1 .TH SAR 1 "DECEMBER 2010" Linux "Linux User's Manual" -*- nroff -*-
3 sar \- Collect, report, or save system activity information.
5 .B sar [ -A ] [ -b ] [ -B ] [ -C ] [ -d ] [ -h ] [ -H ] [ -i
7 .B ] [ -p ] [ -q ] [ -r ] [ -R ] [ -S ] [ -t ] [ -u [ ALL ] ]
8 [ -v ] [ -V ] [ -w ] [ -W ] [ -y ]
17 .B [,...] | SUM | ALL | XALL } ] [ -P {
19 .B [,...] | ALL } ] [ -o [
35 command writes to standard output the contents of selected
36 cumulative activity counters in the operating system. The accounting
37 system, based on the values in the
41 parameters, writes information the specified number of times spaced
42 at the specified intervals in seconds.
45 parameter is set to zero, the
47 command displays the average statistics for the time
48 since the system was started. If the
50 parameter is specified without the
52 parameter, then reports are generated continuously.
53 The collected data can also
54 be saved in the file specified by the
57 flag, in addition to being displayed onto the screen. If
59 is omitted, sar uses the standard system activity daily data file,
62 file, where the dd parameter indicates the current day.
63 By default all the data available from the kernel are saved in the
68 command extracts and writes to standard output records previously
69 saved in a file. This file can be either the one specified by the
71 flag or, by default, the standard system activity daily data file.
77 command reports system-wide (global among all processors) statistics,
78 which are calculated as averages for values expressed as percentages,
79 and as sums otherwise. If the
83 command reports activity which relates to the specified processor or
88 command reports statistics for each individual processor and global
89 statistics among all processors.
91 You can select information about specific system activities using
92 flags. Not specifying any flags selects only CPU activity.
95 flag selects all possible activities.
97 The default version of the
99 command (CPU utilization report) might be one of the first facilities
100 the user runs to begin system activity investigation, because it
101 monitors major system resources. If CPU utilization is near 100 percent
102 (user + nice + system), the workload sampled is CPU-bound.
104 If multiple samples and multiple reports are desired, it is convenient
105 to specify an output file for the
110 command as a background process. The syntax for this is:
112 .B sar -o datafile interval count >/dev/null 2>&1 &
114 All data are captured in binary form and saved to a file (datafile).
115 The data can then be selectively displayed with the
127 second intervals. If the
129 parameter is not set, all the records saved in the
130 file will be selected.
131 Collection of data in this manner is useful to characterize
132 system usage over a period of time and determine peak usage hours.
136 command only reports on local activities.
140 This is equivalent to specifying
141 .BR "-bBdHqrRSuvwWy -I SUM -I XALL -m ALL -n ALL -u ALL -P ALL".
143 Report I/O and transfer rate statistics.
144 The following values are displayed:
149 Total number of transfers per second that were issued to physical devices.
150 A transfer is an I/O request to a physical device. Multiple logical
151 requests can be combined into a single I/O request to the device.
152 A transfer is of indeterminate size.
157 Total number of read requests per second issued to physical devices.
162 Total number of write requests per second issued to physical devices.
167 Total amount of data read from the devices in blocks per second.
168 Blocks are equivalent to sectors
169 and therefore have a size of 512 bytes.
174 Total amount of data written to devices in blocks per second.
178 Report paging statistics.
179 The following values are displayed:
184 Total number of kilobytes the system paged in from disk per second.
189 Total number of kilobytes the system paged out to disk per second.
194 Number of page faults (major + minor) made by the system per second.
195 This is not a count of page faults that generate I/O, because some page
196 faults can be resolved without I/O.
201 Number of major faults the system has made per second, those which
202 have required loading a memory page from disk.
207 Number of pages placed on the free list by the system per second.
212 Number of pages scanned by the kswapd daemon per second.
217 Number of pages scanned directly per second.
222 Number of pages the system has reclaimed from cache (pagecache and
223 swapcache) per second to satisfy its memory demands.
228 Calculated as pgsteal / pgscan, this is a metric of the efficiency of
229 page reclaim. If it is near 100% then almost every page coming off the
230 tail of the inactive list is being reaped. If it gets too low (e.g. less
231 than 30%) then the virtual memory is having some difficulty.
