1 .TH SAR 1 "OCTOBER 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 is equivalent to specifying
96 .BR "-bBdqrRSvwWy -I SUM -I XALL -n ALL -u ALL -P ALL".
98 The default version of the
100 command (CPU utilization report) might be one of the first facilities
101 the user runs to begin system activity investigation, because it
102 monitors major system resources. If CPU utilization is near 100 percent
103 (user + nice + system), the workload sampled is CPU-bound.
105 If multiple samples and multiple reports are desired, it is convenient
106 to specify an output file for the
111 command as a background process. The syntax for this is:
113 .B sar -o datafile interval count >/dev/null 2>&1 &
115 All data are captured in binary form and saved to a file (datafile).
116 The data can then be selectively displayed with the
128 second intervals. If the
130 parameter is not set, all the records saved in the
131 file will be selected.
132 Collection of data in this manner is useful to characterize
133 system usage over a period of time and determine peak usage hours.
137 command only reports on local activities.
141 This is equivalent to specifying
142 .BR "-bBdHqrRSuvwWy -I SUM -I XALL -m ALL -n ALL -u ALL -P ALL".
144 Report I/O and transfer rate statistics.
145 The following values are displayed:
150 Total number of transfers per second that were issued to physical devices.
151 A transfer is an I/O request to a physical device. Multiple logical
152 requests can be combined into a single I/O request to the device.
153 A transfer is of indeterminate size.
158 Total number of read requests per second issued to physical devices.
163 Total number of write requests per second issued to physical devices.
168 Total amount of data read from the devices in blocks per second.
169 Blocks are equivalent to sectors
170 and therefore have a size of 512 bytes.
175 Total amount of data written to devices in blocks per second.
179 Report paging statistics.
180 The following values are displayed:
185 Total number of kilobytes the system paged in from disk per second.
190 Total number of kilobytes the system paged out to disk per second.
195 Number of page faults (major + minor) made by the system per second.
196 This is not a count of page faults that generate I/O, because some page
197 faults can be resolved without I/O.
202 Number of major faults the system has made per second, those which
203 have required loading a memory page from disk.
208 Number of pages placed on the free list by the system per second.
213 Number of pages scanned by the kswapd daemon per second.
218 Number of pages scanned directly per second.
223 Number of pages the system has reclaimed from cache (pagecache and
224 swapcache) per second to satisfy its memory demands.
229 Calculated as pgsteal / pgscan, this is a metric of the efficiency of
230 page reclaim. If it is near 100% then almost every page coming off the
231 tail of the inactive list is being reaped. If it gets too low (e.g. less
232 than 30%) then the virtual memory is having some difficulty.
233 This field is displayed as zero if no pages have been scanned during the
238 When reading data from a file, tell
240 to display comments that have been inserted by
243 Report activity for each block device.
244 When data are displayed, the device specification
250 is the major number of the device and
253 Device names may also be pretty-printed if option -p
255 Note that disk activity depends on sadc options "-S DISK" and "-S XDISK"
256 to be collected. The following values are displayed:
261 Indicate the number of transfers per second that were issued to the device.
262 Multiple logical requests can be combined into a single I/O request to the
263 device. A transfer is of indeterminate size.
268 Number of sectors read from the device. The size of a sector is 512 bytes.
273 Number of sectors written to the device. The size of a sector is 512 bytes.
278 The average size (in sectors) of the requests that were issued to the device.
283 The average queue length of the requests that were issued to the device.
288 The average time (in milliseconds) for I/O requests issued to the device
289 to be served. This includes the time spent by the requests in queue and
290 the time spent servicing them.
295 The average service time (in milliseconds) for I/O requests that were issued
296 to the device. Warning! Do not trust this field any more. This field will be
297 removed in a future sysstat version.
302 Percentage of CPU time during which I/O requests were issued to the device
303 (bandwidth utilization for the device). Device saturation occurs when this
304 value is close to 100%.
307 .IP "-e [ hh:mm:ss ]"
308 Set the ending time of the report. The default ending time is
309 18:00:00. Hours must be given in 24-hour format.
310 This option can be used when data are read from
311 or written to a file (options
316 .IP "-f [ filename ]"
321 flag). The default value of the
323 parameter is the current daily data file, the
325 file. The -f option is exclusive of the -o option.
