1 .TH SAR 1 "JUNE 2012" 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 [
23 .B ] | -[0-9]+ ] [ -s [
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.
72 It is also possible to enter -1, -2 etc. as an argument to sar to
74 of that days ago. For example, -1 will point at the standard system
75 activity file of yesterday.
81 command reports system-wide (global among all processors) statistics,
82 which are calculated as averages for values expressed as percentages,
83 and as sums otherwise. If the
87 command reports activity which relates to the specified processor or
92 command reports statistics for each individual processor and global
93 statistics among all processors.
95 You can select information about specific system activities using
96 flags. Not specifying any flags selects only CPU activity.
99 flag selects all possible activities.
101 The default version of the
103 command (CPU utilization report) might be one of the first facilities
104 the user runs to begin system activity investigation, because it
105 monitors major system resources. If CPU utilization is near 100 percent
106 (user + nice + system), the workload sampled is CPU-bound.
108 If multiple samples and multiple reports are desired, it is convenient
109 to specify an output file for the
114 command as a background process. The syntax for this is:
116 .B sar -o datafile interval count >/dev/null 2>&1 &
118 All data are captured in binary form and saved to a file (datafile).
119 The data can then be selectively displayed with the
131 second intervals. If the
133 parameter is not set, all the records saved in the
134 file will be selected.
135 Collection of data in this manner is useful to characterize
136 system usage over a period of time and determine peak usage hours.
140 command only reports on local activities.
144 This is equivalent to specifying
145 .BR "-bBdHqrRSuvwWy -I SUM -I XALL -m ALL -n ALL -u ALL -P ALL".
147 Report I/O and transfer rate statistics.
148 The following values are displayed:
153 Total number of transfers per second that were issued to physical devices.
154 A transfer is an I/O request to a physical device. Multiple logical
155 requests can be combined into a single I/O request to the device.
156 A transfer is of indeterminate size.
161 Total number of read requests per second issued to physical devices.
166 Total number of write requests per second issued to physical devices.
171 Total amount of data read from the devices in blocks per second.
172 Blocks are equivalent to sectors
173 and therefore have a size of 512 bytes.
178 Total amount of data written to devices in blocks per second.
182 Report paging statistics.
183 The following values are displayed:
188 Total number of kilobytes the system paged in from disk per second.
193 Total number of kilobytes the system paged out to disk per second.
198 Number of page faults (major + minor) made by the system per second.
199 This is not a count of page faults that generate I/O, because some page
200 faults can be resolved without I/O.
205 Number of major faults the system has made per second, those which
206 have required loading a memory page from disk.
211 Number of pages placed on the free list by the system per second.
216 Number of pages scanned by the kswapd daemon per second.
221 Number of pages scanned directly per second.
226 Number of pages the system has reclaimed from cache (pagecache and
227 swapcache) per second to satisfy its memory demands.
232 Calculated as pgsteal / pgscan, this is a metric of the efficiency of
233 page reclaim. If it is near 100% then almost every page coming off the
234 tail of the inactive list is being reaped. If it gets too low (e.g. less
235 than 30%) then the virtual memory is having some difficulty.
236 This field is displayed as zero if no pages have been scanned during the
241 When reading data from a file, tell
243 to display comments that have been inserted by
246 Report activity for each block device.
247 When data are displayed, the device specification
253 is the major number of the device and
256 Device names may also be pretty-printed if option -p
258 Note that disk activity depends on sadc options "-S DISK" and "-S XDISK"
259 to be collected. The following values are displayed:
264 Indicate the number of transfers per second that were issued to the device.
265 Multiple logical requests can be combined into a single I/O request to the
266 device. A transfer is of indeterminate size.
271 Number of sectors read from the device. The size of a sector is 512 bytes.
276 Number of sectors written to the device. The size of a sector is 512 bytes.
281 The average size (in sectors) of the requests that were issued to the device.
286 The average queue length of the requests that were issued to the device.
291 The average time (in milliseconds) for I/O requests issued to the device
292 to be served. This includes the time spent by the requests in queue and
293 the time spent servicing them.