232 This field is displayed as zero if no pages have been scanned during the
237 When reading data from a file, tell
239 to display comments that have been inserted by
242 Report activity for each block device.
243 When data are displayed, the device specification
249 is the major number of the device and
252 Device names may also be pretty-printed if option -p
254 Note that disk activity depends on sadc options "-S DISK" and "-S XDISK"
255 to be collected. The following values are displayed:
260 Indicate the number of transfers per second that were issued to the device.
261 Multiple logical requests can be combined into a single I/O request to the
262 device. A transfer is of indeterminate size.
267 Number of sectors read from the device. The size of a sector is 512 bytes.
272 Number of sectors written to the device. The size of a sector is 512 bytes.
277 The average size (in sectors) of the requests that were issued to the device.
282 The average queue length of the requests that were issued to the device.
287 The average time (in milliseconds) for I/O requests issued to the device
288 to be served. This includes the time spent by the requests in queue and
289 the time spent servicing them.
294 The average service time (in milliseconds) for I/O requests that were issued
295 to the device. Warning! Do not trust this field any more. This field will be
296 removed in a future sysstat version.
301 Percentage of CPU time during which I/O requests were issued to the device
302 (bandwidth utilization for the device). Device saturation occurs when this
303 value is close to 100%.
306 .IP "-e [ hh:mm:ss ]"
307 Set the ending time of the report. The default ending time is
308 18:00:00. Hours must be given in 24-hour format.
309 This option can be used when data are read from
310 or written to a file (options
315 .IP "-f [ filename ]"
320 flag). The default value of the
322 parameter is the current daily data file, the
324 file. The -f option is exclusive of the -o option.
326 Display a short help message then exit.
328 Report hugepages utilization statistics.
329 The following values are displayed:
334 Amount of hugepages memory in kilobytes that is not yet allocated.
339 Amount of hugepages memory in kilobytes that has been allocated.
344 Percentage of total hugepages memory that has been allocated.
348 Select data records at seconds as close as possible to the number specified
352 .IP "-I { int [,...] | SUM | ALL | XALL }"
353 Report statistics for a given interrupt.
355 is the interrupt number. Specifying multiple
357 parameters on the command line will look at multiple independent interrupts.
360 keyword indicates that the total number of interrupts received per second
361 is to be displayed. The
363 keyword indicates that statistics from
364 the first 16 interrupts are to be reported, whereas the
366 keyword indicates that statistics from all interrupts, including potential
367 APIC interrupt sources, are to be reported.
368 Note that interrupt statistics depend on sadc option "-S INT" to be collected.
369 .IP "-m { keyword [,...] | ALL }"
370 Report power management statistics.
371 Note that these statistics depend on sadc option "-S POWER" to be collected.
373 Possible keywords are
383 keyword, statistics about CPU are reported.
384 The following value is displayed:
389 Instantaneous CPU clock frequency in MHz.
394 keyword, statistics about fans speed are reported.
395 The following values are displayed:
399 Fan speed expressed in revolutions per minute.
404 This field is calculated as the difference between current fan speed (rpm)
405 and its low limit (fan_min).
415 keyword, statistics about CPU clock frequency are reported.
416 The following value is displayed:
420 Weighted average CPU clock frequency in MHz.
421 Note that the cpufreq-stats driver must be compiled in the
422 kernel for this option to work.
427 keyword, statistics about voltage inputs are reported.
428 The following values are displayed:
432 Voltage input expressed in Volts.
437 Relative input value. A value of 100% means that
438 voltage input has reached its high limit (in_max) whereas
439 a value of 0% means that it has reached its low limit (in_min).
449 keyword, statistics about devices temperature are reported.
450 The following values are displayed:
454 Device temperature expressed in degrees Celsius.
459 Relative device temperature. A value of 100% means that
460 temperature has reached its high limit (temp_max).
468 .IP "-n { keyword [,...] | ALL }"
469 Report network statistics.
471 Possible keywords are
494 keyword, statistics from the network devices are reported.
495 The following values are displayed:
500 Name of the network interface for which statistics are reported.
505 Total number of packets received per second.
510 Total number of packets transmitted per second.
515 Total number of kilobytes received per second.
520 Total number of kilobytes transmitted per second.
525 Number of compressed packets received per second (for cslip etc.).
530 Number of compressed packets transmitted per second.
535 Number of multicast packets received per second.
540 keyword, statistics on failures (errors) from the network devices are reported.
541 The following values are displayed:
545 Name of the network interface for which statistics are reported.
550 Total number of bad packets received per second.
555 Total number of errors that happened per second while transmitting packets.
560 Number of collisions that happened per second while transmitting packets.
565 Number of received packets dropped per second because of a lack of space in linux buffers.
570 Number of transmitted packets dropped per second because of a lack of space in linux buffers.
575 Number of carrier-errors that happened per second while transmitting packets.
580 Number of frame alignment errors that happened per second on received packets.
585 Number of FIFO overrun errors that happened per second on received packets.
590 Number of FIFO overrun errors that happened per second on transmitted packets.
595 keyword, statistics about NFS client activity are reported.
