1ATOPSAR(1) General Commands Manual ATOPSAR(1)
2
6 atopsar - Advanced System Activity Report (atop related)
7
9 atopsar [-flags...] [-r file|date|- ] [-R cnt ] [-b [YYYYMMDD]hhmm ]
10 [-e [YYYYMMDD]hhmm ]
11 atopsar [-flags...] interval [ samples ]
12
14 The program atopsar can be used to report statistics on system level.
15 In the first synopsis line (no sampling interval specified), atopsar
17 extracts data from a raw logfile that has been recorded previously by
18 the program atop (option -w of the atop program).
19 You can specify the name of the logfile with the -r option of the atop‐
20 sar program. When a daily logfile of atop is used, named
21 /var/log/atop/atop_YYYYMMDD (where YYYYMMDD reflects the date), the re‐
22 quired date of the form YYYYMMDD can be specified with the -r option
23 instead of the filename, or the symbolic name 'y' can be used for yes‐
24 terday's daily logfile (this can be repeated so 'yyyy' indicates the
25 logfile of four days ago), or the filename '-' can be used to read raw
26 data from stdin. If the -r option is not specified at all, today's
27 daily logfile is used by default.
28 The starting and ending times of the report can be defined using the
29 options -b and -e followed by a time argument of the form [YYYYM‐
30 MDD]hhmm.
31 In the second synopsis line, atopsar reads actual activity counters
33 from the kernel with the specified interval (in seconds) and the speci‐
34 fied number of samples (optionally). When atopsar is activated in this
35 way it immediately sends the output for every requested report to stan‐
36 dard output. If only one type of report is requested, the header is
37 printed once and after every interval seconds the statistical counters
38 are shown for that period. If several reports are requested, a header
39 is printed per sample followed by the statistical counters for that pe‐
40 riod.
41 Some generic flags can be specified to influence the behaviour of the
43 atopsar program:
44 -S By default the timestamp at the beginning of a line is suppressed
46 if more lines are shown for one interval. With this flag a time‐
47 stamp is given for every output-line (easier for post-processing).
48 -a By default certain resources as disks and network interfaces are
50 only shown when they were active during the interval. With this
51 flag all resources of a given type are shown, even if they were
52 inactive during the interval.
53 -x By default atopsar only uses colors if output is directed to a
55 terminal (window). These colors might indicate that a critical
56 occupation percentage has been reached (red) or has been almost
57 reached (cyan) for a particular resource. See the man-page of
58 atop for a detailed description of this feature (section COLORS).
59 With the flag -x the use of colors is suppressed unconditionally.
60 -C By default atopsar only uses colors if output is directed to a
62 terminal (window). These colors might indicate that a critical
63 occupation percentage has been reached (red) or has been almost
64 reached (cyan) for a particular resource. See the man-page of
65 atop for a detailed description of this feature (section COLORS).
66 With the flag -C colors will always be used, even if output is not
67 directed to a terminal.
68 -M Use markers at the end of a line to indicate that a critical occu‐
70 pation percentage has been reached ('*') or has been almost
71 reached ('+') for particular resources. The marker '*' is similar
72 to the color red and the marker '+' to the color cyan. See the
73 man-page of atop for a detailed description of these colors (sec‐
74 tion COLORS).
75 -H Repeat the header line within a report for every N detail lines.
77 The value of N is determined dynamically in case of output to a
78 tty/window (depending on the number of lines); for output to a
79 file or pipe this value is 23.
80 -R Summarize cnt samples into one sample. When the logfile contains
82 e.g. samples of 10 minutes, the use of the flag '-R 6' shows a re‐
83 port with one sample for every hour.
84 Other flags are used to define which reports are required:
86 -A Show all possible reports.
88 -c Report about CPU utilization (in total and per cpu).
90 -g Report about GPU utilization (per GPU).
92 -p Report about processor-related matters, like load-averages and
94 hardware interrupts.
95 -P Report about processes.
97 -m Current memory- and swap-occupation.
99 -s Report about paging- and swapping-activity, and overcommitment.
101 -B Report about Pressure Stall Information (PSI).
103 -l Report about utilization of logical volumes.
105 -f Report about utilization of multiple devices.
107 -d Report about utilization of disks.
109 -n Report about NFS mounted filesystems on NFS client.
111 -j Report about NFS client activity.
113 -J Report about NFS server activity.