298 The average service time (in milliseconds) for I/O requests that were issued
299 to the device. Warning! Do not trust this field any more. This field will be
300 removed in a future sysstat version.
305 Percentage of CPU time during which I/O requests were issued to the device
306 (bandwidth utilization for the device). Device saturation occurs when this
307 value is close to 100%.
310 .IP "-e [ hh:mm:ss ]"
311 Set the ending time of the report. The default ending time is
312 18:00:00. Hours must be given in 24-hour format.
313 This option can be used when data are read from
314 or written to a file (options
319 .IP "-f [ filename ]"
324 flag). The default value of the
326 parameter is the current daily data file, the
328 file. The -f option is exclusive of the -o option.
330 Display a short help message then exit.
332 Report hugepages utilization statistics.
333 The following values are displayed:
338 Amount of hugepages memory in kilobytes that is not yet allocated.
343 Amount of hugepages memory in kilobytes that has been allocated.
348 Percentage of total hugepages memory that has been allocated.
352 Select data records at seconds as close as possible to the number specified
356 .IP "-I { int [,...] | SUM | ALL | XALL }"
357 Report statistics for a given interrupt.
359 is the interrupt number. Specifying multiple
361 parameters on the command line will look at multiple independent interrupts.
364 keyword indicates that the total number of interrupts received per second
365 is to be displayed. The
367 keyword indicates that statistics from
368 the first 16 interrupts are to be reported, whereas the
370 keyword indicates that statistics from all interrupts, including potential
371 APIC interrupt sources, are to be reported.
372 Note that interrupt statistics depend on sadc option "-S INT" to be collected.
373 .IP "-m { keyword [,...] | ALL }"
374 Report power management statistics.
375 Note that these statistics depend on sadc option "-S POWER" to be collected.
377 Possible keywords are
388 keyword, statistics about CPU are reported.
389 The following value is displayed:
394 Instantaneous CPU clock frequency in MHz.
399 keyword, statistics about fans speed are reported.
400 The following values are displayed:
404 Fan speed expressed in revolutions per minute.
409 This field is calculated as the difference between current fan speed (rpm)
410 and its low limit (fan_min).
420 keyword, statistics about CPU clock frequency are reported.
421 The following value is displayed:
425 Weighted average CPU clock frequency in MHz.
426 Note that the cpufreq-stats driver must be compiled in the
427 kernel for this option to work.
432 keyword, statistics about voltage inputs are reported.
433 The following values are displayed:
437 Voltage input expressed in Volts.
442 Relative input value. A value of 100% means that
443 voltage input has reached its high limit (in_max) whereas
444 a value of 0% means that it has reached its low limit (in_min).
454 keyword, statistics about devices temperature are reported.
455 The following values are displayed:
459 Device temperature expressed in degrees Celsius.
464 Relative device temperature. A value of 100% means that
465 temperature has reached its high limit (temp_max).
477 command takes a snapshot of all the USB devices currently plugged into
478 the system. At the end of the report,
480 will display a summary of all those USB devices.
481 The following values are displayed:
485 Root hub number of the USB device.
490 Vendor ID number (assigned by USB organization).
495 Product ID number (assigned by Manufacturer).
500 Maximum power consumption of the device (expressed in mA).
513 .IP "-n { keyword [,...] | ALL }"
514 Report network statistics.
516 Possible keywords are
539 keyword, statistics from the network devices are reported.
540 The following values are displayed:
545 Name of the network interface for which statistics are reported.
550 Total number of packets received per second.
555 Total number of packets transmitted per second.
560 Total number of kilobytes received per second.
565 Total number of kilobytes transmitted per second.
570 Number of compressed packets received per second (for cslip etc.).
575 Number of compressed packets transmitted per second.
580 Number of multicast packets received per second.
585 keyword, statistics on failures (errors) from the network devices are reported.