327 Display a short help message then exit.
329 Report hugepages utilization statistics.
330 The following values are displayed:
335 Amount of hugepages memory in kilobytes that is not yet allocated.
340 Amount of hugepages memory in kilobytes that has been allocated.
345 Percentage of total hugepages memory that has been allocated.
349 Select data records at seconds as close as possible to the number specified
353 .IP "-I { int [,...] | SUM | ALL | XALL }"
354 Report statistics for a given interrupt.
356 is the interrupt number. Specifying multiple
358 parameters on the command line will look at multiple independent interrupts.
361 keyword indicates that the total number of interrupts received per second
362 is to be displayed. The
364 keyword indicates that statistics from
365 the first 16 interrupts are to be reported, whereas the
367 keyword indicates that statistics from all interrupts, including potential
368 APIC interrupt sources, are to be reported.
369 Note that interrupt statistics depend on sadc option "-S INT" to be collected.
370 .IP "-m { keyword [,...] | ALL }"
371 Report power management statistics.
372 Note that these statistics depend on sadc option "-S POWER" to be collected.
374 Possible keywords are
384 keyword, statistics about CPU are reported.
385 The following value is displayed:
390 Instantaneous CPU clock frequency in MHz.
395 keyword, statistics about fans speed are reported.
396 The following values are displayed:
400 Fan speed expressed in revolutions per minute.
405 This field is calculated as the difference between current fan speed (rpm)
406 and its low limit (fan_min).
416 keyword, statistics about CPU clock frequency are reported.
417 The following value is displayed:
421 Weighted average CPU clock frequency in MHz.
422 Note that the cpufreq-stats driver must be compiled in the
423 kernel for this option to work.
428 keyword, statistics about voltage inputs are reported.
429 The following values are displayed:
433 Voltage input expressed in Volts.
438 Relative input value. A value of 100% means that
439 voltage input has reached its high limit (in_max) whereas
440 a value of 0% means that it has reached its low limit (in_min).
450 keyword, statistics about devices temperature are reported.
451 The following values are displayed:
455 Device temperature expressed in degrees Celsius.
460 Relative device temperature. A value of 100% means that
461 temperature has reached its high limit (temp_max).
469 .IP "-n { keyword [,...] | ALL }"
470 Report network statistics.
472 Possible keywords are
495 keyword, statistics from the network devices are reported.
496 The following values are displayed:
501 Name of the network interface for which statistics are reported.
506 Total number of packets received per second.
511 Total number of packets transmitted per second.
516 Total number of kilobytes received per second.
521 Total number of kilobytes transmitted per second.
526 Number of compressed packets received per second (for cslip etc.).
531 Number of compressed packets transmitted per second.
536 Number of multicast packets received per second.
541 keyword, statistics on failures (errors) from the network devices are reported.
542 The following values are displayed:
546 Name of the network interface for which statistics are reported.
551 Total number of bad packets received per second.
556 Total number of errors that happened per second while transmitting packets.
561 Number of collisions that happened per second while transmitting packets.
566 Number of received packets dropped per second because of a lack of space in linux buffers.
571 Number of transmitted packets dropped per second because of a lack of space in linux buffers.
576 Number of carrier-errors that happened per second while transmitting packets.
581 Number of frame alignment errors that happened per second on received packets.
586 Number of FIFO overrun errors that happened per second on received packets.
591 Number of FIFO overrun errors that happened per second on transmitted packets.
596 keyword, statistics about NFS client activity are reported.
597 The following values are displayed:
601 Number of RPC requests made per second.
606 Number of RPC requests per second, those which needed to be retransmitted (for
607 example because of a server timeout).
612 Number of 'read' RPC calls made per second.
617 Number of 'write' RPC calls made per second.
622 Number of 'access' RPC calls made per second.
627 Number of 'getattr' RPC calls made per second.
632 keyword, statistics about NFS server activity are reported.
633 The following values are displayed:
637 Number of RPC requests received per second.
642 Number of bad RPC requests received per second, those whose
643 processing generated an error.