115 -i Report about the network interfaces.
117 -I Report about errors for network-interfaces.
119 -w Report about IP version 4 network traffic.
121 -W Report about errors for IP version 4 traffic.
123 -y General report about ICMP version 4 layer activity.
125 -Y Per-type report about ICMP version 4 layer activity.
127 -u Report about UDP version 4 network traffic.
129 -z Report about IP version 6 network traffic.
131 -Z Report about errors for IP version 6 traffic.
133 -k General report about ICMP version 6 layer activity.
135 -K Per-type report about ICMP version 6 layer activity.
137 -U Report about UDP version 6 network traffic.
139 -t Report about TCP network traffic.
141 -T Report about errors for TCP-traffic.
143 -h Report about Infiniband utilization.
145 -O Report about top-3 processes consuming most processor capacity.
147 This report is only available when using a log file (not when
148 specifying an interval).
149 -G Report about top-3 processes consuming most resident memory. This
151 report is only available when using a log file (not when specify‐
152 ing an interval).
153 -D Report about top-3 processes issuing most disk transfers. This
155 report is only available when using a log file (not when specify‐
156 ing an interval).
157 -N Report about top-3 processes issuing most IPv4/IPv6 socket trans‐
159 fers. This report is only available when using a log file (not
160 when specifying an interval).
161
163 Depending on the requested report, a number of columns with output val‐
164 ues are produced. The values are mostly presented as a number of
165 events per second.
166 The output for the flag -c contains the following columns per cpu:
168 usr% Percentage of cpu-time consumed in user mode (program text)
170 for all active processes running with a nice value of zero
171 (default) or a negative nice value (which means a higher
172 priority than usual). The cpu consumption in user mode of
173 processes with a nice value larger than zero (lower prior‐
174 ity) is indicated in the nice%-column.
175 nice% Percentage of cpu time consumed in user mode (i.e. program
177 text) for all processes running witn a nice value larger
178 than zero (which means with a lower priority than average).
179 sys% Percentage of cpu time consumed in system mode (kernel
181 text) for all active processes. A high percentage usually
182 indicates a lot of system calls being issued.
183 irq% Percentage of cpu time consumed for handling of device in‐
185 terrupts.
186 softirq% Percentage of cpu time consumed for soft interrupt han‐
188 dling.
189 steal% Percentage of cpu time stolen by other virtual machines
191 running on the same hardware.
192 guest% Percentage of cpu time used by other virtual machines run‐
194 ning on the same hardware (overlaps with usr%/nice%).
195 wait% Percentage of unused cpu time while at least one of the
197 processes in wait-state awaits completion of disk I/O.
198 idle% Percentage of unused cpu time because all processes are in
200 a wait-state but not waiting for disk-I/O.
201 The output for the flag -g contains the following columns per GPU:
203 busaddr GPU number and bus-ID (separated by '/').
205 gpubusy GPU busy percentage during interval.
207 membusy GPU memory busy percentage during interval, i.e. time to
209 issue read and write accesses on memory.
210 memocc Percentage of memory occupation at this moment.
212 memtot Total memory available.
214 memuse Used GPU memory at this moment.
216 gputype Type of GPU.
218 The output for the flag -p contains the following values:
220 pswch/s Number of process switches (also called context switches)
222 per second on this cpu. A process switch occurs at the mo‐
223 ment that an active thread (i.e. the thread using a cpu)
224 enters a wait state or has used its time slice completely;
225 another thread will then be chosen to use the cpu.
226 devintr/s Number of hardware interrupts handled per second on this
228 cpu.
229 clones/s The number of new threads started per second.
231 loadavg1 Load average reflecting the average number of threads in
233 the runqueue or in non-interruptible wait state (usually
234 waiting for disk or tape I/O) during the last minute.
235 loadavg5 Load average reflecting the average number of threads in
237 the runqueue or in non-interruptible wait state (usually
238 waiting for disk or tape I/O) during the last 5 minutes.
239 loadavg15 Load average reflecting the average number of threads in
241 the runqueue or in non-interruptible wait state (usually
242 waiting for disk or tape I/O) during the last 15 minutes.
243 The output for the flag -P contains information about the processes and
245 threads:
246 clones/s The number of new threads started per second.
248 pexit/s
250 curproc Total number of processes present in the system.
252 curzomb Number of zombie processes present in the system.
