width of 80 positions: process-id, minor and major memory faults,
161 size of virtual shared text, total virtual process size, total
162 resident process size, virtual and resident growth during last
163 interval, memory occupation percentage and process name.
164 When more than 80 positions are available, other information is
166 added.
167 For memory consumption, always all processes are shown (also the
169 processes that were not active during the interval).
170 d Show disk-related output.
172 When "storage accounting" is active in the kernel, the following
174 fields are shown: process-id, amount of data read from disk,
175 amount of data written to disk, amount of data that was written
176 but has been withdrawn again (WCANCL), disk occupation percentage
177 and process name.
178 s Show scheduling characteristics.
180 Per process the following fields are shown in case of a window-
182 width of 80 positions: process-id, number of threads in state
183 'running' (R), number of threads in state 'interruptible sleeping'
184 (S), number of threads in state 'uninterruptible sleeping' (D),
185 scheduling policy (normal timesharing, realtime round-robin, real‐
186 time fifo), nice value, priority, realtime priority, current pro‐
187 cessor, status, exit code, state, the occupation percentage for
188 the chosen resource and the process name.
189 When more than 80 positions are available, other information is
191 added.
192 v Show various process characteristics.
194 Per process the following fields are shown in case of a window-
196 width of 80 positions: process-id, user name and group, start date
197 and time, status (e.g. exit code if the process has finished),
198 state, the occupation percentage for the chosen resource and the
199 process name.
200 When more than 80 positions are available, other information is
202 added.
203 c Show the command line of the process.
205 Per process the following fields are shown: process-id, the occu‐
207 pation percentage for the chosen resource and the command line
208 including arguments.
209 o Show the user-defined line of the process.
211 In the configuration file the keyword ownprocline can be specified
213 with the description of a user-defined output-line.
214 Refer to the man-page of pcp-atoprc for a detailed description.
215 y Show the individual threads within a process (toggle).
217 Single-threaded processes are still shown as one line.
219 For multi-threaded processes, one line represents the process
220 while additional lines show the activity per individual thread (in
221 a different color). Depending on the option 'a' (all or active
222 toggle), all threads are shown or only the threads that were
223 active during the last interval.
224 Whether this key is active or not can be seen in the header line.
225 u Show the process activity accumulated per user.
227 Per user the following fields are shown: number of processes
229 active or terminated during last interval (or in total if combined
230 with command `a'), accumulated cpu consumption during last inter‐
231 val in system and user mode, the current virtual and resident mem‐
232 ory space consumed by active processes (or all processes of the
233 user if combined with command `a').
234 When "storage accounting" is active in the kernel, the accumulated
235 read and write throughput on disk is shown. When the kernel mod‐
236 ule `netatop' has been installed, the number of received and sent
237 network packets are shown.
238 The last columns contain the accumulated occupation percentage for
239 the chosen resource (default: cpu) and the user name.
240 p Show the process activity accumulated per program (i.e. process
242 name).
243 Per program the following fields are shown: number of processes
245 active or terminated during last interval (or in total if combined
246 with command `a'), accumulated cpu consumption during last inter‐
247 val in system and user mode, the current virtual and resident mem‐
248 ory space consumed by active processes (or all processes of the
249 user if combined with command `a').
250 When "storage accounting" is active in the kernel, the accumulated
251 read and write throughput on disk is shown. When the kernel mod‐
252 ule `netatop' has been installed, the number of received and sent
253 network packets are shown.
254 The last columns contain the accumulated occupation percentage for
255 the chosen resource (default: cpu) and the program name.
256 j Show the process activity accumulated per Docker container.
258 Per container the following fields are shown: number of processes
260 active or terminated during last interval (or in total if combined
261 with command `a'), accumulated cpu consumption during last inter‐
262 val in system and user mode, the current virtual and resident mem‐
263 ory space consumed by active processes (or all processes of the
264 user if combined with command `a').
265 When "storage accounting" is active in the kernel, the accumulated
266 read and write throughput on disk is shown. When the kernel mod‐
267 ule `netatop' has been installed, the number of received and sent
268 network packets are shown.
269 The last columns contain the accumulated occupation percentage for
270 the chosen resource (default: cpu) and the Docker container id
271 (CID).
272 C Sort the current list in the order of cpu consumption (default).
274 The one-but-last column changes to ``CPU''.
275 M Sort the current list in the order of resident memory consumption.
277 The one-but-last column changes to ``MEM''. In case of sorting on
278 memory, the full process list will be shown (not only the active
279 processes).
280 D Sort the current list in the order of disk accesses issued. The
282 one-but-last column changes to ``DSK''.
283 N Sort the current list in the order of network bandwidth (received
285 and transmitted). The one-but-last column changes to ``NET''.
286 A Sort the current list automatically in the order of the most busy
288 system resource during this interval. The one-but-last column
289 shows either ``ACPU'', ``AMEM'', ``ADSK'' or ``ANET'' (the preced‐
290 ing 'A' indicates automatic sorting-order). The most busy
291 resource is determined by comparing the weighted busy-percentages
292 of the system resources, as described earlier in the section COL‐
293 ORS.
294 This option remains valid until another sorting-order is explic‐
295 itly selected again.
296 A sorting-order for disk is only possible when "storage account‐
297 ing" is active. A sorting-order for network is only possible when
298 the kernel module `netatop' is loaded.
299 Miscellaneous interactive commands:
301 ? Request for help information (also the key 'h' can be pressed).
303 V Request for version information (version number and date).
305 R Gather and calculate the proportional set size of processes (tog‐
307 gle). Gathering of all values that are needed to calculate the
308 PSIZE of a process is a relatively time-consuming task, so this
309 key should only be active when analyzing the resident memory con‐
310 sumption of processes.
311 x Suppress colors to highlight critical resources (toggle).