586 The following values are displayed:
590 Name of the network interface for which statistics are reported.
595 Total number of bad packets received per second.
600 Total number of errors that happened per second while transmitting packets.
605 Number of collisions that happened per second while transmitting packets.
610 Number of received packets dropped per second because of a lack of space in linux buffers.
615 Number of transmitted packets dropped per second because of a lack of space in linux buffers.
620 Number of carrier-errors that happened per second while transmitting packets.
625 Number of frame alignment errors that happened per second on received packets.
630 Number of FIFO overrun errors that happened per second on received packets.
635 Number of FIFO overrun errors that happened per second on transmitted packets.
640 keyword, statistics about NFS client activity are reported.
641 The following values are displayed:
645 Number of RPC requests made per second.
650 Number of RPC requests per second, those which needed to be retransmitted (for
651 example because of a server timeout).
656 Number of 'read' RPC calls made per second.
661 Number of 'write' RPC calls made per second.
666 Number of 'access' RPC calls made per second.
671 Number of 'getattr' RPC calls made per second.
676 keyword, statistics about NFS server activity are reported.
677 The following values are displayed:
681 Number of RPC requests received per second.
686 Number of bad RPC requests received per second, those whose
687 processing generated an error.
692 Number of network packets received per second.
697 Number of UDP packets received per second.
702 Number of TCP packets received per second.
707 Number of reply cache hits per second.
712 Number of reply cache misses per second.
717 Number of 'read' RPC calls received per second.
722 Number of 'write' RPC calls received per second.
727 Number of 'access' RPC calls received per second.
732 Number of 'getattr' RPC calls received per second.
737 keyword, statistics on sockets in use are reported
739 The following values are displayed:
743 Total number of sockets used by the system.
748 Number of TCP sockets currently in use.
753 Number of UDP sockets currently in use.
758 Number of RAW sockets currently in use.
763 Number of IP fragments currently in queue.
768 Number of TCP sockets in TIME_WAIT state.
773 keyword, statistics about IPv4 network traffic are reported.
774 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
776 The following values are displayed (formal SNMP names between
781 The total number of input datagrams received from interfaces
782 per second, including those received in error [ipInReceives].
787 The number of input datagrams per second, for which this entity was not
788 their final IP destination, as a result of which an attempt
789 was made to find a route to forward them to that final
790 destination [ipForwDatagrams].
795 The total number of input datagrams successfully delivered per second
796 to IP user-protocols (including ICMP) [ipInDelivers].
801 The total number of IP datagrams which local IP user-protocols (including ICMP)
802 supplied per second to IP in requests for transmission [ipOutRequests].
803 Note that this counter does not include any datagrams counted in fwddgm/s.
808 The number of IP fragments received per second which needed to be
809 reassembled at this entity [ipReasmReqds].
814 The number of IP datagrams successfully re-assembled per second [ipReasmOKs].
819 The number of IP datagrams that have been successfully
820 fragmented at this entity per second [ipFragOKs].
825 The number of IP datagram fragments that have been
826 generated per second as a result of fragmentation at this entity [ipFragCreates].
831 keyword, statistics about IPv4 network errors are reported.
832 Note that IPv4 statistics depend on sadc option "-S SNMP" to be
834 The following values are displayed (formal SNMP names between
839 The number of input datagrams discarded per second due to errors in
840 their IP headers, including bad checksums, version number
841 mismatch, other format errors, time-to-live exceeded, errors
842 discovered in processing their IP options, etc. [ipInHdrErrors]
847 The number of input datagrams discarded per second because the IP
848 address in their IP header's destination field was not a
849 valid address to be received at this entity. This count
850 includes invalid addresses (e.g., 0.0.0.0) and addresses of
851 unsupported Classes (e.g., Class E). For entities which are
852 not IP routers and therefore do not forward datagrams, this
853 counter includes datagrams discarded because the destination
854 address was not a local address [ipInAddrErrors].
859 The number of locally-addressed datagrams received
860 successfully but discarded per second because of an unknown or
861 unsupported protocol [ipInUnknownProtos].