596 The following values are displayed:
600 Number of RPC requests made per second.
605 Number of RPC requests per second, those which needed to be retransmitted (for
606 example because of a server timeout).
611 Number of 'read' RPC calls made per second.
616 Number of 'write' RPC calls made per second.
621 Number of 'access' RPC calls made per second.
626 Number of 'getattr' RPC calls made per second.
631 keyword, statistics about NFS server activity are reported.
632 The following values are displayed:
636 Number of RPC requests received per second.
641 Number of bad RPC requests received per second, those whose
642 processing generated an error.
647 Number of network packets received per second.
652 Number of UDP packets received per second.
657 Number of TCP packets received per second.
662 Number of reply cache hits per second.
667 Number of reply cache misses per second.
672 Number of 'read' RPC calls received per second.
677 Number of 'write' RPC calls received per second.
682 Number of 'access' RPC calls received per second.
687 Number of 'getattr' RPC calls received per second.
692 keyword, statistics on sockets in use are reported
694 The following values are displayed:
698 Total number of sockets used by the system.
703 Number of TCP sockets currently in use.
708 Number of UDP sockets currently in use.
713 Number of RAW sockets currently in use.
718 Number of IP fragments currently in queue.
723 Number of TCP sockets in TIME_WAIT state.
728 keyword, statistics about IPv4 network traffic are reported.
729 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
731 The following values are displayed (formal SNMP names between
736 The total number of input datagrams received from interfaces
737 per second, including those received in error [ipInReceives].
742 The number of input datagrams per second, for which this entity was not
743 their final IP destination, as a result of which an attempt
744 was made to find a route to forward them to that final
745 destination [ipForwDatagrams].
750 The total number of input datagrams successfully delivered per second
751 to IP user-protocols (including ICMP) [ipInDelivers].
756 The total number of IP datagrams which local IP user-protocols (including ICMP)
757 supplied per second to IP in requests for transmission [ipOutRequests].
758 Note that this counter does not include any datagrams counted in fwddgm/s.
763 The number of IP fragments received per second which needed to be
764 reassembled at this entity [ipReasmReqds].
769 The number of IP datagrams successfully re-assembled per second [ipReasmOKs].
774 The number of IP datagrams that have been successfully
775 fragmented at this entity per second [ipFragOKs].
780 The number of IP datagram fragments that have been
781 generated per second as a result of fragmentation at this entity [ipFragCreates].
786 keyword, statistics about IPv4 network errors are reported.
787 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
789 The following values are displayed (formal SNMP names between
794 The number of input datagrams discarded per second due to errors in
795 their IP headers, including bad checksums, version number
796 mismatch, other format errors, time-to-live exceeded, errors
797 discovered in processing their IP options, etc. [ipInHdrErrors]
802 The number of input datagrams discarded per second because the IP
803 address in their IP header's destination field was not a
804 valid address to be received at this entity. This count
805 includes invalid addresses (e.g., 0.0.0.0) and addresses of
806 unsupported Classes (e.g., Class E). For entities which are
807 not IP routers and therefore do not forward datagrams, this
808 counter includes datagrams discarded because the destination
809 address was not a local address [ipInAddrErrors].
814 The number of locally-addressed datagrams received
815 successfully but discarded per second because of an unknown or
816 unsupported protocol [ipInUnknownProtos].
821 The number of input IP datagrams per second for which no problems were
822 encountered to prevent their continued processing, but which
823 were discarded (e.g., for lack of buffer space) [ipInDiscards].
824 Note that this counter does not include any datagrams discarded while
825 awaiting re-assembly.
830 The number of output IP datagrams per second for which no problem was
831 encountered to prevent their transmission to their
832 destination, but which were discarded (e.g., for lack of
833 buffer space) [ipOutDiscards].
834 Note that this counter would include
835 datagrams counted in fwddgm/s if any such packets met
836 this (discretionary) discard criterion.
841 The number of IP datagrams discarded per second because no route could
842 be found to transmit them to their destination [ipOutNoRoutes].
843 Note that this counter includes any packets counted in fwddgm/s
844 which meet this 'no-route' criterion.
845 Note that this includes any datagrams which a host cannot route because all
846 of its default routers are down.