313 Whether this key is active or not can be seen in the header line.
314 z The pause key can be used to freeze the current situation in order
316 to investigate the output on the screen. While pcp-atop is paused,
317 the keys described above can be pressed to show other information
318 about the current list of processes. Whenever the pause key is
319 pressed again, pcp-atop will continue with a next sample.
320 i Modify the interval timer (default: 10 seconds). If an interval
322 timer of 0 is entered, the interval timer is switched off. In that
323 case a new sample can only be triggered manually by pressing the
324 key 't'.
325 t Trigger a new sample manually. This key can be pressed if the cur‐
327 rent sample should be finished before the timer has exceeded, or
328 if no timer is set at all (interval timer defined as 0). In the
329 latter case pcp-atop can be used as a stopwatch to measure the
330 load being caused by a particular application transaction, without
331 knowing on beforehand how many seconds this transaction will last.
332 When viewing the contents of a raw file, this key can be used to
334 show the next sample from the file.
335 T When viewing the contents of a raw file, this key can be used to
337 show the previous sample from the file.
338 b When viewing the contents of a raw file, this key can be used to
340 branch to a certain timestamp within the file (either forward or
341 backward).
342 r Reset all counters to zero to see the system and process activity
344 since boot again.
345 When viewing the contents of a raw file, this key can be used to
347 rewind to the beginning of the file again.
348 U Specify a search string for specific user names as a regular
350 expression. From now on, only (active) processes will be shown
351 from a user which matches the regular expression. The system sta‐
352 tistics are still system wide. If the Enter-key is pressed with‐
353 out specifying a name, (active) processes of all users will be
354 shown again.
355 Whether this key is active or not can be seen in the header line.
356 I Specify a list with one or more PIDs to be selected. From now on,
358 only processes will be shown with a PID which matches one of the
359 given list. The system statistics are still system wide. If the
360 Enter-key is pressed without specifying a PID, all (active) pro‐
361 cesses will be shown again.
362 Whether this key is active or not can be seen in the header line.
363 P Specify a search string for specific process names as a regular
365 expression. From now on, only processes will be shown with a name
366 which matches the regular expression. The system statistics are
367 still system wide. If the Enter-key is pressed without specifying
368 a name, all (active) processes will be shown again.
369 Whether this key is active or not can be seen in the header line.
370 / Specify a specific command line search string as a regular expres‐
372 sion. From now on, only processes will be shown with a command
373 line which matches the regular expression. The system statistics
374 are still system wide. If the Enter-key is pressed without speci‐
375 fying a string, all (active) processes will be shown again.
376 Whether this key is active or not can be seen in the header line.
377 J Specify a Docker container id of 12 (hexadecimal) characters.
379 From now on, only processes will be shown that run in that spe‐
380 cific Docker container (CID). The system statistics are still
381 system wide. If the Enter-key is pressed without specifying a
382 container id, all (active) processes will be shown again.
383 Whether this key is active or not can be seen in the header line.
384 S Specify search strings for specific logical volume names, specific
386 disk names and specific network interface names. All search
387 strings are interpreted as a regular expressions. From now on,
388 only those system resources are shown that match the concerning
389 regular expression. If the Enter-key is pressed without specify‐
390 ing a search string, all (active) system resources of that type
391 will be shown again.
392 Whether this key is active or not can be seen in the header line.
393 a The `all/active' key can be used to toggle between only show‐
395 ing/accumulating the processes that were active during the last
396 interval (default) or showing/accumulating all processes.
397 Whether this key is active or not can be seen in the header line.
398 G By default, pcp-atop shows/accumulates the processes that are
400 alive and the processes that are exited during the last interval.
401 With this key (toggle), showing/accumulating the processes that
402 are exited can be suppressed.
403 Whether this key is active or not can be seen in the header line.
404 f Show a fixed (maximum) number of header lines for system resources
406 (toggle). By default only the lines are shown about system
407 resources (CPUs, paging, logical volumes, disks, network inter‐
408 faces) that really have been active during the last interval.
409 With this key you can force pcp-atop to show lines of inactive
410 resources as well.
411 Whether this key is active or not can be seen in the header line.
412 F Suppress sorting of system resources (toggle). By default system
414 resources (CPUs, logical volumes, disks, network interfaces) are
415 sorted on utilization.
416 Whether this key is active or not can be seen in the header line.
417 1 Show relevant counters as an average per second (in the format
419 `..../s') instead of as a total during the interval (toggle).
420 Whether this key is active or not can be seen in the header line.
421 l Limit the number of system level lines for the counters per-cpu,
423 the active disks and the network interfaces. By default lines are
424 shown of all CPUs, disks and network interfaces which have been
425 active during the last interval. Limiting these lines can be use‐
426 ful on systems with huge number CPUs, disks or interfaces in order
427 to be able to run pcp-atop on a screen/window with e.g. only 24
428 lines.
429 For all mentioned resources the maximum number of lines can be
430 specified interactively. When using the flag -l the maximum number
431 of per-cpu lines is set to 0, the maximum number of disk lines to
432 5 and the maximum number of interface lines to 3. These values
433 can be modified again in interactive mode.
434 k Send a signal to an active process (a.k.a. kill a process).
436 q Quit the program.
438 PgDn Show the next page of the process/thread list.
440 With the arrow-down key the list can be scrolled downwards with
441 single lines.
442 ^F Show the next page of the process/thread list (forward).
444 With the arrow-down key the list can be scrolled downwards with
445 single lines.
446 PgUp Show the previous page of the process/thread list.
448 With the arrow-up key the list can be scrolled upwards with single
449 lines.
450 ^B Show the previous page of the process/thread list (backward).
452 With the arrow-up key the list can be scrolled upwards with single
453 lines.
454 ^L Redraw the screen.
456
458 In order to store system and process level statistics for long-term
459 analysis (e.g. to check the system load and the active processes run‐
460 ning yesterday between 3:00 and 4:00 PM), pcp-atop can store the system
461 and process level statistics in the PCP archive format, as an archive
462 folio (see mkaf(1)).
