1PCP-ATOPSAR(1) General Commands Manual PCP-ATOPSAR(1)
2
6 pcp-atopsar - Advanced System Activity Report (pcp-atop related)
7
9 pcp [pcp options] atopsar [atop options] [-r file|date] [-h host] [-R
10 cnt] [-b yy-mm-dd] hh:mm] [-e yy-mm-dd] hh:mm]
11 pcp [pcp options] atopsar [atop options] interval [samples]
12
14 The pcp-atopsar program can be used to report statistics at the system
15 level.
16 In the first synopsis line (no sampling interval specified), pcp-atop‐
18 sar extracts data from a raw logfile that has been recorded previously
19 by pmlogger(1) (or via the -w option of the pcp-atop program).
20 You can specify the name of the logfile with the -r option of the pcp-
21 atopsar program. When a pmlogger daily logfile is used, named
22 $PCP_LOG_DIR/pmlogger/[host]/YYYYMMDD (where YYYYMMDD reflects the
23 date), the required date of the form YYYYMMDD can be specified with the
24 -r option instead of the filename, or the symbolic name 'y' can be used
25 for yesterday's daily logfile (this can be repeated so 'yyyy' indicates
26 the logfile of four days ago). If the -r option is not specified at
27 all, today's daily logfile is used by default.
28 By default, the hostname of the localhost will be used when resolving
29 pmlogger archives, however an alternative host can be specified using
30 the -h option.
31 The starting and ending times of the report can be defined using the
32 options -b and -e followed by a time argument of the form [yy-mm-dd]
33 hh:mm.
34 In the second synopsis line, pcp-atopsar reads actual activity counters
36 from the kernel with the specified interval (in seconds) and the speci‐
37 fied number of samples (optionally). When pcp-atopsar is activated in
38 this way it immediately sends the output for every requested report to
39 standard output. If only one type of report is requested, the header
40 is printed once and after every interval seconds the statistical coun‐
41 ters are shown for that period. If several reports are requested, a
42 header is printed per sample followed by the statistical counters for
43 that period.
44 When invoked via the pcp(1) command, the PCPIntro(1) options -h/--host,
46 -a/--archive, -O/--origin, -s/--samples, -t/--interval, -Z/--timezone
47 and several other pcp options become indirectly available, see PCPIn‐
48 tro(1) for their descriptions.
49 Some generic flags can be specified to influence the behaviour of the
51 pcp-atopsar program:
52 -S By default the timestamp at the beginning of a line is suppressed
54 if more lines are shown for one interval. With this flag a time‐
55 stamp is given for every output-line (easier for post-processing).
56 -a By default certain resources as disks and network interfaces are
58 only shown when they were active during the interval. With this
59 flag all resources of a given type are shown, even if they were
60 inactive during the interval.
61 -x By default pcp-atopsar only uses colors if output is directed to a
63 terminal (window). These colors might indicate that a critical
64 occupation percentage has been reached (red) or has been almost
65 reached (cyan) for a particular resource. See the man-page of
66 atop for a detailed description of this feature (section COLORS).
67 With the flag -x the use of colors is suppressed unconditionally.
68 -C By default pcp-atopsar only uses colors if output is directed to a
70 terminal (window). These colors might indicate that a critical
71 occupation percentage has been reached (red) or has been almost
72 reached (cyan) for a particular resource. See the man-page of
73 atop for a detailed description of this feature (section COLORS).
74 With the flag -C colors will always be used, even if output is not
75 directed to a terminal.
76 -M Use markers at the end of a line to indicate that a critical occu‐
78 pation percentage has been reached ('*') or has been almost
79 reached ('+') for particular resources. The marker '*' is similar
80 to the color red and the marker '+' to the color cyan. See the
81 man-page of atop for a detailed description of these colors (sec‐
82 tion COLORS).
83 -H Repeat the header line within a report for every N detail lines.
85 The value of N is determined dynamically in case of output to a
86 tty/window (depending on the number of lines); for output to a
87 file or pipe this value is 23.
88 -R Summarize cnt samples into one sample. When the logfile contains
90 e.g. samples of 10 minutes, the use of the flag '-R 6' shows a re‐
91 port with one sample for every hour.