851 The number of failures detected per second by the IP re-assembly
852 algorithm (for whatever reason: timed out, errors, etc) [ipReasmFails].
853 Note that this is not necessarily a count of discarded IP
854 fragments since some algorithms can lose track of the number of
855 fragments by combining them as they are received.
860 The number of IP datagrams that have been discarded per second because
861 they needed to be fragmented at this entity but could not
862 be, e.g., because their Don't Fragment flag was set [ipFragFails].
867 keyword, statistics about ICMPv4 network traffic are reported.
868 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
870 The following values are displayed (formal SNMP names between
875 The total number of ICMP messages which the entity
876 received per second [icmpInMsgs].
877 Note that this counter includes all those counted by ierr/s.
882 The total number of ICMP messages which this entity
883 attempted to send per second [icmpOutMsgs].
884 Note that this counter includes all those counted by oerr/s.
889 The number of ICMP Echo (request) messages received per second [icmpInEchos].
894 The number of ICMP Echo Reply messages received per second [icmpInEchoReps].
899 The number of ICMP Echo (request) messages sent per second [icmpOutEchos].
904 The number of ICMP Echo Reply messages sent per second [icmpOutEchoReps].
909 The number of ICMP Timestamp (request) messages received per second [icmpInTimestamps].
914 The number of ICMP Timestamp Reply messages received per second [icmpInTimestampReps].
919 The number of ICMP Timestamp (request) messages sent per second [icmpOutTimestamps].
924 The number of ICMP Timestamp Reply messages sent per second [icmpOutTimestampReps].
929 The number of ICMP Address Mask Request messages received per second [icmpInAddrMasks].
934 The number of ICMP Address Mask Reply messages received per second [icmpInAddrMaskReps].
939 The number of ICMP Address Mask Request messages sent per second [icmpOutAddrMasks].
944 The number of ICMP Address Mask Reply messages sent per second [icmpOutAddrMaskReps].
949 keyword, statistics about ICMPv4 error messages are reported.
950 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
952 The following values are displayed (formal SNMP names between
957 The number of ICMP messages per second which the entity received but
958 determined as having ICMP-specific errors (bad ICMP
959 checksums, bad length, etc.) [icmpInErrors].
964 The number of ICMP messages per second which this entity did not send
965 due to problems discovered within ICMP such as a lack of buffers [icmpOutErrors].
970 The number of ICMP Destination Unreachable messages
971 received per second [icmpInDestUnreachs].
976 The number of ICMP Destination Unreachable messages sent per second [icmpOutDestUnreachs].
981 The number of ICMP Time Exceeded messages received per second [icmpInTimeExcds].
986 The number of ICMP Time Exceeded messages sent per second [icmpOutTimeExcds].
991 The number of ICMP Parameter Problem messages received per second [icmpInParmProbs].
996 The number of ICMP Parameter Problem messages sent per second [icmpOutParmProbs].
1001 The number of ICMP Source Quench messages received per second [icmpInSrcQuenchs].
1006 The number of ICMP Source Quench messages sent per second [icmpOutSrcQuenchs].
1011 The number of ICMP Redirect messages received per second [icmpInRedirects].
1016 The number of ICMP Redirect messages sent per second [icmpOutRedirects].
1021 keyword, statistics about TCPv4 network traffic are reported.
1022 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1024 The following values are displayed (formal SNMP names between
1029 The number of times TCP connections have made a direct
1030 transition to the SYN-SENT state from the CLOSED state per second [tcpActiveOpens].
1035 The number of times TCP connections have made a direct
1036 transition to the SYN-RCVD state from the LISTEN state per second [tcpPassiveOpens].
1041 The total number of segments received per second, including those
1042 received in error [tcpInSegs]. This count includes segments received on
1043 currently established connections.
1048 The total number of segments sent per second, including those on
1049 current connections but excluding those containing only
1050 retransmitted octets [tcpOutSegs].
1055 keyword, statistics about TCPv4 network errors are reported.
1056 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1058 The following values are displayed (formal SNMP names between
1063 The number of times per second TCP connections have made a direct
1064 transition to the CLOSED state from either the SYN-SENT
1065 state or the SYN-RCVD state, plus the number of times per second TCP
1066 connections have made a direct transition to the LISTEN
1067 state from the SYN-RCVD state [tcpAttemptFails].
1072 The number of times per second TCP connections have made a direct
1073 transition to the CLOSED state from either the ESTABLISHED
1074 state or the CLOSE-WAIT state [tcpEstabResets].