463 All processes/threads are stored in the raw file.
464 The interval (default: 10 seconds) and number of samples (default:
465 infinite) can be passed as last arguments. Instead of the number of
466 samples, the flag -S can be used to indicate that pcp-atop should fin‐
467 ish anyhow before midnight.
468 A PCP archive can be read and visualized again with the flag -r . The
470 argument is a comma-separated list of names, each of which may be the
471 base name of an archive or the name of a directory containing one or
472 more archives. If no argument is specified, the file
473 $PCP_LOG_DIR/pmlogger/HOST/YYYYMMDD is opened for input (where YYYYMMDD
474 are digits representing the current date, and HOST is the hostname of
475 the machine being logged). If a filename is specified in the format
476 YYYYMMDD (representing any valid date), the file $PCP_LOG_DIR/pmlog‐
477 ger/HOST/YYYYMMDD is opened. If a filename with the symbolic name y is
478 specified, yesterday's daily logfile is opened (this can be repeated so
479 'yyyy' indicates the logfile of four days ago).
480 The samples from the file can be viewed interactively by using the key
481 't' to show the next sample, the key 'T' to show the previous sample,
482 the key 'b' to branch to a particular time or the key 'r' to rewind to
483 the begin of the file.
484 When output is redirected to a file or pipe, pcp-atop prints all sam‐
485 ples in plain ASCII. The default line length is 80 characters in that
486 case; with the flag -L followed by an alternate line length, more (or
487 less) columns will be shown.
488 With the flag -b (begin time) and/or -e (end time) followed by a time
489 argument of the form HH:MM, a certain time period within the raw file
490 can be selected.
491
493 The first sample shows the system level activity since boot (the
494 elapsed time in the header shows the time since boot). Note that par‐
495 ticular counters could have reached their maximum value (several times)
496 and started by zero again, so do not rely on these figures.
497 For every sample pcp-atop first shows the lines related to system level
499 activity. If a particular system resource has not been used during the
500 interval, the entire line related to this resource is suppressed. So
501 the number of system level lines may vary for each sample.
502 After that a list is shown of processes which have been active during
503 the last interval. This list is by default sorted on cpu consumption,
504 but this order can be changed by the keys which are previously
505 described.
506 If values have to be shown by pcp-atop which do not fit in the column
508 width, another format is used. If e.g. a cpu-consumption of 233216 mil‐
509 liseconds should be shown in a column width of 4 positions, it is shown
510 as `233s' (in seconds). For large memory figures, another unit is cho‐
511 sen if the value does not fit (Mb instead of Kb, Gb instead of Mb, Tb
512 instead of Gb, ...). For other values, a kind of exponent notation is
513 used (value 123456789 shown in a column of 5 positions gives 123e6).
514
516 The system level information consists of the following output lines:
517 PRC Process and thread level totals.
519 This line contains the total cpu time consumed in system mode
520 (`sys') and in user mode (`user'), the total number of processes
521 present at this moment (`#proc'), the total number of threads
522 present at this moment in state `running' (`#trun'), `sleeping
523 interruptible' (`#tslpi') and `sleeping uninterruptible'
524 (`#tslpu'), the number of zombie processes (`#zombie'), the number
525 of clone system calls (`clones'), and the number of processes that
526 ended during the interval (`#exit') when process accounting is
527 used. Instead of `#exit` the last column may indicate that process
528 accounting could not be activated (`no procacct`).
529 If the screen-width does not allow all of these counters, only a
530 relevant subset is shown.
531 CPU CPU utilization.
533 At least one line is shown for the total occupation of all CPUs
534 together.
535 In case of a multi-processor system, an additional line is shown
536 for every individual processor (with `cpu' in lower case), sorted
537 on activity. Inactive CPUs will not be shown by default. The
538 lines showing the per-cpu occupation contain the cpu number in the
539 last field.
540 Every line contains the percentage of cpu time spent in kernel
542 mode by all active processes (`sys'), the percentage of cpu time
543 consumed in user mode (`user') for all active processes (including
544 processes running with a nice value larger than zero), the per‐
545 centage of cpu time spent for interrupt handling (`irq') including
546 softirq, the percentage of unused cpu time while no processes were
547 waiting for disk-I/O (`idle'), and the percentage of unused cpu
548 time while at least one process was waiting for disk-I/O (`wait').
549 In case of per-cpu occupation, the last column shows the cpu num‐
550 ber and the wait percentage (`w') for that cpu. The number of
551 lines showing the per-cpu occupation can be limited.
552 For virtual machines the steal-percentage is shown (`steal'),
554 reflecting the percentage of cpu time stolen by other virtual
555 machines running on the same hardware.
556 For physical machines hosting one or more virtual machines, the
557 guest-percentage is shown (`guest'), reflecting the percentage of
558 cpu time used by the virtual machines. Notice that this percentage
559 overlaps the user-percentage.
560 In case of frequency-scaling, all previously mentioned CPU-per‐
562 centages are relative to the used scaling of the CPU during the
563 interval. If a CPU has been active for e.g. 50% in user mode dur‐
564 ing the interval while the frequency-scaling of that CPU was 40%,
565 only 20% of the full capacity of the CPU has been used in user
566 mode.
567 If the screen-width does not allow all of these counters, only a
569 relevant subset is shown.
570 CPL CPU load information.
572 This line contains the load average figures reflecting the number
573 of threads that are available to run on a CPU (i.e. part of the
574 runqueue) or that are waiting for disk I/O. These figures are
575 averaged over 1 (`avg1'), 5 (`avg5') and 15 (`avg15') minutes.
576 Furthermore the number of context switches (`csw'), the number of
577 serviced interrupts (`intr') and the number of available CPUs are
578 shown.
579 If the screen-width does not allow all of these counters, only a
581 relevant subset is shown.