254 thrrun Total number of threads present in the system in state
256 'running'.
257 thrslpi Total number of threads present in the system in state 'in‐
259 terruptible sleeping'.
260 thrslpu Total number of threads present in the system in state 'un‐
262 interruptible sleeping'.
263 The output for the flag -m contains information about the memory- and
265 swap-utilization:
266 memtotal Total usable main memory size.
268 memfree Available main memory size at this moment (snapshot).
270 buffers Main memory used at this moment to cache metadata-blocks
272 (snapshot).
273 cached Main memory used at this moment to cache data-blocks (snap‐
275 shot).
276 dirty Amount of memory in the page cache that still has to be
278 flushed to disk at this moment (snapshot).
279 slabmem Main memory used at this moment for dynamically allocated
281 memory by the kernel (snapshot).
282 swptotal Total swap space size at this moment (snapshot).
284 swpfree Available swap space at this moment (snapshot).
286 The output for the flag -s contains information about the frequency of
288 swapping:
289 pagescan/s Number of scanned pages per second due to the fact that
291 free memory drops below a particular threshold.
292 swapin/s The number of memory-pages the system read from the swap-
294 device per second.
295 swapout/s The number of memory-pages the system wrote to the swap-de‐
297 vice per second.
298 oomkill The number of processes being killed during the last inter‐
300 val due to lack of memory/swap. The value -1 means that
301 this counter is not supported by the current kernel ver‐
302 sion.
303 commitspc The committed virtual memory space i.e. the reserved vir‐
305 tual space for all allocations of private memory space for
306 processes.
307 commitlim The maximum limit for the committed space, which is by de‐
309 fault swap size plus 50% of memory size. The kernel only
310 verifies whether the committed space exceeds the limit if
311 strict overcommit handling is configured (vm.overcom‐
312 mit_memory is 2).
313 The output for the flag -B contains the Pressure Stall Information
315 (PSI):
316 cpusome Average pressure percentage during the interval for the
318 category 'CPU some'.
319 memsome Average pressure percentage during the interval for the
321 category 'memory some'.
322 memfull Average pressure percentage during the interval for the
324 category 'memory full'.
325 iosome Average pressure percentage during the interval for the
327 category 'I/O some'.
328 iofull Average pressure percentage during the interval for the
330 category 'I/O full'.
331 The output for the flags -l (LVM), -f (MD), and -d (hard disk) contains
333 the following columns per active unit:
334 disk Name.
336 busy Busy-percentage of the unit (i.e. the portion of time that
338 the device was busy handling requests).
339 read/s Number of read-requests issued per second on this unit.
341 KB/read Average number of Kbytes transferred per read-request for
343 this unit.
344 writ/s Number of write-requests (including discard requests) is‐
346 sued per second on this unit.
347 KB/writ Average number of Kbytes transferred per write-request for
349 this unit.
350 avque Average number of requests outstanding in the queue during
352 the time that the unit is busy.
353 avserv Average number of milliseconds needed by a request on this
355 unit (seek, latency and data-transfer).
356 The output for the flag -n contains information about activity on NFS
358 mounted filesystems (client):
359 mounted_device
361 Mounted device containing server name and server directory
362 being mounted.
363 physread/s Kilobytes data physically read from the NFS server by pro‐
365 cesses running on the NFS client.
366 KBwrite/s Kilobytes data physically written to the NFS server by pro‐
368 cesses running on the NFS client.
369 When the NFS filesystem was mounted during the interval,
370 the state 'M' is shown.
371 The output for the flag -j contains information about NFS client activ‐
373 ity:
374 rpc/s Number of RPC calls per second issued to NFS server(s).
376 rpcread/s Number of read RPC calls per second issued to NFS
378 server(s).
379 rpcwrite/s Number of write RPC calls per second issued to NFS
381 server(s).
382 retrans/s Number of retransmitted RPC calls per second.
384 autrefresh/s
386 Number of authorization refreshes per second.
387 The output for the flag -J contains information about NFS server activ‐
389 ity:
390 rpc/s Number of RPC calls per second received from NFS client(s).
392 rpcread/s Number of read RPC calls per second received from NFS
394 client(s).
395 rpcwrite/s Number of write RPC calls per second received from NFS
397 client(s).
398 MBcr/s Number of Megabytes per second returned to read requests by
400 clients.
401 MBcw/s Number of Megabytes per second passed in write requests by
403 clients.