92 Other flags are used to define which reports are required:
94 -A Show all possible reports.
96 -c Report about CPU utilization (in total and per cpu).
98 -g Report about GPU utilization (per GPU).
100 -p Report about processor-related matters, like load-averages and
102 hardware interrupts.
103 -P Report about processes.
105 -m Current memory- and swap-occupation.
107 -s Report about paging- and swapping-activity, and overcommitment.
109 -B Report about Pressure Stall Information (PSI).
111 -l Report about utilization of logical volumes.
113 -f Report about utilization of multiple devices.
115 -d Report about utilization of disks.
117 -n Report about NFS mounted filesystems on NFS client.
119 -j Report about NFS client activity.
121 -J Report about NFS server activity.
123 -i Report about the network interfaces.
125 -I Report about errors for network-interfaces.
127 -w Report about IP version 4 network traffic.
129 -W Report about errors for IP version 4 traffic.
131 -y General report about ICMP version 4 layer activity.
133 -Y Per-type report about ICMP version 4 layer activity.
135 -u Report about UDP version 4 network traffic.
137 -z Report about IP version 6 network traffic.
139 -Z Report about errors for IP version 6 traffic.
141 -k General report about ICMP version 6 layer activity.
143 -K Per-type report about ICMP version 6 layer activity.
145 -U Report about UDP version 6 network traffic.
147 -t Report about TCP network traffic.
149 -T Report about errors for TCP-traffic.
151 -h Report about Infiniband utilization.
153 -O Report about top-3 processes consuming most processor capacity.
155 This report is only available when using a log file (not when
156 specifying an interval).
157 -G Report about top-3 processes consuming most resident memory. This
159 report is only available when using a log file (not when specify‐
160 ing an interval).
161 -D Report about top-3 processes issueing most disk transfers. This
163 report is only available when using a log file (not when specify‐
164 ing an interval).
165 -N Report about top-3 processes issueing most IPv4/IPv6 socket trans‐
167 fers. This report is only available when using a log file (not
168 when specifying an interval).
169
171 The following additional PCP command line long options are also avail‐
172 able:
173 --align=align
175 Force the initial sample to be aligned on the boundary of a natu‐
176 ral time unit align. Refer to PCPIntro(1) for a complete descrip‐
177 tion of the syntax for align.
178 --archive=archive
180 Performance metric values are retrieved from the set of Perfor‐
181 mance Co-Pilot (PCP) archive log files identified by the argument
182 archive, which is a comma-separated list of names, each of which
183 may be the base name of an archive or the name of a directory con‐
184 taining one or more archives.
185 --finish=endtime
187 When reporting archived metrics, the report will be restricted to
188 those records logged before or at endtime. Refer to PCPIntro(1)
189 for a complete description of the syntax for endtime.
190 --host=host
192 Fetch performance metrics from pmcd(1) on host, rather than from
193 the default localhost.
194 --hostzone
196 Use the local timezone of the host that is the source of the per‐
197 formance metrics, as identified by either the --host or the --ar‐
198 chive options. The default is to use the timezone of the local
199 host.
200 --hotproc
202 Use the pmdaproc(1) hotproc metrics.
203 --interval=interval
205 Set the reporting interval to something other than the default 1
206 second. The interval argument follows the syntax described in
207 PCPIntro(1), and in the simplest form may be an unsigned integer
208 (the implied units in this case are seconds).
209 --samples=samples
211 The samples option defines the number of samples to be retrieved
212 and reported.
213 --start=starttime
215 When reporting archived metrics, the report will be restricted to
216 those records logged at or after starttime. Refer to PCPIntro(1)
217 for a complete description of the syntax for starttime.
218 --timezone=timezone
220 Use timezone for the date and time. Timezone is in the format of
221 the environment variable TZ as described in environ(7).
222 --version
224 Display version number and exit.
225
227 Depending on the requested report, a number of columns with output val‐
228 ues are produced. The values are mostly presented as a number of
229 events per second.