582 MEM Memory occupation.
584 This line contains the total amount of physical memory (`tot'),
585 the amount of memory which is currently free (`free'), the amount
586 of memory in use as page cache including the total resident shared
587 memory (`cache'), the amount of memory within the page cache that
588 has to be flushed to disk (`dirty'), the amount of memory used for
589 filesystem meta data (`buff'), the amount of memory being used for
590 kernel mallocs (`slab'), the amount of slab memory that is
591 reclaimable (`slrec'), the resident size of shared memory includ‐
592 ing tmpfs (`shmem`), the resident size of shared memory (`shrss`)
593 the amount of shared memory that is currently swapped (`shswp`),
594 the amount of memory that is currently claimed by vmware's balloon
595 driver (`vmbal`), the amount of memory that is claimed for huge
596 pages (`hptot`), and the amount of huge page memory that is really
597 in use (`hpuse`).
598 If the screen-width does not allow all of these counters, only a
600 relevant subset is shown.
601 SWP Swap occupation and overcommit info.
603 This line contains the total amount of swap space on disk (`tot')
604 and the amount of free swap space (`free').
605 Furthermore the committed virtual memory space (`vmcom') and the
606 maximum limit of the committed space (`vmlim', which is by default
607 swap size plus 50% of memory size) is shown. The committed space
608 is the reserved virtual space for all allocations of private mem‐
609 ory space for processes. The kernel only verifies whether the com‐
610 mitted space exceeds the limit if strict overcommit handling is
611 configured (vm.overcommit_memory is 2).
612 PAG Paging frequency.
614 This line contains the number of scanned pages (`scan') due to the
615 fact that free memory drops below a particular threshold and the
616 number times that the kernel tries to reclaim pages due to an
617 urgent need (`stall').
618 Also the number of memory pages the system read from swap space
619 (`swin') and the number of memory pages the system wrote to swap
620 space (`swout') are shown.
621 LVM/MDD/DSK
623 Logical volume/multiple device/disk utilization.
624 Per active unit one line is produced, sorted on unit activity.
625 Such line shows the name (e.g. VolGroup00-lvtmp for a logical vol‐
626 ume or sda for a hard disk), the busy percentage i.e. the portion
627 of time that the unit was busy handling requests (`busy'), the
628 number of read requests issued (`read'), the number of write
629 requests issued (`write'), the number of KiBytes per read
630 (`KiB/r'), the number of KiBytes per write (`KiB/w'), the number
631 of MiBytes per second throughput for reads (`MBr/s'), the number
632 of MiBytes per second throughput for writes (`MBw/s'), the average
633 queue depth (`avq') and the average number of milliseconds needed
634 by a request (`avio') for seek, latency and data transfer.
635 If the screen-width does not allow all of these counters, only a
636 relevant subset is shown.
637 The number of lines showing the units can be limited per class
639 (LVM, MDD or DSK) with the 'l' key or statically (see separate
640 man-page of pcp-atoprc(5)). By specifying the value 0 for a par‐
641 ticular class, no lines will be shown any more for that class.
642 NFM Network Filesystem (NFS) mount at the client side.
644 For each NFS-mounted filesystem, a line is shown that contains the
645 mounted server directory, the name of the server (`srv'), the
646 total number of bytes physically read from the server (`read') and
647 the total number of bytes physically written to the server
648 (`write'). Data transfer is subdivided in the number of bytes
649 read via normal read() system calls (`nread'), the number of bytes
650 written via normal read() system calls (`nwrit'), the number of
651 bytes read via direct I/O (`dread'), the number of bytes written
652 via direct I/O (`dwrit'), the number of bytes read via memory
653 mapped I/O pages (`mread'), and the number of bytes written via
654 memory mapped I/O pages (`mwrit').
655 NFC Network Filesystem (NFS) client side counters.
657 This line contains the number of RPC calls issues by local pro‐
658 cesses (`rpc'), the number of read RPC calls (`read`) and write
659 RPC calls (`rpwrite') issued to the NFS server, the number of RPC
660 calls being retransmitted (`retxmit') and the number of authoriza‐
661 tion refreshes (`autref').
662 NFS Network Filesystem (NFS) server side counters.
664 This line contains the number of RPC calls received from NFS
665 clients (`rpc'), the number of read RPC calls received (`cread`),
666 the number of write RPC calls received (`cwrit'), the number of
667 network requests handled via TCP (`nettcp'), the number of network
668 requests handled via UDP (`netudp'), the number of Megabytes/sec‐
669 ond returned to read requests by clients (`MBcr/s`), the number of
670 Megabytes/second passed in write requests by clients (`MBcw/s`),
671 the number of reply cache hits (`rchits'), the number of reply
672 cache misses (`rcmiss') and the number of uncached requests
673 (`rcnoca'). Furthermore some error counters indicating the number
674 of requests with a bad format (`badfmt') or a bad authorization
675 (`badaut'), and a counter indicating the number of bad clients
676 (`badcln'). and the number of authorization refreshes (`autref').
677 NET Network utilization (TCP/IP).
679 One line is shown for activity of the transport layer (TCP and
680 UDP), one line for the IP layer and one line per active interface.
681 For the transport layer, counters are shown concerning the number
682 of received TCP segments including those received in error
683 (`tcpi'), the number of transmitted TCP segments excluding those
684 containing only retransmitted octets (`tcpo'), the number of UDP
685 datagrams received (`udpi'), the number of UDP datagrams transmit‐
686 ted (`udpo'), the number of active TCP opens (`tcpao'), the number
687 of passive TCP opens (`tcppo'), the number of TCP output retrans‐
688 missions (`tcprs'), the number of TCP input errors (`tcpie'), the
689 number of TCP output resets (`tcpor'), the number of UDP no ports
690 (`udpnp'), and the number of UDP input errors (`udpie').
691 If the screen-width does not allow all of these counters, only a
692 relevant subset is shown.