1079 The total number of segments retransmitted per second - that is, the
1080 number of TCP segments transmitted containing one or more
1081 previously transmitted octets [tcpRetransSegs].
1086 The total number of segments received in error (e.g., bad
1087 TCP checksums) per second [tcpInErrs].
1092 The number of TCP segments sent per second containing the RST flag [tcpOutRsts].
1097 keyword, statistics about UDPv4 network traffic are reported.
1098 Note that UDPv4 statistics depend on sadc option "-S SNMP" to be
1100 The following values are displayed (formal SNMP names between
1105 The total number of UDP datagrams delivered per second to UDP users [udpInDatagrams].
1110 The total number of UDP datagrams sent per second from this entity [udpOutDatagrams].
1115 The total number of received UDP datagrams per second for which there
1116 was no application at the destination port [udpNoPorts].
1121 The number of received UDP datagrams per second that could not be
1122 delivered for reasons other than the lack of an application
1123 at the destination port [udpInErrors].
1128 keyword, statistics on sockets in use are reported (IPv6).
1129 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1131 The following values are displayed:
1135 Number of TCPv6 sockets currently in use.
1140 Number of UDPv6 sockets currently in use.
1145 Number of RAWv6 sockets currently in use.
1150 Number of IPv6 fragments currently in use.
1155 keyword, statistics about IPv6 network traffic are reported.
1156 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1158 The following values are displayed (formal SNMP names between
1163 The total number of input datagrams received from
1164 interfaces per second, including those received in error
1165 [ipv6IfStatsInReceives].
1170 The number of output datagrams per second which this
1171 entity received and forwarded to their final
1172 destinations [ipv6IfStatsOutForwDatagrams].
1177 The total number of datagrams successfully
1178 delivered per second to IPv6 user-protocols (including ICMP)
1179 [ipv6IfStatsInDelivers].
1184 The total number of IPv6 datagrams which local IPv6
1185 user-protocols (including ICMP) supplied per second to IPv6 in
1186 requests for transmission [ipv6IfStatsOutRequests].
1187 Note that this counter
1188 does not include any datagrams counted in fwddgm6/s.
1193 The number of IPv6 fragments received per second which needed
1194 to be reassembled at this interface [ipv6IfStatsReasmReqds].
1199 The number of IPv6 datagrams successfully
1200 reassembled per second [ipv6IfStatsReasmOKs].
1205 The number of multicast packets received per second
1206 by the interface [ipv6IfStatsInMcastPkts].
1211 The number of multicast packets transmitted per second
1212 by the interface [ipv6IfStatsOutMcastPkts].
1217 The number of IPv6 datagrams that have been
1218 successfully fragmented at this output interface per second
1219 [ipv6IfStatsOutFragOKs].
1224 The number of output datagram fragments that have
1225 been generated per second as a result of fragmentation at
1226 this output interface [ipv6IfStatsOutFragCreates].
1231 keyword, statistics about IPv6 network errors are reported.
1232 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1234 The following values are displayed (formal SNMP names between
1239 The number of input datagrams discarded per second due to
1240 errors in their IPv6 headers, including version
1241 number mismatch, other format errors, hop count
1242 exceeded, errors discovered in processing their
1243 IPv6 options, etc. [ipv6IfStatsInHdrErrors]
1248 The number of input datagrams discarded per second because
1249 the IPv6 address in their IPv6 header's destination
1250 field was not a valid address to be received at
1251 this entity. This count includes invalid
1252 addresses (e.g., ::0) and unsupported addresses
1253 (e.g., addresses with unallocated prefixes). For
1254 entities which are not IPv6 routers and therefore
1255 do not forward datagrams, this counter includes
1256 datagrams discarded because the destination address
1257 was not a local address [ipv6IfStatsInAddrErrors].
1262 The number of locally-addressed datagrams
1263 received successfully but discarded per second because of an
1264 unknown or unsupported protocol [ipv6IfStatsInUnknownProtos].
1269 The number of input datagrams that could not be
1270 forwarded per second because their size exceeded the link MTU
1271 of outgoing interface [ipv6IfStatsInTooBigErrors].
1276 The number of input IPv6 datagrams per second for which no
1277 problems were encountered to prevent their
1278 continued processing, but which were discarded
1279 (e.g., for lack of buffer space)
1280 [ipv6IfStatsInDiscards]. Note that this
1281 counter does not include any datagrams discarded
1282 while awaiting re-assembly.