404 nettcp/s Number of requests per second handled via TCP.
406 netudp/s Number of requests per second handled via UDP.
408 The output for the flag -i provides information about utilization of
410 network interfaces:
411 interf Name of interface.
413 busy Busy percentage for this interface. If the linespeed of
415 this interface could not be determined (for virtual inter‐
416 faces or in case that atop or atopsar had no root-privi‐
417 leges), a question mark is shown.
418 ipack/s Number of packets received from this interface per second.
420 opack/s Number of packets transmitted to this interface per second.
422 iKbyte/s Number of Kbytes received from this interface per second.
424 oKbyte/s Number of Kbytes transmitted via this interface per second.
426 imbps/s Effective number of megabits received per second.
428 ombps/s Effective number of megabits transmitted per second.
430 maxmbps/s Linespeed as number of megabits per second. If the line‐
432 speed could not be determined (for virtual interfaces or in
433 case that atop or atopsar had no root-privileges), value 0
434 is shown.
435 The linespeed is followed by the indication 'f' (full du‐
436 plex) or 'h' (half duplex).
437 The output for the flag -I provides information about the failures that
439 were detected for network interfaces:
440 interf Name of interface.
442 ierr/s Number of bad packets received from this interface per sec‐
444 ond.
445 oerr/s Number of times that packet transmission to this interface
447 failed per second.
448 coll/s Number of collisions encountered per second while transmit‐
450 ting packets.
451 idrop/s Number of received packets dropped per second due to lack
453 of buffer-space in the local system.
454 odrop/s Number of transmitted packets dropped per second due to
456 lack of buffer-space in the local system.
457 iframe/s Number of frame alignment-errors encountered per second on
459 received packets.
460 ocarrier/s Number of carrier-errors encountered per second on trans‐
462 mitted packets.
463 The output for the flag -w provides information about the utilization
465 of the IPv4-layer (formal SNMP-names between brackets):
466 inrecv/s Number of IP datagrams received from interfaces per second,
468 including those received in error (ipInReceives).
469 outreq/s Number of IP datagrams that local higher-layer protocols
471 supplied to IP in requests for transmission per second
472 (ipOutRequests).
473 indeliver/s Number of received IP datagrams that have been successfully
475 delivered to higher protocol-layers per second (ipInDeliv‐
476 ers).
477 forward/s Number of received IP datagrams per second for which this
479 entity was not their final IP destination, as a result of
480 which an attempt was made to forward (ipForwDatagrams).
481 reasmok/s Number of IP datagrams successfully reassembled per second
483 (ipReasmOKs).
484 fragcreat/s Number of IP datagram fragments generated per second at
486 this entity (ipFragCreates).
487 The output for the flag -W provides information about the failures that
489 were detected in the IPv4-layer (formal SNMP-names between brackets):
490 in: dsc/s Number of input IP datagrams per second for which no prob‐
492 lems were encountered to prevent their continued processing
493 but that were discarded, e.g. for lack of buffer space
494 (ipInDiscards).
495 in: hder/s Number of input IP datagrams per second discarded due to
497 errors in the IP header (ipInHdrErrors).
498 in: ader/s Number of input IP datagrams per second discarded because
500 the IP address in the destination field was not valid to be
501 received by this entity (ipInAddrErrors).
502 in: unkp/s Number of inbound packets per second that were discarded
504 because of an unknown or unsupported protocol (ipInUnknown‐
505 Protos).
506 in: ratim/s Number of timeout-situations per second while other frag‐
508 ments were expected for successful reassembly (ipReasmTime‐
509 out).
510 in: rfail/s Number of failures detected per second by the IP reassembly
512 algorithm (ipReasmFails).
513 out: dsc/s Number of output IP datagrams per second for which no prob‐
515 lems were encountered to prevent their continued processing
516 but that were discarded, e.g. for lack of buffer space
517 (ipOutDiscards).
518 out: nrt/s Number of IP datagrams per second discarded because no
520 route could be found (ipOutNoRoutes).
521 The output for the flag -y provides information about the general uti‐
523 lization of the ICMPv4-layer and some information per type of ICMP-mes‐
524 sage (formal SNMP-names between brackets):
525 intot/s Number of ICMP messages (any type) received per second at
527 this entity (icmpInMsgs).
528 outtot/s Number of ICMP messages (any type) transmitted per second
530 from this entity (icmpOutMsgs).