693 These counters are related to IPv4 and IPv6 combined.
694 For the IP layer, counters are shown concerning the number of IP
696 datagrams received from interfaces, including those received in
697 error (`ipi'), the number of IP datagrams that local higher-layer
698 protocols offered for transmission (`ipo'), the number of received
699 IP datagrams which were forwarded to other interfaces (`ipfrw'),
700 the number of IP datagrams which were delivered to local higher-
701 layer protocols (`deliv'), the number of received ICMP datagrams
702 (`icmpi'), and the number of transmitted ICMP datagrams (`icmpo').
703 If the screen-width does not allow all of these counters, only a
704 relevant subset is shown.
705 These counters are related to IPv4 and IPv6 combined.
706 For every active network interface one line is shown, sorted on
708 the interface activity. Such line shows the name of the interface
709 and its busy percentage in the first column. The busy percentage
710 for half duplex is determined by comparing the interface speed
711 with the number of bits transmitted and received per second; for
712 full duplex the interface speed is compared with the highest of
713 either the transmitted or the received bits. When the interface
714 speed can not be determined (e.g. for the loopback interface),
715 `---' is shown instead of the percentage.
716 Furthermore the number of received packets (`pcki'), the number of
717 transmitted packets (`pcko'), the line speed of the interface
718 (`sp'), the effective amount of bits received per second (`si'),
719 the effective amount of bits transmitted per second (`so'), the
720 number of collisions (`coll'), the number of received multicast
721 packets (`mlti'), the number of errors while receiving a packet
722 (`erri'), the number of errors while transmitting a packet
723 (`erro'), the number of received packets dropped (`drpi'), and the
724 number of transmitted packets dropped (`drpo').
725 If the screen-width does not allow all of these counters, only a
726 relevant subset is shown.
727 The number of lines showing the network interfaces can be limited.
728
730 Following the system level information, the processes are shown from
731 which the resource utilization has changed during the last interval.
732 These processes might have used cpu time or issued disk or network
733 requests. However a process is also shown if part of it has been paged
734 out due to lack of memory (while the process itself was in sleep
735 state).
736 Per process the following fields may be shown (in alphabetical order),
738 depending on the current output mode as described in the section INTER‐
739 ACTIVE COMMANDS and depending on the current width of your window:
740 AVGRSZ The average size of one read-action on disk.
742 AVGWSZ The average size of one write-action on disk.
744 CID Container ID (Docker) of 12 hexadecimal digits, referring to
746 the container in which the process/thread is running. If a
747 process has been started and finished during the last inter‐
748 val, a `?' is shown because the container ID is not part of
749 the standard process accounting record.
750 CMD The name of the process. This name can be surrounded by
752 "less/greater than" signs (`<name>') which means that the
753 process has finished during the last interval.
754 Behind the abbreviation `CMD' in the header line, the current
755 page number and the total number of pages of the
756 process/thread list are shown.
757 COMMAND-LINE
759 The full command line of the process (including arguments). If
760 the length of the command line exceeds the length of the
761 screen line, the arrow keys -> and <- can be used for horizon‐
762 tal scroll.
763 Behind the verb `COMMAND-LINE' in the header line, the current
764 page number and the total number of pages of the
765 process/thread list are shown.
766 CPU The occupation percentage of this process related to the
768 available capacity for this resource on system level.
769 CPUNR The identification of the CPU the (main) thread is running on
771 or has recently been running on.
772 CTID Container ID (OpenVZ). If a process has been started and fin‐
774 ished during the last interval, a `?' is shown because the
775 container ID is not part of the standard process accounting
776 record.
777 DSK The occupation percentage of this process related to the total
779 load that is produced by all processes (i.e. total disk
780 accesses by all processes during the last interval).
781 This information is shown when per process "storage account‐
782 ing" is active in the kernel.
783 EGID Effective group-id under which this process executes.
785 ENDATE Date that the process has been finished. If the process is
787 still running, this field shows `active'.
788 ENTIME Time that the process has been finished. If the process is
790 still running, this field shows `active'.
791 ENVID Virtual environment identified (OpenVZ only).
793 EUID Effective user-id under which this process executes.
795 EXC The exit code of a terminated process (second position of col‐
797 umn `ST' is E) or the fatal signal number (second position of
798 column `ST' is S or C).
799 FSGID Filesystem group-id under which this process executes.
801 FSUID Filesystem user-id under which this process executes.
803 MAJFLT The number of page faults issued by this process that have
805 been solved by creating/loading the requested memory page.
806 MEM The occupation percentage of this process related to the
808 available capacity for this resource on system level.
809 MINFLT The number of page faults issued by this process that have
811 been solved by reclaiming the requested memory page from the
812 free list of pages.
813 NET The occupation percentage of this process related to the total
815 load that is produced by all processes (i.e. consumed network
816 bandwidth of all processes during the last interval).
817 This information will only be shown when kernel module
818 `netatop' is loaded.
819 NICE The more or less static priority that can be given to a
821 process on a scale from -20 (high priority) to +19 (low prior‐
822 ity).
823 NPROCS The number of active and terminated processes accumulated for
825 this user or program.
826 PID Process-id.
828 POLI The policies 'norm' (normal, which is SCHED_OTHER), 'btch'
830 (batch) and 'idle' refer to timesharing processes. The poli‐
831 cies 'fifo' (SCHED_FIFO) and 'rr' (round robin, which is
832 SCHED_RR) refer to realtime processes.
833 PPID Parent process-id.
835 PRI The process' priority ranges from 0 (highest priority) to 139
837 (lowest priority). Priority 0 to 99 are used for realtime pro‐
838 cesses (fixed priority independent of their behavior) and pri‐
839 ority 100 to 139 for timesharing processes (variable priority
840 depending on their recent CPU consumption and the nice value).
841 PSIZE The proportional memory size of this process (or user).
843 Every process shares resident memory with other processes.
844 E.g. when a particular program is started several times, the
845 code pages (text) are only loaded once in memory and shared by
846 all incarnations. Also the code of shared libraries is shared
847 by all processes using that shared library, as well as shared
848 memory and memory-mapped files. For the PSIZE calculation of
849 a process, the resident memory of a process that is shared
850 with other processes is divided by the number of sharers.