1287 The number of output IPv6 datagrams per second for which no
1288 problem was encountered to prevent their
1289 transmission to their destination, but which were
1290 discarded (e.g., for lack of buffer space)
1291 [ipv6IfStatsOutDiscards]. Note
1292 that this counter would include datagrams counted
1293 in fwddgm6/s if any such packets
1294 met this (discretionary) discard criterion.
1299 The number of input datagrams discarded per second because no
1300 route could be found to transmit them to their
1301 destination [ipv6IfStatsInNoRoutes].
1306 The number of locally generated IP datagrams discarded per second
1307 because no route could be found to transmit them to their
1308 destination [unknown formal SNMP name].
1313 The number of failures detected per second by the IPv6
1314 re-assembly algorithm (for whatever reason: timed
1315 out, errors, etc.) [ipv6IfStatsReasmFails].
1316 Note that this is not
1317 necessarily a count of discarded IPv6 fragments
1318 since some algorithms
1319 can lose track of the number of fragments
1320 by combining them as they are received.
1325 The number of IPv6 datagrams that have been
1326 discarded per second because they needed to be fragmented
1327 at this output interface but could not be
1328 [ipv6IfStatsOutFragFails].
1333 The number of input datagrams discarded per second because
1334 datagram frame didn't carry enough data
1335 [ipv6IfStatsInTruncatedPkts].
1340 keyword, statistics about ICMPv6 network traffic are reported.
1341 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1343 The following values are displayed (formal SNMP names between
1348 The total number of ICMP messages received
1349 by the interface per second which includes all those
1350 counted by ierr6/s [ipv6IfIcmpInMsgs].
1355 The total number of ICMP messages which this
1356 interface attempted to send per second [ipv6IfIcmpOutMsgs].
1361 The number of ICMP Echo (request) messages
1362 received by the interface per second [ipv6IfIcmpInEchos].
1367 The number of ICMP Echo Reply messages received
1368 by the interface per second [ipv6IfIcmpInEchoReplies].
1373 The number of ICMP Echo Reply messages sent
1374 by the interface per second [ipv6IfIcmpOutEchoReplies].
1379 The number of ICMPv6 Group Membership Query
1380 messages received by the interface per second
1381 [ipv6IfIcmpInGroupMembQueries].
1386 The number of ICMPv6 Group Membership Response messages
1387 received by the interface per second
1388 [ipv6IfIcmpInGroupMembResponses].
1393 The number of ICMPv6 Group Membership Response
1394 messages sent per second
1395 [ipv6IfIcmpOutGroupMembResponses].
1400 The number of ICMPv6 Group Membership Reduction messages
1401 received by the interface per second
1402 [ipv6IfIcmpInGroupMembReductions].
1407 The number of ICMPv6 Group Membership Reduction
1408 messages sent per second
1409 [ipv6IfIcmpOutGroupMembReductions].
1414 The number of ICMP Router Solicit messages
1415 received by the interface per second
1416 [ipv6IfIcmpInRouterSolicits].
1421 The number of ICMP Router Solicitation messages
1422 sent by the interface per second
1423 [ipv6IfIcmpOutRouterSolicits].
1428 The number of ICMP Router Advertisement messages
1429 received by the interface per second
1430 [ipv6IfIcmpInRouterAdvertisements].
1435 The number of ICMP Neighbor Solicit messages
1436 received by the interface per second
1437 [ipv6IfIcmpInNeighborSolicits].
1442 The number of ICMP Neighbor Solicitation
1443 messages sent by the interface per second
1444 [ipv6IfIcmpOutNeighborSolicits].
1449 The number of ICMP Neighbor Advertisement
1450 messages received by the interface per second
1451 [ipv6IfIcmpInNeighborAdvertisements].
1456 The number of ICMP Neighbor Advertisement
1457 messages sent by the interface per second
1458 [ipv6IfIcmpOutNeighborAdvertisements].
1463 keyword, statistics about ICMPv6 error messages are reported.
1464 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1466 The following values are displayed (formal SNMP names between
1471 The number of ICMP messages per second which the interface
1472 received but determined as having ICMP-specific
1473 errors (bad ICMP checksums, bad length, etc.)
1474 [ipv6IfIcmpInErrors]
1479 The number of ICMP Destination Unreachable
1480 messages received by the interface per second
1481 [ipv6IfIcmpInDestUnreachs].
1486 The number of ICMP Destination Unreachable
1487 messages sent by the interface per second
1488 [ipv6IfIcmpOutDestUnreachs].
1493 The number of ICMP Time Exceeded messages
1494 received by the interface per second
1495 [ipv6IfIcmpInTimeExcds].