230 The output for the flag -c contains the following columns per cpu:
232 usr% Percentage of cpu-time consumed in user mode (program text)
234 for all active processes running with a nice value of zero
235 (default) or a negative nice value (which means a higher
236 priority than usual). The cpu consumption in user mode of
237 processes with a nice value larger than zero (lower prior‐
238 ity) is indicated in the nice%-column.
239 nice% Percentage of cpu time consumed in user mode (i.e. program
241 text) for all processes running witn a nice value larger
242 than zero (which means with a lower priority than average).
243 sys% Percentage of cpu time consumed in system mode (kernel
245 text) for all active processes. A high percentage usually
246 indicates a lot of system calls being issued.
247 irq% Percentage of cpu time consumed for handling of device in‐
249 terrupts.
250 softirq% Percentage of cpu time consumed for soft interrupt han‐
252 dling.
253 steal% Percentage of cpu time stolen by other virtual machines
255 running on the same hardware.
256 guest% Percentage of cpu time used by other virtual machines run‐
258 ning on the same hardware (overlaps with usr%/nice%).
259 wait% Percentage of unused cpu time while at least one of the
261 processes in wait-state awaits completion of disk I/O.
262 idle% Percentage of unused cpu time because all processes are in
264 a wait-state but not waiting for disk-I/O.
265 The output for the flag -g contains the following columns per GPU:
267 busaddr GPU number and bus-ID (separated by '/').
269 gpubusy GPU busy percentage during interval.
271 membusy GPU memory busy percentage during interval, i.e. time to
273 issue read and write accesses on memory.
274 memocc Percentage of memory occupation at this moment.
276 memtot Total memory available.
278 memuse Used GPU memory at this moment.
280 gputype Type of GPU.
282 The output for the flag -p contains the following values:
284 pswch/s Number of process switches (also called context switches)
286 per second on this cpu. A process switch occurs at the mo‐
287 ment that an active thread (i.e. the thread using a cpu)
288 enters a wait state or has used its time slice completely;
289 another thread will then be chosen to use the cpu.
290 devintr/s Number of hardware interrupts handled per second on this
292 cpu.
293 clones/s The number of new threads started per second.
295 loadavg1 Load average reflecting the average number of threads in
297 the runqueue or in non-interruptible wait state (usually
298 waiting for disk or tape I/O) during the last minute.
299 loadavg5 Load average reflecting the average number of threads in
301 the runqueue or in non-interruptible wait state (usually
302 waiting for disk or tape I/O) during the last 5 minutes.
303 loadavg15 Load average reflecting the average number of threads in
305 the runqueue or in non-interruptible wait state (usually
306 waiting for disk or tape I/O) during the last 15 minutes.
307 The output for the flag -P contains information about the processes and
309 threads:
310 clones/s The number of new threads started per second.
312 pexit/s
314 curproc Total number of processes present in the system.
316 curzomb Number of zombie processes present in the system.
318 thrrun Total number of threads present in the system in state
320 'running'.
321 thrslpi Total number of threads present in the system in state 'in‐
323 terruptible sleeping'.
324 thrslpu Total number of threads present in the system in state 'un‐
326 interruptible sleeping'.
327 The output for the flag -m contains information about the memory- and
329 swap-utilization:
330 memtotal Total usable main memory size.
332 memfree Available main memory size at this moment (snapshot).
334 buffers Main memory used at this moment to cache metadata-blocks
336 (snapshot).
337 cached Main memory used at this moment to cache data-blocks (snap‐
339 shot).
340 dirty Amount of memory in the page cache that still has to be
342 flushed to disk at this moment (snapshot).
343 slabmem Main memory used at this moment for dynamically allocated
345 memory by the kernel (snapshot).
346 swptotal Total swap space size at this moment (snapshot).
348 swpfree Available swap space at this moment (snapshot).
350 The output for the flag -s contains information about the frequency of
352 swapping:
353 pagescan/s Number of scanned pages per second due to the fact that
355 free memory drops below a particular threshold.
356 swapin/s The number of memory-pages the system read from the swap-
358 device per second.
359 swapout/s The number of memory-pages the system wrote to the swap-de‐
361 vice per second.
362 commitspc The committed virtual memory space i.e. the reserved vir‐
364 tual space for all allocations of private memory space for
365 processes.