866 The number of input IP datagrams per second for which no problems were
867 encountered to prevent their continued processing, but which
868 were discarded (e.g., for lack of buffer space) [ipInDiscards].
869 Note that this counter does not include any datagrams discarded while
870 awaiting re-assembly.
875 The number of output IP datagrams per second for which no problem was
876 encountered to prevent their transmission to their
877 destination, but which were discarded (e.g., for lack of
878 buffer space) [ipOutDiscards].
879 Note that this counter would include
880 datagrams counted in fwddgm/s if any such packets met
881 this (discretionary) discard criterion.
886 The number of IP datagrams discarded per second because no route could
887 be found to transmit them to their destination [ipOutNoRoutes].
888 Note that this counter includes any packets counted in fwddgm/s
889 which meet this 'no-route' criterion.
890 Note that this includes any datagrams which a host cannot route because all
891 of its default routers are down.
896 The number of failures detected per second by the IP re-assembly
897 algorithm (for whatever reason: timed out, errors, etc) [ipReasmFails].
898 Note that this is not necessarily a count of discarded IP
899 fragments since some algorithms can lose track of the number of
900 fragments by combining them as they are received.
905 The number of IP datagrams that have been discarded per second because
906 they needed to be fragmented at this entity but could not
907 be, e.g., because their Don't Fragment flag was set [ipFragFails].
912 keyword, statistics about ICMPv4 network traffic are reported.
913 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
915 The following values are displayed (formal SNMP names between
920 The total number of ICMP messages which the entity
921 received per second [icmpInMsgs].
922 Note that this counter includes all those counted by ierr/s.
927 The total number of ICMP messages which this entity
928 attempted to send per second [icmpOutMsgs].
929 Note that this counter includes all those counted by oerr/s.
934 The number of ICMP Echo (request) messages received per second [icmpInEchos].
939 The number of ICMP Echo Reply messages received per second [icmpInEchoReps].
944 The number of ICMP Echo (request) messages sent per second [icmpOutEchos].
949 The number of ICMP Echo Reply messages sent per second [icmpOutEchoReps].
954 The number of ICMP Timestamp (request) messages received per second [icmpInTimestamps].
959 The number of ICMP Timestamp Reply messages received per second [icmpInTimestampReps].
964 The number of ICMP Timestamp (request) messages sent per second [icmpOutTimestamps].
969 The number of ICMP Timestamp Reply messages sent per second [icmpOutTimestampReps].
974 The number of ICMP Address Mask Request messages received per second [icmpInAddrMasks].
979 The number of ICMP Address Mask Reply messages received per second [icmpInAddrMaskReps].
984 The number of ICMP Address Mask Request messages sent per second [icmpOutAddrMasks].
989 The number of ICMP Address Mask Reply messages sent per second [icmpOutAddrMaskReps].
994 keyword, statistics about ICMPv4 error messages are reported.
995 Note that ICMPv4 statistics depend on sadc option "-S SNMP" to be
997 The following values are displayed (formal SNMP names between
1002 The number of ICMP messages per second which the entity received but
1003 determined as having ICMP-specific errors (bad ICMP
1004 checksums, bad length, etc.) [icmpInErrors].
1009 The number of ICMP messages per second which this entity did not send
1010 due to problems discovered within ICMP such as a lack of buffers [icmpOutErrors].
1015 The number of ICMP Destination Unreachable messages
1016 received per second [icmpInDestUnreachs].
1021 The number of ICMP Destination Unreachable messages sent per second [icmpOutDestUnreachs].
1026 The number of ICMP Time Exceeded messages received per second [icmpInTimeExcds].
1031 The number of ICMP Time Exceeded messages sent per second [icmpOutTimeExcds].
1036 The number of ICMP Parameter Problem messages received per second [icmpInParmProbs].
1041 The number of ICMP Parameter Problem messages sent per second [icmpOutParmProbs].
1046 The number of ICMP Source Quench messages received per second [icmpInSrcQuenchs].
1051 The number of ICMP Source Quench messages sent per second [icmpOutSrcQuenchs].
1056 The number of ICMP Redirect messages received per second [icmpInRedirects].
1061 The number of ICMP Redirect messages sent per second [icmpOutRedirects].
1066 keyword, statistics about TCPv4 network traffic are reported.