1500 The number of ICMP Time Exceeded messages sent
1501 by the interface per second
1502 [ipv6IfIcmpOutTimeExcds].
1507 The number of ICMP Parameter Problem messages
1508 received by the interface per second
1509 [ipv6IfIcmpInParmProblems].
1514 The number of ICMP Parameter Problem messages
1515 sent by the interface per second
1516 [ipv6IfIcmpOutParmProblems].
1521 The number of Redirect messages received
1522 by the interface per second
1523 [ipv6IfIcmpInRedirects].
1528 The number of Redirect messages sent by
1529 the interface by second
1530 [ipv6IfIcmpOutRedirects].
1535 The number of ICMP Packet Too Big messages
1536 received by the interface per second
1537 [ipv6IfIcmpInPktTooBigs].
1542 The number of ICMP Packet Too Big messages sent
1543 by the interface per second
1544 [ipv6IfIcmpOutPktTooBigs].
1549 keyword, statistics about UDPv6 network traffic are reported.
1550 Note that UDPv6 statistics depend on sadc option "-S IPV6" to be
1552 The following values are displayed (formal SNMP names between
1557 The total number of UDP datagrams delivered per second to UDP users
1563 The total number of UDP datagrams sent per second from this
1564 entity [udpOutDatagrams].
1569 The total number of received UDP datagrams per second for which there
1570 was no application at the destination port [udpNoPorts].
1575 The number of received UDP datagrams per second that could not be
1576 delivered for reasons other than the lack of an application
1577 at the destination port [udpInErrors].
1582 keyword is equivalent to specifying all the keywords above and therefore all the network
1583 activities are reported.
1586 .IP "-o [ filename ]"
1587 Save the readings in the file in binary form. Each reading
1588 is in a separate record. The default value of the
1590 parameter is the current daily data file, the
1592 file. The -o option is exclusive of the -f option.
1593 All the data available from the kernel are saved in the file (in fact,
1595 calls its data collector
1597 with the option "-S ALL". See sadc(8) manual page).
1598 .IP "-P { cpu [,...] | ALL }"
1599 Report per-processor statistics for the specified processor or processors.
1602 keyword reports statistics for each individual processor, and globally for
1604 Note that processor 0 is the first processor.
1606 Pretty-print device names. Use this option in conjunction with option -d.
1607 By default names are printed as
1609 where m and n are the major and minor numbers for the device.
1610 Use of this option displays the names of the devices as they (should) appear
1611 in /dev. Name mappings are controlled by
1612 .B @SYSCONFIG_DIR@/sysstat.ioconf.
1614 Report queue length and load averages. The following values are displayed:
1619 Run queue length (number of tasks waiting for run time).
1624 Number of tasks in the task list.
1629 System load average for the last minute.
1630 The load average is calculated as the average number of runnable or
1631 running tasks (R state), and the number of tasks in uninterruptible
1632 sleep (D state) over the specified interval.
1637 System load average for the past 5 minutes.
1642 System load average for the past 15 minutes.
1647 Number of tasks currently blocked, waiting for I/O to complete.
1651 Report memory utilization statistics.
1652 The following values are displayed:
1657 Amount of free memory available in kilobytes.
1662 Amount of used memory in kilobytes. This does not take into account memory
1663 used by the kernel itself.
1668 Percentage of used memory.
1673 Amount of memory used as buffers by the kernel in kilobytes.
1678 Amount of memory used to cache data by the kernel in kilobytes.
1683 Amount of memory in kilobytes needed for current workload. This is an estimate of how much
1684 RAM/swap is needed to guarantee that there never is out of memory.
1689 Percentage of memory needed for current workload in relation to the total amount of memory (RAM+swap).
1690 This number may be greater than 100% because the kernel usually overcommits memory.
1695 Report memory statistics. The following values are displayed:
1700 Number of memory pages freed by the system per second.
1701 A negative value represents a number of pages allocated by the system.
1702 Note that a page has a size of 4 kB or 8 kB according to the machine architecture.
1707 Number of additional memory pages used as buffers by the system per second.
1708 A negative value means fewer pages used as buffers by the system.
1713 Number of additional memory pages cached by the system per second.
1714 A negative value means fewer pages in the cache.
1717 .IP "-s [ hh:mm:ss ]"
1718 Set the starting time of the data, causing the
1720 command to extract records time-tagged at, or following, the time
1721 specified. The default starting time is 08:00:00.