366 commitlim The maximum limit for the committed space, which is by de‐
368 fault swap size plus 50% of memory size. The kernel only
369 verifies whether the committed space exceeds the limit if
370 strict overcommit handling is configured (vm.overcom‐
371 mit_memory is 2).
372 The output for the flag -B contains the Pressure Stall Information
374 (PSI):
375 cpusome Average pressure percentage during the interval for the
377 category 'CPU some'.
378 memsome Average pressure percentage during the interval for the
380 category 'memory some'.
381 memfull Average pressure percentage during the interval for the
383 category 'memory full'.
384 iosome Average pressure percentage during the interval for the
386 category 'I/O some'.
387 iofull Average pressure percentage during the interval for the
389 category 'I/O full'.
390 The output for the flags -l (LVM), -f (MD), and -d (hard disk) contains
392 the following columns per active unit:
393 disk Name.
395 busy Busy-percentage of the unit (i.e. the portion of time that
397 the device was busy handling requests).
398 read/s Number of read-requests issued per second on this unit.
400 KB/read Average number of Kbytes transferred per read-request for
402 this unit.
403 writ/s Number of write-requests issued per second on this unit.
405 KB/writ Average number of Kbytes transferred per write-request for
407 this unit.
408 avque Average number of requests outstanding in the queue during
410 the time that the unit is busy.
411 avserv Average number of milliseconds needed by a request on this
413 unit (seek, latency and data-transfer).
414 The output for the flag -n contains information about activity on NFS
416 mounted filesystems (client):
417 mounted_device
419 Mounted device containing server name and server directory
420 being mounted.
421 physread/s Kilobytes data physically read from the NFS server by pro‐
423 cesses running on the NFS client.
424 KBwrite/s Kilobytes data physically written to the NFS server by pro‐
426 cesses running on the NFS client.
427 When the NFS filesystem was mounted during the interval,
428 the state 'M' is shown.
429 The output for the flag -j contains information about NFS client activ‐
431 ity:
432 rpc/s Number of RPC calls per second issued to NFS server(s).
434 rpcread/s Number of read RPC calls per second issued to NFS
436 server(s).
437 rpcwrite/s Number of write RPC calls per second issued to NFS
439 server(s).
440 retrans/s Number of retransmitted RPC calls per second.
442 autrefresh/s
444 Number of authorization refreshes per second.
445 The output for the flag -J contains information about NFS server activ‐
447 ity:
448 rpc/s Number of RPC calls per second received from NFS client(s).
450 rpcread/s Number of read RPC calls per second received from NFS
452 client(s).
453 rpcwrite/s Number of write RPC calls per second received from NFS
455 client(s).
456 MBcr/s Number of Megabytes per second returned to read requests by
458 clients.
459 MBcw/s Number of Megabytes per second passed in write requests by
461 clients.
462 nettcp/s Number of requests per second handled via TCP.
464 netudp/s Number of requests per second handled via UDP.
466 The output for the flag -i provides information about utilization of
468 network interfaces:
469 interf Name of interface.
471 busy Busy percentage for this interface. If the linespeed of
473 this interface could not be determined (e.g. for virtual
474 interfaces), a question mark is shown.
475 ipack/s Number of packets received from this interface per second.
477 opack/s Number of packets transmitted to this interface per second.
479 iKbyte/s Number of Kbytes received from this interface per second.
481 oKbyte/s Number of Kbytes transmitted via this interface per second.
483 imbps/s Effective number of megabits received per second.
485 ombps/s Effective number of megabits transmitted per second.
487 maxmbps/s Linespeed as number of megabits per second. If the line‐
489 speed could not be determined (e.g. virtual interfaces),
490 value 0 is shown.
491 The linespeed is followed by the indication 'f' (full du‐
492 plex) or 'h' (half duplex).
493 The output for the flag -I provides information about the failures that
495 were detected for network interfaces:
496 interf Name of interface.
498 ierr/s Number of bad packets received from this interface per sec‐
500 ond.
501 oerr/s Number of times that packet transmission to this interface
503 failed per second.
504 coll/s Number of collisions encountered per second while transmit‐
506 ting packets.