531 inecho/s Number of ICMP Echo (request) messages received per second
533 (icmpInEchos).
534 inerep/s Number of ICMP Echo-Reply messages received per second
536 (icmpInEchoReps).
537 otecho/s Number of ICMP Echo (request) messages transmitted per sec‐
539 ond (icmpOutEchos).
540 oterep/s Number of ICMP Echo-Reply messages transmitted per second
542 (icmpOutEchoReps).
543 The output for the flag -Y provides information about other types of
545 ICMPv4-messages (formal SNMP-names between brackets):
546 ierr/s Number of ICMP messages received per second but determined
548 to have ICMP-specific errors (icmpInErrors).
549 isq/s Number of ICMP Source Quench messages received per second
551 (icmpInSrcQuenchs).
552 ird/s Number of ICMP Redirect messages received per second (icmp‐
554 InRedirects).
555 idu/s Number of ICMP Destination Unreachable messages received
557 per second (icmpInDestUnreachs).
558 ite/s Number of ICMP Time Exceeded messages received per second
560 (icmpOutTimeExcds).
561 oerr/s Number of ICMP messages transmitted per second but deter‐
563 mined to have ICMP-specific errors (icmpOutErrors).
564 osq/s Number of ICMP Source Quench messages transmitted per sec‐
566 ond (icmpOutSrcQuenchs).
567 ord/s Number of ICMP Redirect messages transmitted per second
569 (icmpOutRedirects).
570 odu/s Number of ICMP Destination Unreachable messages transmitted
572 per second (icmpOutDestUnreachs).
573 ote/s Number of ICMP Time Exceeded messages transmitted per sec‐
575 ond (icmpOutTimeExcds).
576 The output for the flag -u provides information about the utilization
578 of the UDPv4-layer (formal SNMP-names between brackets):
579 indgram/s Number of UDP datagrams per second delivered to UDP users
581 (udpInDatagrams).
582 outdgram/s Number of UDP datagrams transmitted per second from this
584 entity (udpOutDatagrams).
585 inerr/s Number of received UDP datagrams per second that could not
587 be delivered for reasons other than the lack of an applica‐
588 tion at the destination port (udpInErrors).
589 noport/s Number of received UDP datagrams per second for which there
591 was no application at the destination port (udpNoPorts).
592 The output for the flag -z provides information about the utilization
594 of the IPv6-layer (formal SNMP-names between brackets):
595 inrecv/s Number of input IPv6-datagrams received from interfaces per
597 second, including those received in error (ipv6IfStatsInRe‐
598 ceives).
599 outreq/s Number of IPv6-datagrams per second that local higher-layer
601 protocols supplied to IP in requests for transmission
602 (ipv6IfStatsOutRequests). This counter does not include
603 any forwarded datagrams.
604 inmc/s Number of multicast packets per second that have been re‐
606 ceived by the interface (ipv6IfStatsInMcastPkts).
607 outmc/s Number of multicast packets per second that have been
609 transmitted to the interface (ipv6IfStatsOutMcastPkts).
610 indeliv/s Number of IP datagrams successfully delivered per second to
612 IPv6 user-protocols, including ICMP (ipv6IfStatsInDeliv‐
613 ers).
614 reasmok/s Number of IPv6 datagrams successfully reassembled per sec‐
616 ond (ipv6IfStatsReasmOKs).
617 fragcre/s Number of IPv6 datagram fragments generated per second at
619 this entity (ipv6IfStatsOutFragCreates).
620 The output for the flag -Z provides information about the failures that
622 were detected in the IPv6-layer (formal SNMP-names between brackets):
623 in: dsc/s Number of input IPv6 datagrams per second for which no
625 problems were encountered to prevent their continued pro‐
626 cessing but that were discarded, e.g. for lack of buffer
627 space (ipv6IfStatsInDiscards).
628 in: hder/s Number of input datagrams per second discarded due to er‐
630 rors in the IPv6 header (ipv6IfStatsInHdrErrors).
631 in: ader/s Number of input datagrams per second discarded because the
633 IPv6 address in the destination field was not valid to be
634 received by this entity (ipv6IfStatsInAddrErrors).
635 in: unkp/s Number of locally-addressed datagrams per second that were
637 discarded because of an unknown or unsupported protocol
638 (ipv6IfStatsInUnknownProtos).