1722 Hours must be given in 24-hour format. This option can be
1723 used only when data are read from a file (option
1727 Report swap space utilization statistics.
1728 The following values are displayed:
1733 Amount of free swap space in kilobytes.
1738 Amount of used swap space in kilobytes.
1743 Percentage of used swap space.
1748 Amount of cached swap memory in kilobytes.
1749 This is memory that once was swapped out, is swapped back in
1750 but still also is in the swap area (if memory is needed it doesn't need
1751 to be swapped out again because it is already in the swap area. This
1757 Percentage of cached swap memory in relation to the amount of used swap space.
1761 When reading data from a daily data file, indicate that
1763 should display the timestamps in the original locale time of
1764 the data file creator. Without this option, the
1766 command displays the timestamps in the user's locale time.
1768 Report CPU utilization. The
1770 keyword indicates that all the CPU fields should be displayed.
1771 The report may show the following fields:
1776 Percentage of CPU utilization that occurred while executing at the user
1777 level (application). Note that this field includes time spent running
1783 Percentage of CPU utilization that occurred while executing at the user
1784 level (application). Note that this field does NOT include time spent
1785 running virtual processors.
1790 Percentage of CPU utilization that occurred while executing at the user
1791 level with nice priority.
1796 Percentage of CPU utilization that occurred while executing at the system
1797 level (kernel). Note that this field includes time spent servicing
1798 hardware and software interrupts.
1803 Percentage of CPU utilization that occurred while executing at the system
1804 level (kernel). Note that this field does NOT include time spent servicing
1805 hardware or software interrupts.
1810 Percentage of time that the CPU or CPUs were idle during which
1811 the system had an outstanding disk I/O request.
1816 Percentage of time spent in involuntary wait by the virtual CPU
1817 or CPUs while the hypervisor was servicing another virtual processor.
1822 Percentage of time spent by the CPU or CPUs to service hardware interrupts.
1827 Percentage of time spent by the CPU or CPUs to service software interrupts.
1832 Percentage of time spent by the CPU or CPUs to run a virtual processor.
1837 Percentage of time that the CPU or CPUs were idle and the system
1838 did not have an outstanding disk I/O request.
1841 Note: On SMP machines a processor that does not have any activity at all
1842 (0.00 for every field) is a disabled (offline) processor.
1845 Report status of inode, file and other kernel tables.
1846 The following values are displayed:
1851 Number of unused cache entries in the directory cache.
1856 Number of file handles used by the system.
1861 Number of inode handlers used by the system.
1866 Number of pseudo-terminals used by the system.
1870 Print version number then exit.
1872 Report task creation and system switching activity.
1877 Total number of tasks created per second.
1882 Total number of context switches per second.
1886 Report swapping statistics. The following values are displayed:
1891 Total number of swap pages the system brought in per second.
1896 Total number of swap pages the system brought out per second.
1900 Report TTY device activity. The following values are displayed:
1905 Number of receive interrupts per second for current serial line. Serial line number
1906 is given in the TTY column.
1911 Number of transmit interrupts per second for current serial line.
1916 Number of frame errors per second for current serial line.
1921 Number of parity errors per second for current serial line.
1926 Number of breaks per second for current serial line.
1931 Number of overrun errors per second for current serial line.
1938 command takes into account the following environment variables:
1941 If this variable exists and its value is
1943 then the current locale will be ignored when printing the date in the report header.
1946 command will use the ISO 8601 format (YYYY-MM-DD) instead.
1949 If this variable exists and its value is
1953 will save its data in UTC time (data will still be displayed in local time).
1955 will also use UTC time instead of local time to determine the current daily
1956 data file located in the
1958 directory. This variable may be useful for servers with users located across
1963 Report CPU utilization for each 2 seconds. 5 lines are displayed.
1966 .B sar -I 14 -o int14.file 2 10
1968 Report statistics on IRQ 14 for each 2 seconds. 10 lines are displayed.
1969 Data are stored in a file called
1973 .B sar -r -n DEV -f @SA_DIR@/sa16
1975 Display memory and network statistics saved in daily data file 'sa16'.
1980 Display all the statistics saved in current daily data file.
1983 filesystem must be mounted for the
1987 All the statistics are not necessarily available, depending on the kernel version used.
1989 assumes that you are using at least a 2.6 kernel.
1993 Indicate the daily data file, where the
1995 parameter is a number representing the day of the month.
1999 contains various files with system statistics.
2001 Sebastien Godard (sysstat <at> orange.fr)
2012 .I http://pagesperso-orange.fr/sebastien.godard/