507 idrop/s Number of received packets dropped per second due to lack
509 of buffer-space in the local system.
510 odrop/s Number of transmitted packets dropped per second due to
512 lack of buffer-space in the local system.
513 iframe/s Number of frame alignment-errors encountered per second on
515 received packets.
516 ocarrier/s Number of carrier-errors encountered per second on trans‐
518 mitted packets.
519 The output for the flag -w provides information about the utilization
521 of the IPv4-layer (formal SNMP-names between brackets):
522 inrecv/s Number of IP datagrams received from interfaces per second,
524 including those received in error (ipInReceives).
525 outreq/s Number of IP datagrams that local higher-layer protocols
527 supplied to IP in requests for transmission per second
528 (ipOutRequests).
529 indeliver/s Number of received IP datagrams that have been successfully
531 delivered to higher protocol-layers per second (ipInDeliv‐
532 ers).
533 forward/s Number of received IP datagrams per second for which this
535 entity was not their final IP destination, as a result of
536 which an attempt was made to forward (ipForwDatagrams).
537 reasmok/s Number of IP datagrams successfully reassembled per second
539 (ipReasmOKs).
540 fragcreat/s Number of IP datagram fragments generated per second at
542 this entity (ipFragCreates).
543 The output for the flag -W provides information about the failures that
545 were detected in the IPv4-layer (formal SNMP-names between brackets):
546 in: dsc/s Number of input IP datagrams per second for which no prob‐
548 lems were encountered to prevent their continued processing
549 but that were discarded, e.g. for lack of buffer space
550 (ipInDiscards).
551 in: hder/s Number of input IP datagrams per second discarded due to
553 errors in the IP header (ipInHdrErrors).
554 in: ader/s Number of input IP datagrams per second discarded because
556 the IP address in the destination field was not valid to be
557 received by this entity (ipInAddrErrors).
558 in: unkp/s Number of inbound packets per second that were discarded
560 because of an unknown or unsupported protocol (ipInUnknown‐
561 Protos).
562 in: ratim/s Number of timeout-situations per second while other frag‐
564 ments were expected for successful reassembly (ipReasmTime‐
565 out).
566 in: rfail/s Number of failures detected per second by the IP reassembly
568 algorithm (ipReasmFails).
569 out: dsc/s Number of output IP datagrams per second for which no prob‐
571 lems were encountered to prevent their continued processing
572 but that were discarded, e.g. for lack of buffer space
573 (ipOutDiscards).
574 out: nrt/s Number of IP datagrams per second discarded because no
576 route could be found (ipOutNoRoutes).
577 The output for the flag -y provides information about the general uti‐
579 lization of the ICMPv4-layer and some information per type of ICMP-mes‐
580 sage (formal SNMP-names between brackets):
581 intot/s Number of ICMP messages (any type) received per second at
583 this entity (icmpInMsgs).
584 outtot/s Number of ICMP messages (any type) transmitted per second
586 from this entity (icmpOutMsgs).
587 inecho/s Number of ICMP Echo (request) messages received per second
589 (icmpInEchos).
590 inerep/s Number of ICMP Echo-Reply messages received per second
592 (icmpInEchoReps).
593 otecho/s Number of ICMP Echo (request) messages transmitted per sec‐
595 ond (icmpOutEchos).
596 oterep/s Number of ICMP Echo-Reply messages transmitted per second
598 (icmpOutEchoReps).
599 The output for the flag -Y provides information about other types of
601 ICMPv4-messages (formal SNMP-names between brackets):
602 ierr/s Number of ICMP messages received per second but determined
604 to have ICMP-specific errors (icmpInErrors).
605 isq/s Number of ICMP Source Quench messages received per second
607 (icmpInSrcQuenchs).
608 ird/s Number of ICMP Redirect messages received per second (icmp‐
610 InRedirects).
611 idu/s Number of ICMP Destination Unreachable messages received
613 per second (icmpInDestUnreachs).
614 ite/s Number of ICMP Time Exceeded messages received per second
616 (icmpOutTimeExcds).
617 oerr/s Number of ICMP messages transmitted per second but deter‐
619 mined to have ICMP-specific errors (icmpOutErrors).