648 Number of network packets received per second.
653 Number of UDP packets received per second.
658 Number of TCP packets received per second.
663 Number of reply cache hits per second.
668 Number of reply cache misses per second.
673 Number of 'read' RPC calls received per second.
678 Number of 'write' RPC calls received per second.
683 Number of 'access' RPC calls received per second.
688 Number of 'getattr' RPC calls received per second.
693 keyword, statistics on sockets in use are reported
695 The following values are displayed:
699 Total number of sockets used by the system.
704 Number of TCP sockets currently in use.
709 Number of UDP sockets currently in use.
714 Number of RAW sockets currently in use.
719 Number of IP fragments currently in use.
724 Number of TCP sockets in TIME_WAIT state.
729 keyword, statistics about IPv4 network traffic are reported.
730 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
732 The following values are displayed (formal SNMP names between
737 The total number of input datagrams received from interfaces
738 per second, including those received in error [ipInReceives].
743 The number of input datagrams per second, for which this entity was not
744 their final IP destination, as a result of which an attempt
745 was made to find a route to forward them to that final
746 destination [ipForwDatagrams].
751 The total number of input datagrams successfully delivered per second
752 to IP user-protocols (including ICMP) [ipInDelivers].
757 The total number of IP datagrams which local IP user-protocols (including ICMP)
758 supplied per second to IP in requests for transmission [ipOutRequests].
759 Note that this counter does not include any datagrams counted in fwddgm/s.
764 The number of IP fragments received per second which needed to be
765 reassembled at this entity [ipReasmReqds].
770 The number of IP datagrams successfully re-assembled per second [ipReasmOKs].
775 The number of IP datagrams that have been successfully
776 fragmented at this entity per second [ipFragOKs].
781 The number of IP datagram fragments that have been
782 generated per second as a result of fragmentation at this entity [ipFragCreates].
787 keyword, statistics about IPv4 network errors are reported.
788 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
790 The following values are displayed (formal SNMP names between
795 The number of input datagrams discarded per second due to errors in
796 their IP headers, including bad checksums, version number
797 mismatch, other format errors, time-to-live exceeded, errors
798 discovered in processing their IP options, etc. [ipInHdrErrors]
803 The number of input datagrams discarded per second because the IP
804 address in their IP header's destination field was not a
805 valid address to be received at this entity. This count
806 includes invalid addresses (e.g., 0.0.0.0) and addresses of
807 unsupported Classes (e.g., Class E). For entities which are
808 not IP routers and therefore do not forward datagrams, this
809 counter includes datagrams discarded because the destination
810 address was not a local address [ipInAddrErrors].
815 The number of locally-addressed datagrams received
816 successfully but discarded per second because of an unknown or
817 unsupported protocol [ipInUnknownProtos].
822 The number of input IP datagrams per second for which no problems were
823 encountered to prevent their continued processing, but which
824 were discarded (e.g., for lack of buffer space) [ipInDiscards].
825 Note that this counter does not include any datagrams discarded while
826 awaiting re-assembly.
831 The number of output IP datagrams per second for which no problem was
832 encountered to prevent their transmission to their
833 destination, but which were discarded (e.g., for lack of
834 buffer space) [ipOutDiscards].
835 Note that this counter would include
836 datagrams counted in fwddgm/s if any such packets met
837 this (discretionary) discard criterion.
842 The number of IP datagrams discarded per second because no route could
843 be found to transmit them to their destination [ipOutNoRoutes].
844 Note that this counter includes any packets counted in fwddgm/s
845 which meet this 'no-route' criterion.
846 Note that this includes any datagrams which a host cannot route because all
847 of its default routers are down.
852 The number of failures detected per second by the IP re-assembly
853 algorithm (for whatever reason: timed out, errors, etc) [ipReasmFails].
854 Note that this is not necessarily a count of discarded IP
855 fragments since some algorithms can lose track of the number of
856 fragments by combining them as they are received.
861 The number of IP datagrams that have been discarded per second because
862 they needed to be fragmented at this entity but could not
863 be, e.g., because their Don't Fragment flag was set [ipFragFails].
868 keyword, statistics about ICMPv4 network traffic are reported.
869 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
871 The following values are displayed (formal SNMP names between
876 The total number of ICMP messages which the entity
877 received per second [icmpInMsgs].
878 Note that this counter includes all those counted by ierr/s.
883 The total number of ICMP messages which this entity
884 attempted to send per second [icmpOutMsgs].