851 This means, that every process is accounted for a proportional
852 part of that memory. Accumulating the PSIZE values of all pro‐
853 cesses in the system gives a reliable impression of the total
854 resident memory consumed by all processes.
855 Since gathering of all values that are needed to calculate the
856 PSIZE is a relatively time-consuming task, the 'R' key (or
857 '-R' flag) should be active. Gathering these values also
858 requires superuser privileges (otherwise '?K' is shown in the
859 output).
860 RDDSK When the kernel maintains standard io statistics (>= 2.6.20):
862 The read data transfer issued physically on disk (so reading
863 from the disk cache is not accounted for).
864 Unfortunately, the kernel aggregates the data tranfer of a
865 process to the data transfer of its parent process when termi‐
866 nating, so you might see transfers for (parent) processes like
867 cron, bash or init, that are not really issued by them.
868 RGID The real group-id under which the process executes.
870 RGROW The amount of resident memory that the process has grown dur‐
872 ing the last interval. A resident growth can be caused by
873 touching memory pages which were not physically created/loaded
874 before (load-on-demand). Note that a resident growth can also
875 be negative e.g. when part of the process is paged out due to
876 lack of memory or when the process frees dynamically allocated
877 memory. For a process which started during the last interval,
878 the resident growth reflects the total resident size of the
879 process at that moment.
880 RSIZE The total resident memory usage consumed by this process (or
882 user). Notice that the RSIZE of a process includes all resi‐
883 dent memory used by that process, even if certain memory parts
884 are shared with other processes (see also the explanation of
885 PSIZE).
886 RTPR Realtime priority according the POSIX standard. Value can be
888 0 for a timesharing process (policy 'norm', 'btch' or 'idle')
889 or ranges from 1 (lowest) till 99 (highest) for a realtime
890 process (policy 'rr' or 'fifo').
891 RUID The real user-id under which the process executes.
893 S The current state of the (main) thread: `R' for running (cur‐
895 rently processing or in the runqueue), `S' for sleeping inter‐
896 ruptible (wait for an event to occur), `D' for sleeping non-
897 interruptible, `Z' for zombie (waiting to be synchronized with
898 its parent process), `T' for stopped (suspended or traced),
899 `W' for swapping, and `E' (exit) for processes which have fin‐
900 ished during the last interval.
901 SGID The saved group-id of the process.
903 ST The status of a process.
905 The first position indicates if the process has been started
906 during the last interval (the value N means 'new process').
907 The second position indicates if the process has been finished
909 during the last interval.
910 The value E means 'exit' on the process' own initiative; the
911 exit code is displayed in the column `EXC'.
912 The value S means that the process has been terminated unvol‐
913 untarily by a signal; the signal number is displayed in the in
914 the column `EXC'.
915 The value C means that the process has been terminated unvol‐
916 untarily by a signal, producing a core dump in its current
917 directory; the signal number is displayed in the column `EXC'.
918 STDATE The start date of the process.
920 STTIME The start time of the process.
922 SUID The saved user-id of the process.
924 SWAPSZ The swap space consumed by this process (or user).
926 SYSCPU CPU time consumption of this process in system mode (kernel
928 mode), usually due to system call handling.
929 THR Total number of threads within this process. All related
931 threads are contained in a thread group, represented by pcp-
932 atop as one line or as a separate line when the 'y' key (or -y
933 flag) is active.
934 On Linux 2.4 systems it is hardly possible to determine which
936 threads (i.e. processes) are related to the same thread group.
937 Every thread is represented by pcp-atop as a separate line.
938 TID Thread-id. All threads within a process run with the same PID
940 but with a different TID. This value is shown for individual
941 threads in multi-threaded processes (when using the key 'y').
942 TRUN Number of threads within this process that are in the state
944 'running' (R).
945 TSLPI Number of threads within this process that are in the state
947 'interruptible sleeping' (S).
948 TSLPU Number of threads within this process that are in the state
950 'uninterruptible sleeping' (D).
951 USRCPU CPU time consumption of this process in user mode, due to pro‐
953 cessing the own program text.
954 VDATA The virtual memory size of the private data used by this
956 process (including heap and shared library data).
957 VGROW The amount of virtual memory that the process has grown during
959 the last interval. A virtual growth can be caused by e.g.
960 issueing a malloc() or attaching a shared memory segment. Note
961 that a virtual growth can also be negative by e.g. issueing a
962 free() or detaching a shared memory segment. For a process
963 which started during the last interval, the virtual growth
964 reflects the total virtual size of the process at that moment.
965 VPID Virtual process-id (within an OpenVZ container). If a process
967 has been started and finished during the last interval, a `?'
968 is shown because the virtual process-id is not part of the
969 standard process accounting record.
970 VSIZE The total virtual memory usage consumed by this process (or
972 user).
973 VSLIBS The virtual memory size of the (shared) text of all shared
975 libraries used by this process.
976 VSTACK The virtual memory size of the (private) stack used by this
978 process
979 VSTEXT The virtual memory size of the (shared) text of the executable
981 program.
982 WRDSK When the kernel maintains standard io statistics (>= 2.6.20):
984 The write data transfer issued physically on disk (so writing
985 to the disk cache is not accounted for). This counter is
986 maintained for the application process that writes its data to
987 the cache (assuming that this data is physically transferred
988 to disk later on). Notice that disk I/O needed for swapping is
989 not taken into account.
990 Unfortunately, the kernel aggregates the data tranfer of a
991 process to the data transfer of its parent process when termi‐
992 nating, so you might see transfers for (parent) processes like
993 cron, bash or init, that are not really issued by them.
994 WCANCL When the kernel maintains standard io statistics (>= 2.6.20):
996 The write data transfer previously accounted for this process
997 or another process that has been cancelled. Suppose that a
998 process writes new data to a file and that data is removed
999 again before the cache buffers have been flushed to disk.