620 osq/s Number of ICMP Source Quench messages transmitted per sec‐
622 ond (icmpOutSrcQuenchs).
623 ord/s Number of ICMP Redirect messages transmitted per second
625 (icmpOutRedirects).
626 odu/s Number of ICMP Destination Unreachable messages transmitted
628 per second (icmpOutDestUnreachs).
629 ote/s Number of ICMP Time Exceeded messages transmitted per sec‐
631 ond (icmpOutTimeExcds).
632 The output for the flag -u provides information about the utilization
634 of the UDPv4-layer (formal SNMP-names between brackets):
635 indgram/s Number of UDP datagrams per second delivered to UDP users
637 (udpInDatagrams).
638 outdgram/s Number of UDP datagrams transmitted per second from this
640 entity (udpOutDatagrams).
641 inerr/s Number of received UDP datagrams per second that could not
643 be delivered for reasons other than the lack of an applica‐
644 tion at the destination port (udpInErrors).
645 noport/s Number of received UDP datagrams per second for which there
647 was no application at the destination port (udpNoPorts).
648 The output for the flag -z provides information about the utilization
650 of the IPv6-layer (formal SNMP-names between brackets):
651 inrecv/s Number of input IPv6-datagrams received from interfaces per
653 second, including those received in error (ipv6IfStatsInRe‐
654 ceives).
655 outreq/s Number of IPv6-datagrams per second that local higher-layer
657 protocols supplied to IP in requests for transmission
658 (ipv6IfStatsOutRequests). This counter does not include
659 any forwarded datagrams.
660 inmc/s Number of multicast packets per second that have been re‐
662 ceived by the interface (ipv6IfStatsInMcastPkts).
663 outmc/s Number of multicast packets per second that have been
665 transmitted to the interface (ipv6IfStatsOutMcastPkts).
666 indeliv/s Number of IP datagrams successfully delivered per second to
668 IPv6 user-protocols, including ICMP (ipv6IfStatsInDeliv‐
669 ers).
670 reasmok/s Number of IPv6 datagrams successfully reassembled per sec‐
672 ond (ipv6IfStatsReasmOKs).
673 fragcre/s Number of IPv6 datagram fragments generated per second at
675 this entity (ipv6IfStatsOutFragCreates).
676 The output for the flag -Z provides information about the failures that
678 were detected in the IPv6-layer (formal SNMP-names between brackets):
679 in: dsc/s Number of input IPv6 datagrams per second for which no
681 problems were encountered to prevent their continued pro‐
682 cessing but that were discarded, e.g. for lack of buffer
683 space (ipv6IfStatsInDiscards).
684 in: hder/s Number of input datagrams per second discarded due to er‐
686 rors in the IPv6 header (ipv6IfStatsInHdrErrors).
687 in: ader/s Number of input datagrams per second discarded because the
689 IPv6 address in the destination field was not valid to be
690 received by this entity (ipv6IfStatsInAddrErrors).
691 in: unkp/s Number of locally-addressed datagrams per second that were
693 discarded because of an unknown or unsupported protocol
694 (ipv6IfStatsInUnknownProtos).
695 in: ratim/s Number of timeout-situations per second while other IPv6
697 fragments were expected for successful reassembly
698 (ipv6ReasmTimeout).
699 in: rfail/s Number of failures detected per second by the IPv6 reassem‐
701 bly-algorithm (ipv6IfStatsReasmFails).
702 out: dsc/s Number of output IPv6 datagrams per second for which no
704 problems were encountered to prevent their continued pro‐
705 cessing but that were discarded, e.g. for lack of buffer
706 space (ipv6IfStatsOutDiscards).
707 out: nrt/s Number of IPv6 datagrams per second discarded because no
709 route could be found (ipv6IfStatsInNoRoutes).
710 The output for the flag -k provides information about the general uti‐
712 lization of the ICMPv6-layer and some information per type of ICMP-mes‐
713 sage (formal SNMP-names between brackets):
714 intot/s Number of ICMPv6 messages (any type) received per second at
716 the interface (ipv6IfIcmpInMsgs).