885 Note that this counter includes all those counted by oerr/s.
890 The number of ICMP Echo (request) messages received per second [icmpInEchos].
895 The number of ICMP Echo Reply messages received per second [icmpInEchoReps].
900 The number of ICMP Echo (request) messages sent per second [icmpOutEchos].
905 The number of ICMP Echo Reply messages sent per second [icmpOutEchoReps].
910 The number of ICMP Timestamp (request) messages received per second [icmpInTimestamps].
915 The number of ICMP Timestamp Reply messages received per second [icmpInTimestampReps].
920 The number of ICMP Timestamp (request) messages sent per second [icmpOutTimestamps].
925 The number of ICMP Timestamp Reply messages sent per second [icmpOutTimestampReps].
930 The number of ICMP Address Mask Request messages received per second [icmpInAddrMasks].
935 The number of ICMP Address Mask Reply messages received per second [icmpInAddrMaskReps].
940 The number of ICMP Address Mask Request messages sent per second [icmpOutAddrMasks].
945 The number of ICMP Address Mask Reply messages sent per second [icmpOutAddrMaskReps].
950 keyword, statistics about ICMPv4 error messages are reported.
951 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
953 The following values are displayed (formal SNMP names between
958 The number of ICMP messages per second which the entity received but
959 determined as having ICMP-specific errors (bad ICMP
960 checksums, bad length, etc.) [icmpInErrors].
965 The number of ICMP messages per second which this entity did not send
966 due to problems discovered within ICMP such as a lack of buffers [icmpOutErrors].
971 The number of ICMP Destination Unreachable messages
972 received per second [icmpInDestUnreachs].
977 The number of ICMP Destination Unreachable messages sent per second [icmpOutDestUnreachs].
982 The number of ICMP Time Exceeded messages received per second [icmpInTimeExcds].
987 The number of ICMP Time Exceeded messages sent per second [icmpOutTimeExcds].
992 The number of ICMP Parameter Problem messages received per second [icmpInParmProbs].
997 The number of ICMP Parameter Problem messages sent per second [icmpOutParmProbs].
1002 The number of ICMP Source Quench messages received per second [icmpInSrcQuenchs].
1007 The number of ICMP Source Quench messages sent per second [icmpOutSrcQuenchs].
1012 The number of ICMP Redirect messages received per second [icmpInRedirects].
1017 The number of ICMP Redirect messages sent per second [icmpOutRedirects].
1022 keyword, statistics about TCPv4 network traffic are reported.
1023 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1025 The following values are displayed (formal SNMP names between
1030 The number of times TCP connections have made a direct
1031 transition to the SYN-SENT state from the CLOSED state per second [tcpActiveOpens].
1036 The number of times TCP connections have made a direct
1037 transition to the SYN-RCVD state from the LISTEN state per second [tcpPassiveOpens].
1042 The total number of segments received per second, including those
1043 received in error [tcpInSegs]. This count includes segments received on
1044 currently established connections.
1049 The total number of segments sent per second, including those on
1050 current connections but excluding those containing only
1051 retransmitted octets [tcpOutSegs].
1056 keyword, statistics about TCPv4 network errors are reported.
1057 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1059 The following values are displayed (formal SNMP names between
1064 The number of times per second TCP connections have made a direct
1065 transition to the CLOSED state from either the SYN-SENT
1066 state or the SYN-RCVD state, plus the number of times per second TCP
1067 connections have made a direct transition to the LISTEN
1068 state from the SYN-RCVD state [tcpAttemptFails].
1073 The number of times per second TCP connections have made a direct
1074 transition to the CLOSED state from either the ESTABLISHED
1075 state or the CLOSE-WAIT state [tcpEstabResets].
1080 The total number of segments retransmitted per second - that is, the
1081 number of TCP segments transmitted containing one or more
1082 previously transmitted octets [tcpRetransSegs].
1087 The total number of segments received in error (e.g., bad
1088 TCP checksums) per second [tcpInErrs].
1093 The number of TCP segments sent per second containing the RST flag [tcpOutRsts].
1098 keyword, statistics about UDPv4 network traffic are reported.
1099 Note that UDPv4 statistics depend on sadc option "-S SNMP" to be
1101 The following values are displayed (formal SNMP names between
1106 The total number of UDP datagrams delivered per second to UDP users [udpInDatagrams].
1111 The total number of UDP datagrams sent per second from this entity [udpOutDatagrams].
1116 The total number of received UDP datagrams per second for which there
1117 was no application at the destination port [udpNoPorts].