1000 Then the original process shows the written data as WRDSK,
1001 while the process that removes/truncates the file shows the
1002 unflushed removed data as WCANCL.
1003
1005 With the flag -P followed by a list of one or more labels (comma-sepa‐
1006 rated), parseable output is produced for each sample. The labels that
1007 can be specified for system-level statistics correspond to the labels
1008 (first verb of each line) that can be found in the interactive output:
1009 "CPU", "cpu" "CPL" "MEM", "SWP", "PAG", "LVM", "MDD", "DSK", "NFM",
1010 "NFC", "NFS" and "NET".
1011 For process-level statistics special labels are introduced: "PRG" (gen‐
1012 eral), "PRC" (cpu), "PRM" (memory), "PRD" (disk, only if "storage
1013 accounting" is active) and "PRN" (network, only if the kernel module
1014 'netatop' has been installed).
1015 With the label "ALL", all system and process level statistics are
1016 shown.
1017 For every interval all requested lines are shown whereafter pcp-atop
1019 shows a line just containing the label "SEP" as a separator before the
1020 lines for the next sample are generated.
1021 When a sample contains the values since boot, pcp-atop shows a line
1022 just containing the label "RESET" before the lines for this sample are
1023 generated.
1024 The first part of each output-line consists of the following six
1026 fields: label (the name of the label), host (the name of this machine),
1027 epoch (the time of this interval as number of seconds since 1-1-1970),
1028 date (date of this interval in format YYYY/MM/DD), time (time of this
1029 interval in format HH:MM:SS), and interval (number of seconds elapsed
1030 for this interval).
1031 The subsequent fields of each output-line depend on the label:
1033 CPU Subsequent fields: total number of clock-ticks per second for
1035 this machine, number of processors, consumption for all CPUs
1036 in system mode (clock-ticks), consumption for all CPUs in user
1037 mode (clock-ticks), consumption for all CPUs in user mode for
1038 niced processes (clock-ticks), consumption for all CPUs in
1039 idle mode (clock-ticks), consumption for all CPUs in wait mode
1040 (clock-ticks), consumption for all CPUs in irq mode (clock-
1041 ticks), consumption for all CPUs in softirq mode (clock-
1042 ticks), consumption for all CPUs in steal mode (clock-ticks),
1043 consumption for all CPUs in guest mode (clock-ticks) overlap‐
1044 ping user mode, frequency of all CPUs and frequency percentage
1045 of all CPUs.
1046 cpu Subsequent fields: total number of clock-ticks per second for
1048 this machine, processor-number, consumption for this CPU in
1049 system mode (clock-ticks), consumption for this CPU in user
1050 mode (clock-ticks), consumption for this CPU in user mode for
1051 niced processes (clock-ticks), consumption for this CPU in
1052 idle mode (clock-ticks), consumption for this CPU in wait mode
1053 (clock-ticks), consumption for this CPU in irq mode (clock-
1054 ticks), consumption for this CPU in softirq mode (clock-
1055 ticks), consumption for this CPU in steal mode (clock-ticks),
1056 consumption for this CPU in guest mode (clock-ticks) overlap‐
1057 ping user mode, frequency of this CPU and frequency percentage
1058 of this CPU.
1059 CPL Subsequent fields: number of processors, load average for last
1061 minute, load average for last five minutes, load average for
1062 last fifteen minutes, number of context-switches, and number
1063 of device interrupts.
1064 MEM Subsequent fields: page size for this machine (in bytes), size
1066 of physical memory (pages), size of free memory (pages), size
1067 of page cache (pages), size of buffer cache (pages), size of
1068 slab (pages), dirty pages in cache (pages), reclaimable part
1069 of slab (pages), total size of vmware's balloon pages (pages),
1070 total size of shared memory (pages), size of resident shared
1071 memory (pages), size of swapped shared memory (pages), huge
1072 page size (in bytes), total size of huge pages (huge pages),
1073 and size of free huge pages (huge pages).
1074 SWP Subsequent fields: page size for this machine (in bytes), size
1076 of swap (pages), size of free swap (pages), 0 (future use),
1077 size of committed space (pages), and limit for committed space
1078 (pages).
1079 PAG Subsequent fields: page size for this machine (in bytes), num‐
1081 ber of page scans, number of allocstalls, 0 (future use), num‐
1082 ber of swapins, and number of swapouts.
1083 LVM/MDD/DSK
1085 For every logical volume/multiple device/hard disk one line is
1086 shown.
1087 Subsequent fields: name, number of milliseconds spent for I/O,
1088 number of reads issued, number of sectors transferred for
1089 reads, number of writes issued, and number of sectors trans‐
1090 ferred for write.
1091 NFM Subsequent fields: mounted NFS filesystem, total number of
1093 bytes read, total number of bytes written, number of bytes
1094 read by normal system calls, number of bytes written by normal
1095 system calls, number of bytes read by direct I/O, number of
1096 bytes written by direct I/O, number of pages read by memory-
1097 mapped I/O, and number of pages written by memory-mapped I/O.
1098 NFC Subsequent fields: number of transmitted RPCs, number of
1100 transmitted read RPCs, number of transmitted write RPCs, num‐
1101 ber of RPC retransmissions, and number of authorization
1102 refreshes.
1103 NFS Subsequent fields: number of handled RPCs, number of received
1105 read RPCs, number of received write RPCs, number of bytes read
1106 by clients, number of bytes written by clients, number of RPCs
1107 with bad format, number of RPCs with bad authorization, number
1108 of RPCs from bad client, total number of handled network
1109 requests, number of handled network requests via TCP, number
1110 of handled network requests via UDP, number of handled TCP
1111 connections, number of hits on reply cache, number of misses
1112 on reply cache, and number of uncached requests.
1113 NET First one line is produced for the upper layers of the TCP/IP
1115 stack.