1067 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1069 The following values are displayed (formal SNMP names between
1074 The number of times TCP connections have made a direct
1075 transition to the SYN-SENT state from the CLOSED state per second [tcpActiveOpens].
1080 The number of times TCP connections have made a direct
1081 transition to the SYN-RCVD state from the LISTEN state per second [tcpPassiveOpens].
1086 The total number of segments received per second, including those
1087 received in error [tcpInSegs]. This count includes segments received on
1088 currently established connections.
1093 The total number of segments sent per second, including those on
1094 current connections but excluding those containing only
1095 retransmitted octets [tcpOutSegs].
1100 keyword, statistics about TCPv4 network errors are reported.
1101 Note that TCPv4 statistics depend on sadc option "-S SNMP" to be
1103 The following values are displayed (formal SNMP names between
1108 The number of times per second TCP connections have made a direct
1109 transition to the CLOSED state from either the SYN-SENT
1110 state or the SYN-RCVD state, plus the number of times per second TCP
1111 connections have made a direct transition to the LISTEN
1112 state from the SYN-RCVD state [tcpAttemptFails].
1117 The number of times per second TCP connections have made a direct
1118 transition to the CLOSED state from either the ESTABLISHED
1119 state or the CLOSE-WAIT state [tcpEstabResets].
1124 The total number of segments retransmitted per second - that is, the
1125 number of TCP segments transmitted containing one or more
1126 previously transmitted octets [tcpRetransSegs].
1131 The total number of segments received in error (e.g., bad
1132 TCP checksums) per second [tcpInErrs].
1137 The number of TCP segments sent per second containing the RST flag [tcpOutRsts].
1142 keyword, statistics about UDPv4 network traffic are reported.
1143 Note that UDPv4 statistics depend on sadc option "-S SNMP" to be
1145 The following values are displayed (formal SNMP names between
1150 The total number of UDP datagrams delivered per second to UDP users [udpInDatagrams].
1155 The total number of UDP datagrams sent per second from this entity [udpOutDatagrams].
1160 The total number of received UDP datagrams per second for which there
1161 was no application at the destination port [udpNoPorts].
1166 The number of received UDP datagrams per second that could not be
1167 delivered for reasons other than the lack of an application
1168 at the destination port [udpInErrors].
1173 keyword, statistics on sockets in use are reported (IPv6).
1174 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1176 The following values are displayed:
1180 Number of TCPv6 sockets currently in use.
1185 Number of UDPv6 sockets currently in use.
1190 Number of RAWv6 sockets currently in use.
1195 Number of IPv6 fragments currently in use.
1200 keyword, statistics about IPv6 network traffic are reported.
1201 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1203 The following values are displayed (formal SNMP names between
1208 The total number of input datagrams received from
1209 interfaces per second, including those received in error
1210 [ipv6IfStatsInReceives].
1215 The number of output datagrams per second which this
1216 entity received and forwarded to their final
1217 destinations [ipv6IfStatsOutForwDatagrams].
1222 The total number of datagrams successfully
1223 delivered per second to IPv6 user-protocols (including ICMP)
1224 [ipv6IfStatsInDelivers].
1229 The total number of IPv6 datagrams which local IPv6
1230 user-protocols (including ICMP) supplied per second to IPv6 in
1231 requests for transmission [ipv6IfStatsOutRequests].
1232 Note that this counter
1233 does not include any datagrams counted in fwddgm6/s.
1238 The number of IPv6 fragments received per second which needed
1239 to be reassembled at this interface [ipv6IfStatsReasmReqds].
1244 The number of IPv6 datagrams successfully
1245 reassembled per second [ipv6IfStatsReasmOKs].
1250 The number of multicast packets received per second
1251 by the interface [ipv6IfStatsInMcastPkts].