1122 The number of received UDP datagrams per second that could not be
1123 delivered for reasons other than the lack of an application
1124 at the destination port [udpInErrors].
1129 keyword, statistics on sockets in use are reported (IPv6).
1130 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1132 The following values are displayed:
1136 Number of TCPv6 sockets currently in use.
1141 Number of UDPv6 sockets currently in use.
1146 Number of RAWv6 sockets currently in use.
1151 Number of IPv6 fragments currently in use.
1156 keyword, statistics about IPv6 network traffic are reported.
1157 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1159 The following values are displayed (formal SNMP names between
1164 The total number of input datagrams received from
1165 interfaces per second, including those received in error
1166 [ipv6IfStatsInReceives].
1171 The number of output datagrams per second which this
1172 entity received and forwarded to their final
1173 destinations [ipv6IfStatsOutForwDatagrams].
1178 The total number of datagrams successfully
1179 delivered per second to IPv6 user-protocols (including ICMP)
1180 [ipv6IfStatsInDelivers].
1185 The total number of IPv6 datagrams which local IPv6
1186 user-protocols (including ICMP) supplied per second to IPv6 in
1187 requests for transmission [ipv6IfStatsOutRequests].
1188 Note that this counter
1189 does not include any datagrams counted in fwddgm6/s.
1194 The number of IPv6 fragments received per second which needed
1195 to be reassembled at this interface [ipv6IfStatsReasmReqds].
1200 The number of IPv6 datagrams successfully
1201 reassembled per second [ipv6IfStatsReasmOKs].
1206 The number of multicast packets received per second
1207 by the interface [ipv6IfStatsInMcastPkts].
1212 The number of multicast packets transmitted per second
1213 by the interface [ipv6IfStatsOutMcastPkts].
1218 The number of IPv6 datagrams that have been
1219 successfully fragmented at this output interface per second
1220 [ipv6IfStatsOutFragOKs].
1225 The number of output datagram fragments that have
1226 been generated per second as a result of fragmentation at
1227 this output interface [ipv6IfStatsOutFragCreates].
1232 keyword, statistics about IPv6 network errors are reported.
1233 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1235 The following values are displayed (formal SNMP names between
1240 The number of input datagrams discarded per second due to
1241 errors in their IPv6 headers, including version
1242 number mismatch, other format errors, hop count
1243 exceeded, errors discovered in processing their
1244 IPv6 options, etc. [ipv6IfStatsInHdrErrors]
1249 The number of input datagrams discarded per second because
1250 the IPv6 address in their IPv6 header's destination
1251 field was not a valid address to be received at
1252 this entity. This count includes invalid
1253 addresses (e.g., ::0) and unsupported addresses
1254 (e.g., addresses with unallocated prefixes). For
1255 entities which are not IPv6 routers and therefore
1256 do not forward datagrams, this counter includes
1257 datagrams discarded because the destination address
1258 was not a local address [ipv6IfStatsInAddrErrors].
1263 The number of locally-addressed datagrams
1264 received successfully but discarded per second because of an
1265 unknown or unsupported protocol [ipv6IfStatsInUnknownProtos].
1270 The number of input datagrams that could not be
1271 forwarded per second because their size exceeded the link MTU
1272 of outgoing interface [ipv6IfStatsInTooBigErrors].
1277 The number of input IPv6 datagrams per second for which no
1278 problems were encountered to prevent their
1279 continued processing, but which were discarded
1280 (e.g., for lack of buffer space)
1281 [ipv6IfStatsInDiscards]. Note that this
1282 counter does not include any datagrams discarded
1283 while awaiting re-assembly.
1288 The number of output IPv6 datagrams per second for which no
1289 problem was encountered to prevent their
1290 transmission to their destination, but which were
1291 discarded (e.g., for lack of buffer space)
1292 [ipv6IfStatsOutDiscards]. Note
1293 that this counter would include datagrams counted
1294 in fwddgm6/s if any such packets
1295 met this (discretionary) discard criterion.
1300 The number of input datagrams discarded per second because no
1301 route could be found to transmit them to their
1302 destination [ipv6IfStatsInNoRoutes].
1307 The number of locally generated IP datagrams discarded per second
1308 because no route could be found to transmit them to their
1309 destination [unknown formal SNMP name].