1116 Subsequent fields: the verb "upper", number of packets
1117 received by TCP, number of packets transmitted by TCP, number
1118 of packets received by UDP, number of packets transmitted by
1119 UDP, number of packets received by IP, number of packets
1120 transmitted by IP, number of packets delivered to higher lay‐
1121 ers by IP, and number of packets forwarded by IP.
1122 Next one line is shown for every interface.
1124 Subsequent fields: name of the interface, number of packets
1125 received by the interface, number of bytes received by the
1126 interface, number of packets transmitted by the interface,
1127 number of bytes transmitted by the interface, interface speed,
1128 and duplex mode (0=half, 1=full).
1129 PRG For every process one line is shown.
1131 Subsequent fields: PID (unique ID of task), name (between
1132 brackets), state, real uid, real gid, TGID (group number of
1133 related tasks/threads), total number of threads, exit code,
1134 start time (epoch), full command line (between brackets),
1135 PPID, number of threads in state 'running' (R), number of
1136 threads in state 'interruptible sleeping' (S), number of
1137 threads in state 'uninterruptible sleeping' (D), effective
1138 uid, effective gid, saved uid, saved gid, filesystem uid,
1139 filesystem gid, elapsed time (hertz), is_process (y/n), OpenVZ
1140 virtual pid (VPID), OpenVZ container id (CTID) and Docker con‐
1141 tainer id (CID).
1142 PRC For every process one line is shown.
1144 Subsequent fields: PID, name (between brackets), state, total
1145 number of clock-ticks per second for this machine, CPU-con‐
1146 sumption in user mode (clockticks), CPU-consumption in system
1147 mode (clockticks), nice value, priority, realtime priority,
1148 scheduling policy, current CPU, sleep average, TGID (group
1149 number of related tasks/threads) and is_process (y/n).
1150 PRM For every process one line is shown.
1152 Subsequent fields: PID, name (between brackets), state, page
1153 size for this machine (in bytes), virtual memory size
1154 (Kbytes), resident memory size (Kbytes), shared text memory
1155 size (Kbytes), virtual memory growth (Kbytes), resident memory
1156 growth (Kbytes), number of minor page faults, number of major
1157 page faults, virtual library exec size (Kbytes), virtual data
1158 size (Kbytes), virtual stack size (Kbytes), swap space used
1159 (Kbytes), TGID (group number of related tasks/threads),
1160 is_process (y/n) and proportional set size (Kbytes) if in 'R'
1161 option is specified.
1162 PRD For every process one line is shown.
1164 Subsequent fields: PID, name (between brackets), state, obso‐
1165 leted kernel patch installed ('n'), standard io statistics
1166 used ('y' or 'n'), number of reads on disk, cumulative number
1167 of sectors read, number of writes on disk, cumulative number
1168 of sectors written, cancelled number of written sectors, TGID
1169 (group number of related tasks/threads) and is_process (y/n).
1170 If the standard I/O statistics (>= 2.6.20) are not used, the
1171 disk I/O counters per process are not relevant. The counters
1172 'number of reads on disk' and 'number of writes on disk' are
1173 obsoleted anyhow.
1174 PRN For every process one line is shown.
1176 Subsequent fields: PID, name (between brackets), state, kernel
1177 module 'netatop' loaded ('y' or 'n'), number of TCP-packets
1178 transmitted, cumulative size of TCP-packets transmitted, num‐
1179 ber of TCP-packets received, cumulative size of TCP-packets
1180 received, number of UDP-packets transmitted, cumulative size
1181 of UDP-packets transmitted, number of UDP-packets received,
1182 cumulative size of UDP-packets transmitted, number of raw
1183 packets transmitted (obsolete, always 0), number of raw pack‐
1184 ets received (obsolete, always 0), TGID (group number of
1185 related tasks/threads) and is_process (y/n).
1186
1188 To monitor the current system load interactively with an interval of 5
1189 seconds:
1190 pcp atop 5
1192 To monitor the system load and write it to a file (in plain ASCII) with
1194 an interval of one minute during half an hour with active processes
1195 sorted on memory consumption:
1196 pcp atop -M 60 30 > /log/pcp-atop.mem
1198 Store information about the system and process activity in a PCP ar‐
1200 chive folio with an interval of ten minutes during an hour:
1201 pcp atop -w /tmp/pcp-atop 600 6
1203 View the contents of this file interactively:
1205 pcp atop -r /tmp/pcp-atop
1207 View the processor and disk utilization of this file in parseable for‐
1209 mat:
1210 pcp atop -PCPU,DSK -r /tmp/pcp-atop.raw
1212 View the contents of today's standard logfile interactively:
1214 pcp atop -r
1216 View the contents of the standard logfile of the day before yesterday
1218 interactively:
1219 pcp atop -r yy
1221 View the contents of the standard logfile of 2014, June 7 from 02:00 PM
1223 onwards interactively:
1224 pcp atop -r 20140607 -b 14:00
1226
1228 /etc/atoprc
1229 Configuration file containing system-wide default values. See
1230 related man-page.
1231 ~/.atoprc
1233 Configuration file containing personal default values. See
1234 related man-page.
1235
1237 pcp-atop is based on the source code of the atop(1) command from
1238 http://atoptool.nl and aims to be command line and output compatible
1239 with it as much as possible. Some features of that atop command are
1240 not available in pcp-atop.
1241 Some features of pcp-atop (such as reporting on the Apache HTTP daemon,
1243 and NFS client mounts) are only activated if the corresonding PCP met‐
1244 rics are available. Refer to the documentation for pmdaapache(1) and
1245 pmdanfsclient(1) for further details on activating these metrics.
1246
1248 pcp(1), pcp-atopsar(1), pmdaapache(1), pmdanfsclient(1), pmdaproc(1),
1249 mkaf(1), pmlogger(1), pmlogger_daily(1), PCPIntro(1) and pcp-atoprc(5).
1250Performance Co-Pilot PCP PCP-ATOP(1)