717 outtot/s Number of ICMPv6 messages (any type) transmitted per second
719 from this entity (ipv6IfIcmpOutMsgs).
720 inerr/s Number of ICMPv6 messages received per second that had
722 ICMP-specific errors, such as bad ICMP checksums, bad
723 length, etc (ipv6IfIcmpInErrors).
724 innsol/s Number of ICMP Neighbor Solicit messages received per sec‐
726 ond (ipv6IfIcmpInNeighborSolicits).
727 innadv/s Number of ICMP Neighbor Advertisement messages received per
729 second (ipv6IfIcmpInNeighborAdvertisements).
730 otnsol/s Number of ICMP Neighbor Solicit messages transmitted per
732 second (ipv6IfIcmpOutNeighborSolicits).
733 otnadv/s Number of ICMP Neighbor Advertisement messages transmitted
735 per second (ipv6IfIcmpOutNeighborAdvertisements).
736 The output for the flag -K provides information about other types of
738 ICMPv6-messages (formal SNMP-names between brackets):
739 iecho/s Number of ICMP Echo (request) messages received per second
741 (ipv6IfIcmpInEchos).
742 ierep/s Number of ICMP Echo-Reply messages received per second
744 (ipv6IfIcmpInEchoReplies).
745 oerep/s Number of ICMP Echo-Reply messages transmitted per second
747 (ipv6IfIcmpOutEchoReplies).
748 idu/s Number of ICMP Destination Unreachable messages received
750 per second (ipv6IfIcmpInDestUnreachs).
751 odu/s Number of ICMP Destination Unreachable messages transmitted
753 per second (ipv6IfIcmpOutDestUnreachs).
754 ird/s Number of ICMP Redirect messages received per second
756 (ipv6IfIcmpInRedirects).
757 ord/s Number of ICMP Redirect messages transmitted per second
759 (ipv6IfIcmpOutRedirect).
760 ite/s Number of ICMP Time Exceeded messages received per second
762 (ipv6IfIcmpInTimeExcds).
763 ote/s Number of ICMP Time Exceeded messages transmitted per sec‐
765 ond (ipv6IfIcmpOutTimeExcds).
766 The output for the flag -U provides information about the utilization
768 of the UDPv6-layer (formal SNMP-names between brackets):
769 indgram/s Number of UDPv6 datagrams per second delivered to UDP users
771 (udpInDatagrams),
772 outdgram/s Number of UDPv6 datagrams transmitted per second from this
774 entity (udpOutDatagrams),
775 inerr/s Number of received UDPv6 datagrams per second that could
777 not be delivered for reasons other than the lack of an ap‐
778 plication at the destination port (udpInErrors).
779 noport/s Number of received UDPv6 datagrams per second for which
781 there was no application at the destination port (udpNo‐
782 Ports).
783 The output for the flag -t provides information about the utilization
785 of the TCP-layer (formal SNMP-names between brackets):
786 insegs/s Number of received segments per second, including those re‐
788 ceived in error (tcpInSegs).
789 outsegs/s Number of transmitted segments per second, excluding those
791 containing only retransmitted octets (tcpOutSegs).
792 actopen/s Number of active opens per second that have been supported
794 by this entity (tcpActiveOpens).
795 pasopen/s Number of passive opens per second that have been supported
797 by this entity (tcpPassiveOpens).
798 nowopen Number of connections currently open (snapshot), for which
800 the state is either ESTABLISHED or CLOSE-WAIT (tcpCur‐
801 rEstab).
802 The output for the flag -T provides information about the failures that
804 were detected in the TCP-layer (formal SNMP-names between brackets):
805 inerr/s Number of received segments per second received in error
807 (tcpInErrs).
808 retrans/s Number of retransmitted segments per second (tcpRe‐
810 transSegs).
811 attfail/s Number of failed connection attempts per second that have
813 occurred at this entity (tcpAttemptFails).
814 estabreset/s
816 Number of resets per second that have occurred at this en‐
817 tity (tcpEstabResets).
818 outreset/s Number of transmitted segments per second containing the
820 RST flag (tcpOutRsts).
821 The output for the flag -h provides information about utilization of
823 Infiniband ports:
824 controller Name of controller.