1314 The number of failures detected per second by the IPv6
1315 re-assembly algorithm (for whatever reason: timed
1316 out, errors, etc.) [ipv6IfStatsReasmFails].
1317 Note that this is not
1318 necessarily a count of discarded IPv6 fragments
1319 since some algorithms
1320 can lose track of the number of fragments
1321 by combining them as they are received.
1326 The number of IPv6 datagrams that have been
1327 discarded per second because they needed to be fragmented
1328 at this output interface but could not be
1329 [ipv6IfStatsOutFragFails].
1334 The number of input datagrams discarded per second because
1335 datagram frame didn't carry enough data
1336 [ipv6IfStatsInTruncatedPkts].
1341 keyword, statistics about ICMPv6 network traffic are reported.
1342 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1344 The following values are displayed (formal SNMP names between
1349 The total number of ICMP messages received
1350 by the interface per second which includes all those
1351 counted by ierr6/s [ipv6IfIcmpInMsgs].
1356 The total number of ICMP messages which this
1357 interface attempted to send per second [ipv6IfIcmpOutMsgs].
1362 The number of ICMP Echo (request) messages
1363 received by the interface per second [ipv6IfIcmpInEchos].
1368 The number of ICMP Echo Reply messages received
1369 by the interface per second [ipv6IfIcmpInEchoReplies].
1374 The number of ICMP Echo Reply messages sent
1375 by the interface per second [ipv6IfIcmpOutEchoReplies].
1380 The number of ICMPv6 Group Membership Query
1381 messages received by the interface per second
1382 [ipv6IfIcmpInGroupMembQueries].
1387 The number of ICMPv6 Group Membership Response messages
1388 received by the interface per second
1389 [ipv6IfIcmpInGroupMembResponses].
1394 The number of ICMPv6 Group Membership Response
1395 messages sent per second
1396 [ipv6IfIcmpOutGroupMembResponses].
1401 The number of ICMPv6 Group Membership Reduction messages
1402 received by the interface per second
1403 [ipv6IfIcmpInGroupMembReductions].
1408 The number of ICMPv6 Group Membership Reduction
1409 messages sent per second
1410 [ipv6IfIcmpOutGroupMembReductions].
1415 The number of ICMP Router Solicit messages
1416 received by the interface per second
1417 [ipv6IfIcmpInRouterSolicits].
1422 The number of ICMP Router Solicitation messages
1423 sent by the interface per second
1424 [ipv6IfIcmpOutRouterSolicits].
1429 The number of ICMP Router Advertisement messages
1430 received by the interface per second
1431 [ipv6IfIcmpInRouterAdvertisements].
1436 The number of ICMP Neighbor Solicit messages
1437 received by the interface per second
1438 [ipv6IfIcmpInNeighborSolicits].
1443 The number of ICMP Neighbor Solicitation
1444 messages sent by the interface per second
1445 [ipv6IfIcmpOutNeighborSolicits].
1450 The number of ICMP Neighbor Advertisement
1451 messages received by the interface per second
1452 [ipv6IfIcmpInNeighborAdvertisements].
1457 The number of ICMP Neighbor Advertisement
1458 messages sent by the interface per second
1459 [ipv6IfIcmpOutNeighborAdvertisements].
1464 keyword, statistics about ICMPv6 error messages are reported.
1465 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1467 The following values are displayed (formal SNMP names between
1472 The number of ICMP messages per second which the interface
1473 received but determined as having ICMP-specific
1474 errors (bad ICMP checksums, bad length, etc.)
1475 [ipv6IfIcmpInErrors]
1480 The number of ICMP Destination Unreachable
1481 messages received by the interface per second
1482 [ipv6IfIcmpInDestUnreachs].
1487 The number of ICMP Destination Unreachable
1488 messages sent by the interface per second
1489 [ipv6IfIcmpOutDestUnreachs].
1494 The number of ICMP Time Exceeded messages
1495 received by the interface per second
1496 [ipv6IfIcmpInTimeExcds].
1501 The number of ICMP Time Exceeded messages sent
1502 by the interface per second
1503 [ipv6IfIcmpOutTimeExcds].
1508 The number of ICMP Parameter Problem messages
1509 received by the interface per second
1510 [ipv6IfIcmpInParmProblems].