639 in: ratim/s Number of timeout-situations per second while other IPv6
641 fragments were expected for successful reassembly
642 (ipv6ReasmTimeout).
643 in: rfail/s Number of failures detected per second by the IPv6 reassem‐
645 bly-algorithm (ipv6IfStatsReasmFails).
646 out: dsc/s Number of output IPv6 datagrams per second for which no
648 problems were encountered to prevent their continued pro‐
649 cessing but that were discarded, e.g. for lack of buffer
650 space (ipv6IfStatsOutDiscards).
651 out: nrt/s Number of IPv6 datagrams per second discarded because no
653 route could be found (ipv6IfStatsInNoRoutes).
654 The output for the flag -k provides information about the general uti‐
656 lization of the ICMPv6-layer and some information per type of ICMP-mes‐
657 sage (formal SNMP-names between brackets):
658 intot/s Number of ICMPv6 messages (any type) received per second at
660 the interface (ipv6IfIcmpInMsgs).
661 outtot/s Number of ICMPv6 messages (any type) transmitted per second
663 from this entity (ipv6IfIcmpOutMsgs).
664 inerr/s Number of ICMPv6 messages received per second that had
666 ICMP-specific errors, such as bad ICMP checksums, bad
667 length, etc (ipv6IfIcmpInErrors).
668 innsol/s Number of ICMP Neighbor Solicit messages received per sec‐
670 ond (ipv6IfIcmpInNeighborSolicits).
671 innadv/s Number of ICMP Neighbor Advertisement messages received per
673 second (ipv6IfIcmpInNeighborAdvertisements).
674 otnsol/s Number of ICMP Neighbor Solicit messages transmitted per
676 second (ipv6IfIcmpOutNeighborSolicits).
677 otnadv/s Number of ICMP Neighbor Advertisement messages transmitted
679 per second (ipv6IfIcmpOutNeighborAdvertisements).
680 The output for the flag -K provides information about other types of
682 ICMPv6-messages (formal SNMP-names between brackets):
683 iecho/s Number of ICMP Echo (request) messages received per second
685 (ipv6IfIcmpInEchos).
686 ierep/s Number of ICMP Echo-Reply messages received per second
688 (ipv6IfIcmpInEchoReplies).
689 oerep/s Number of ICMP Echo-Reply messages transmitted per second
691 (ipv6IfIcmpOutEchoReplies).
692 idu/s Number of ICMP Destination Unreachable messages received
694 per second (ipv6IfIcmpInDestUnreachs).
695 odu/s Number of ICMP Destination Unreachable messages transmitted
697 per second (ipv6IfIcmpOutDestUnreachs).
698 ird/s Number of ICMP Redirect messages received per second
700 (ipv6IfIcmpInRedirects).
701 ord/s Number of ICMP Redirect messages transmitted per second
703 (ipv6IfIcmpOutRedirect).
704 ite/s Number of ICMP Time Exceeded messages received per second
706 (ipv6IfIcmpInTimeExcds).
707 ote/s Number of ICMP Time Exceeded messages transmitted per sec‐
709 ond (ipv6IfIcmpOutTimeExcds).
710 The output for the flag -U provides information about the utilization
712 of the UDPv6-layer (formal SNMP-names between brackets):
713 indgram/s Number of UDPv6 datagrams per second delivered to UDP users
715 (udpInDatagrams),
716 outdgram/s Number of UDPv6 datagrams transmitted per second from this
718 entity (udpOutDatagrams),
719 inerr/s Number of received UDPv6 datagrams per second that could
721 not be delivered for reasons other than the lack of an ap‐
722 plication at the destination port (udpInErrors).
723 noport/s Number of received UDPv6 datagrams per second for which
725 there was no application at the destination port (udpNo‐
726 Ports).
727 The output for the flag -t provides information about the utilization
729 of the TCP-layer (formal SNMP-names between brackets):
730 insegs/s Number of received segments per second, including those re‐
732 ceived in error (tcpInSegs).
733 outsegs/s Number of transmitted segments per second, excluding those
735 containing only retransmitted octets (tcpOutSegs).
736 actopen/s Number of active opens per second that have been supported
738 by this entity (tcpActiveOpens).
739 pasopen/s Number of passive opens per second that have been supported
741 by this entity (tcpPassiveOpens).