826 port Controller port.
828 busy Busy percentage for this port.
830 ipack/s Number of packets received from this port per second.
832 opack/s Number of packets transmitted to this port per second.
834 igbps/s Effective number of gigabits received per second.
836 ogbps/s Effective number of gigabits transmitted per second.
838 maxgbps/s Maximum rate as number of gigabits per second.
840 lanes Number of lanes.
842 The output for the flag -O provides information about the top-3 of pro‐
844 cesses with the highest processor consumption:
845 pid Process-id (if zero, the process has exited while the pid
847 could not be determined).
848 command The name of the process.
850 cpu% The percentage of cpu-capacity being consumed. This value
852 can exceed 100% for a multithreaded process running on a
853 multiprocessor machine.
854 The output for the flag -G provides information about the top-3 of pro‐
856 cesses with the highest memory consumption:
857 pid Process-id (if zero, the process has exited while the pid
859 could not be determined).
860 command The name of the process.
862 mem% The percentage of resident memory-utilization by this
864 process.
865 The output for the flag -D provides information about the top-3 of pro‐
867 cesses that issue the most read and write accesses to disk:
868 pid Process-id (if zero, the process has exited while the pid
870 could not be determined).
871 command The name of the process.
873 dsk% The percentage of read and write accesses related to the
875 total number of read and write accesses issued on disk by
876 all processes, so a high percentage does not imply a high
877 disk load on system level.
878 The output for the flag -N provides information about the top-3 of pro‐
880 cesses that issue the most socket transfers for IPv4/IPv6:
881 pid Process-id (if zero, the process has exited while the pid
883 could not be determined).
884 command The name of the process.
886 net% The percentage of socket transfers related to the total
888 number of transfers issued by all processes, so a high per‐
889 centage does not imply a high network load on system level.
890
892 To see today's cpu-activity so far (supposed that atop is logging in
893 the background):
894 pcp-atopsar
896 To see the memory occupation for June 5, 2018 between 10:00 and 12:30
898 (supposed that pmlogger has been logging daily in the background on
899 host acme.com):
900 pcp-atopsar -m -r $PCP_LOG_DIR/pmlogger/acme.com/20180605 -b 10:00 -e
902 12:30
903 or
905 pcp-atopsar -m -r 20180605 -b 10:00 -e 12:30
907 or, suppose it is June 8, 2018 at this moment
909 pcp-atopsar -m -r yyy -b 10:00 -e 12:30
911 Write a logfile with atop to record the system behaviour for 30 minutes
913 (30 samples of one minute) and produce all available reports after‐
914 wards:
915 pcp-atop -w /tmp/atoplog 60 30
917 pcp-atopsar -A -r /tmp/atoplog
919 To watch TCP activity evolve for ten minutes (10 samples with sixty
921 seconds interval):
922 pcp-atopsar -t 60 10
924 To watch the header-lines ('_' as last character) of all reports with
926 only the detail-lines showing critical resource consumption (marker '*'
927 or '+' as last character):
928 pcp-atopsar -AM | grep '[_*+]$'
930
932 /etc/atoprc
933 Configuration file containing system-wide default values (mainly
934 flags). See related man-page.
935 ~/.atoprc
937 Configuration file containing personal default values (mainly
938 flags). See related man-page.
939 $PCP_LOG_DIR/pmlogger/HOST/YYYYMMDD
941 Daily data file, where YYYYMMDD are digits representing the date,
942 and HOST is the hostname of the machine being logged.
943
945 Environment variables with the prefix PCP_ are used to parameterize the
946 file and directory names used by PCP. On each installation, the file
947 /etc/pcp.conf contains the local values for these variables. The
948 $PCP_CONF variable may be used to specify an alternative configuration
949 file, as described in pcp.conf(5).
950 For environment variables affecting PCP tools, see pmGetOptions(3).
952
954 PCPIntro(1), pcp(1), pcp-atop(1), mkaf(1), pmlogger(1), pmlog‐
955 ger_daily(1) and pcp-atoprc(5).
956Performance Co-Pilot PCP PCP-ATOPSAR(1)