1515 The number of ICMP Parameter Problem messages
1516 sent by the interface per second
1517 [ipv6IfIcmpOutParmProblems].
1522 The number of Redirect messages received
1523 by the interface per second
1524 [ipv6IfIcmpInRedirects].
1529 The number of Redirect messages sent by
1530 the interface by second
1531 [ipv6IfIcmpOutRedirects].
1536 The number of ICMP Packet Too Big messages
1537 received by the interface per second
1538 [ipv6IfIcmpInPktTooBigs].
1543 The number of ICMP Packet Too Big messages sent
1544 by the interface per second
1545 [ipv6IfIcmpOutPktTooBigs].
1550 keyword, statistics about UDPv6 network traffic are reported.
1551 Note that UDPv6 statistics depend on sadc option "-S IPV6" to be
1553 The following values are displayed (formal SNMP names between
1558 The total number of UDP datagrams delivered per second to UDP users
1564 The total number of UDP datagrams sent per second from this
1565 entity [udpOutDatagrams].
1570 The total number of received UDP datagrams per second for which there
1571 was no application at the destination port [udpNoPorts].
1576 The number of received UDP datagrams per second that could not be
1577 delivered for reasons other than the lack of an application
1578 at the destination port [udpInErrors].
1583 keyword is equivalent to specifying all the keywords above and therefore all the network
1584 activities are reported.
1587 .IP "-o [ filename ]"
1588 Save the readings in the file in binary form. Each reading
1589 is in a separate record. The default value of the
1591 parameter is the current daily data file, the
1593 file. The -o option is exclusive of the -f option.
1594 All the data available from the kernel are saved in the file (in fact,
1596 calls its data collector
1598 with the option "-S ALL". See sadc(8) manual page).
1599 .IP "-P { cpu [,...] | ALL }"
1600 Report per-processor statistics for the specified processor or processors.
1603 keyword reports statistics for each individual processor, and globally for
1605 Note that processor 0 is the first processor.
1607 Pretty-print device names. Use this option in conjunction with option -d.
1608 By default names are printed as
1610 where m and n are the major and minor numbers for the device.
1611 Use of this option displays the names of the devices as they (should) appear
1612 in /dev. Name mappings are controlled by
1613 .B @SYSCONFIG_DIR@/sysstat.ioconf.
1615 Report queue length and load averages. The following values are displayed:
1620 Run queue length (number of tasks waiting for run time).
1625 Number of tasks in the task list.
1630 System load average for the last minute.
1631 The load average is calculated as the average number of runnable or
1632 running tasks (R state), and the number of tasks in uninterruptible
1633 sleep (D state) over the specified interval.
1638 System load average for the past 5 minutes.
1643 System load average for the past 15 minutes.
1647 Report memory utilization statistics.
1648 The following values are displayed:
1653 Amount of free memory available in kilobytes.
1658 Amount of used memory in kilobytes. This does not take into account memory
1659 used by the kernel itself.
1664 Percentage of used memory.
1669 Amount of memory used as buffers by the kernel in kilobytes.
1674 Amount of memory used to cache data by the kernel in kilobytes.
1679 Amount of memory in kilobytes needed for current workload. This is an estimate of how much
1680 RAM/swap is needed to guarantee that there never is out of memory.
1685 Percentage of memory needed for current workload in relation to the total amount of memory (RAM+swap).
1686 This number may be greater than 100% because the kernel usually overcommits memory.
1691 Report memory statistics. The following values are displayed:
1696 Number of memory pages freed by the system per second.
1697 A negative value represents a number of pages allocated by the system.
1698 Note that a page has a size of 4 kB or 8 kB according to the machine architecture.
1703 Number of additional memory pages used as buffers by the system per second.
1704 A negative value means fewer pages used as buffers by the system.
1709 Number of additional memory pages cached by the system per second.
1710 A negative value means fewer pages in the cache.
1713 .IP "-s [ hh:mm:ss ]"
1714 Set the starting time of the data, causing the
1716 command to extract records time-tagged at, or following, the time
1717 specified. The default starting time is 08:00:00.