742 nowopen Number of connections currently open (snapshot), for which
744 the state is either ESTABLISHED or CLOSE-WAIT (tcpCur‐
745 rEstab).
746 The output for the flag -T provides information about the failures that
748 were detected in the TCP-layer (formal SNMP-names between brackets):
749 inerr/s Number of received segments per second received in error
751 (tcpInErrs).
752 retrans/s Number of retransmitted segments per second (tcpRe‐
754 transSegs).
755 attfail/s Number of failed connection attempts per second that have
757 occurred at this entity (tcpAttemptFails).
758 estabreset/s
760 Number of resets per second that have occurred at this en‐
761 tity (tcpEstabResets).
762 outreset/s Number of transmitted segments per second containing the
764 RST flag (tcpOutRsts).
765 The output for the flag -h provides information about utilization of
767 Infiniband ports:
768 controller Name of controller.
770 port Controller port.
772 busy Busy percentage for this port.
774 ipack/s Number of packets received from this port per second.
776 opack/s Number of packets transmitted to this port per second.
778 igbps/s Effective number of gigabits received per second.
780 ogbps/s Effective number of gigabits transmitted per second.
782 maxgbps/s Maximum rate as number of gigabits per second.
784 lanes Number of lanes.
786 The output for the flag -O provides information about the top-3 of pro‐
788 cesses with the highest processor consumption:
789 pid Process-id (if zero, the process has exited while the pid
791 could not be determined).
792 command The name of the process.
794 cpu% The percentage of cpu-capacity being consumed. This value
796 can exceed 100% for a multithreaded process running on a
797 multiprocessor machine.
798 The output for the flag -G provides information about the top-3 of pro‐
800 cesses with the highest memory consumption:
801 pid Process-id (if zero, the process has exited while the pid
803 could not be determined).
804 command The name of the process.
806 mem% The percentage of resident memory-utilization by this
808 process.
809 The output for the flag -D provides information about the top-3 of pro‐
811 cesses that issue the most read and write accesses to disk:
812 pid Process-id (if zero, the process has exited while the pid
814 could not be determined).
815 command The name of the process.
817 dsk% The percentage of read and write accesses related to the
819 total number of read and write accesses issued on disk by
820 all processes, so a high percentage does not imply a high
821 disk load on system level.
822 The output for the flag -N provides information about the top-3 of pro‐
824 cesses that issue the most socket transfers for IPv4/IPv6:
825 pid Process-id (if zero, the process has exited while the pid
827 could not be determined).
828 command The name of the process.
830 net% The percentage of socket transfers related to the total
832 number of transfers issued by all processes, so a high per‐
833 centage does not imply a high network load on system level.
834
836 To see today's cpu-activity so far (supposed that atop is logging in
837 the background):
838 atopsar
840 To see the memory occupation for June 5, 2018 between 10:00 and 12:30
842 (supposed that atop has been logging daily in the background):
843 atopsar -m -r /var/log/atop_20180605 -b 10:00 -e 12:30
845 or
847 atopsar -m -r 20180605 -b 10:00 -e 12:30
849 or, suppose it is June 8, 2018 at this moment
851 atopsar -m -r yyy -b 10:00 -e 12:30
853 Write a logfile with atop to record the system behaviour for 30 minutes
855 (30 samples of one minute) and produce all available reports after‐
856 wards:
857 atop -w /tmp/atoplog 60 30
859 atopsar -A -r /tmp/atoplog
861 To watch TCP activity evolve for ten minutes (10 samples with sixty
863 seconds interval):
864 atopsar -t 60 10
866 To watch the header-lines ('_' as last character) of all reports with
868 only the detail-lines showing critical resource consumption (marker '*'
869 or '+' as last character):
870 atopsar -AM | grep '[_*+]$'
872
874 /etc/atoprc
875 Configuration file containing system-wide default values (mainly
876 flags). See related man-page.
877 ~/.atoprc
879 Configuration file containing personal default values (mainly
880 flags). See related man-page.
881 /var/log/atop/atop_YYYYMMDD
883 Daily data file, where YYYYMMDD are digits representing the date.
884
886 atop(1), atopconvert(1), atopcat(1), atoprc(5), atopacctd(8), ne‐
887 tatop(4), netatopd(8)
888 https://www.atoptool.nl
889
891 Gerlof Langeveld (gerlof.langeveld@atoptool.nl)
892Linux December 2022 ATOPSAR(1)