1718 Hours must be given in 24-hour format. This option can be
1719 used only when data are read from a file (option
1723 Report swap space utilization statistics.
1724 The following values are displayed:
1729 Amount of free swap space in kilobytes.
1734 Amount of used swap space in kilobytes.
1739 Percentage of used swap space.
1744 Amount of cached swap memory in kilobytes.
1745 This is memory that once was swapped out, is swapped back in
1746 but still also is in the swap area (if memory is needed it doesn't need
1747 to be swapped out again because it is already in the swap area. This
1753 Percentage of cached swap memory in relation to the amount of used swap space.
1757 When reading data from a daily data file, indicate that
1759 should display the timestamps in the original locale time of
1760 the data file creator. Without this option, the
1762 command displays the timestamps in the user's locale time.
1764 Report CPU utilization. The
1766 keyword indicates that all the CPU fields should be displayed.
1767 The report may show the following fields:
1772 Percentage of CPU utilization that occurred while executing at the user
1773 level (application). Note that this field includes time spent running
1779 Percentage of CPU utilization that occurred while executing at the user
1780 level (application). Note that this field does NOT include time spent
1781 running virtual processors.
1786 Percentage of CPU utilization that occurred while executing at the user
1787 level with nice priority.
1792 Percentage of CPU utilization that occurred while executing at the system
1793 level (kernel). Note that this field includes time spent servicing
1794 hardware and software interrupts.
1799 Percentage of CPU utilization that occurred while executing at the system
1800 level (kernel). Note that this field does NOT include time spent servicing
1801 hardware or software interrupts.
1806 Percentage of time that the CPU or CPUs were idle during which
1807 the system had an outstanding disk I/O request.
1812 Percentage of time spent in involuntary wait by the virtual CPU
1813 or CPUs while the hypervisor was servicing another virtual processor.
1818 Percentage of time spent by the CPU or CPUs to service hardware interrupts.
1823 Percentage of time spent by the CPU or CPUs to service software interrupts.
1828 Percentage of time spent by the CPU or CPUs to run a virtual processor.
1833 Percentage of time that the CPU or CPUs were idle and the system
1834 did not have an outstanding disk I/O request.
1837 Note: On SMP machines a processor that does not have any activity at all
1838 (0.00 for every field) is a disabled (offline) processor.
1841 Report status of inode, file and other kernel tables.
1842 The following values are displayed:
1847 Number of unused cache entries in the directory cache.
1852 Number of file handles used by the system.
1857 Number of inode handlers used by the system.
1862 Number of pseudo-terminals used by the system.
1866 Print version number then exit.
1868 Report task creation and system switching activity.
1873 Total number of tasks created per second.
1878 Total number of context switches per second.
1882 Report swapping statistics. The following values are displayed:
1887 Total number of swap pages the system brought in per second.
1892 Total number of swap pages the system brought out per second.
1896 Report TTY device activity. The following values are displayed:
1901 Number of receive interrupts per second for current serial line. Serial line number
1902 is given in the TTY column.
1907 Number of transmit interrupts per second for current serial line.
1912 Number of frame errors per second for current serial line.
1917 Number of parity errors per second for current serial line.
1922 Number of breaks per second for current serial line.
1927 Number of overrun errors per second for current serial line.
1934 command takes into account the following environment variables:
1937 If this variable exists and its value is
1939 then the current locale will be ignored when printing the date in the report header.
1942 command will use the ISO 8601 format (YYYY-MM-DD) instead.
1945 If this variable exists and its value is
1949 will save its data in UTC time (data will still be displayed in local time).
1951 will also use UTC time instead of local time to determine the current daily
1952 data file located in the
1954 directory. This variable may be useful for servers with users located across
1959 Report CPU utilization for each 2 seconds. 5 lines are displayed.
1962 .B sar -I 14 -o int14.file 2 10
1964 Report statistics on IRQ 14 for each 2 seconds. 10 lines are displayed.
1965 Data are stored in a file called
1969 .B sar -r -n DEV -f @SA_DIR@/sa16
1971 Display memory and network statistics saved in daily data file 'sa16'.
1976 Display all the statistics saved in current daily data file.
1979 filesystem must be mounted for the
1983 All the statistics are not necessarily available, depending on the kernel version used.
1985 assumes that you are using at least a 2.6 kernel.
1989 Indicate the daily data file, where the
1991 parameter is a number representing the day of the month.
1995 contains various files with system statistics.
1997 Sebastien Godard (sysstat <at> orange.fr)
2008 .I http://pagesperso-orange.fr/sebastien.godard/