1256 The number of multicast packets transmitted per second
1257 by the interface [ipv6IfStatsOutMcastPkts].
1262 The number of IPv6 datagrams that have been
1263 successfully fragmented at this output interface per second
1264 [ipv6IfStatsOutFragOKs].
1269 The number of output datagram fragments that have
1270 been generated per second as a result of fragmentation at
1271 this output interface [ipv6IfStatsOutFragCreates].
1276 keyword, statistics about IPv6 network errors are reported.
1277 Note that IPv6 statistics depend on sadc option "-S IPV6" to be
1279 The following values are displayed (formal SNMP names between
1284 The number of input datagrams discarded per second due to
1285 errors in their IPv6 headers, including version
1286 number mismatch, other format errors, hop count
1287 exceeded, errors discovered in processing their
1288 IPv6 options, etc. [ipv6IfStatsInHdrErrors]
1293 The number of input datagrams discarded per second because
1294 the IPv6 address in their IPv6 header's destination
1295 field was not a valid address to be received at
1296 this entity. This count includes invalid
1297 addresses (e.g., ::0) and unsupported addresses
1298 (e.g., addresses with unallocated prefixes). For
1299 entities which are not IPv6 routers and therefore
1300 do not forward datagrams, this counter includes
1301 datagrams discarded because the destination address
1302 was not a local address [ipv6IfStatsInAddrErrors].
1307 The number of locally-addressed datagrams
1308 received successfully but discarded per second because of an
1309 unknown or unsupported protocol [ipv6IfStatsInUnknownProtos].
1314 The number of input datagrams that could not be
1315 forwarded per second because their size exceeded the link MTU
1316 of outgoing interface [ipv6IfStatsInTooBigErrors].
1321 The number of input IPv6 datagrams per second for which no
1322 problems were encountered to prevent their
1323 continued processing, but which were discarded
1324 (e.g., for lack of buffer space)
1325 [ipv6IfStatsInDiscards]. Note that this
1326 counter does not include any datagrams discarded
1327 while awaiting re-assembly.
1332 The number of output IPv6 datagrams per second for which no
1333 problem was encountered to prevent their
1334 transmission to their destination, but which were
1335 discarded (e.g., for lack of buffer space)
1336 [ipv6IfStatsOutDiscards]. Note
1337 that this counter would include datagrams counted
1338 in fwddgm6/s if any such packets
1339 met this (discretionary) discard criterion.
1344 The number of input datagrams discarded per second because no
1345 route could be found to transmit them to their
1346 destination [ipv6IfStatsInNoRoutes].
1351 The number of locally generated IP datagrams discarded per second
1352 because no route could be found to transmit them to their
1353 destination [unknown formal SNMP name].
1358 The number of failures detected per second by the IPv6
1359 re-assembly algorithm (for whatever reason: timed
1360 out, errors, etc.) [ipv6IfStatsReasmFails].
1361 Note that this is not
1362 necessarily a count of discarded IPv6 fragments
1363 since some algorithms
1364 can lose track of the number of fragments
1365 by combining them as they are received.
1370 The number of IPv6 datagrams that have been
1371 discarded per second because they needed to be fragmented
1372 at this output interface but could not be
1373 [ipv6IfStatsOutFragFails].
1378 The number of input datagrams discarded per second because
1379 datagram frame didn't carry enough data
1380 [ipv6IfStatsInTruncatedPkts].
1385 keyword, statistics about ICMPv6 network traffic are reported.
1386 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1388 The following values are displayed (formal SNMP names between
1393 The total number of ICMP messages received
1394 by the interface per second which includes all those
1395 counted by ierr6/s [ipv6IfIcmpInMsgs].
1400 The total number of ICMP messages which this
1401 interface attempted to send per second [ipv6IfIcmpOutMsgs].
1406 The number of ICMP Echo (request) messages
1407 received by the interface per second [ipv6IfIcmpInEchos].
1412 The number of ICMP Echo Reply messages received
1413 by the interface per second [ipv6IfIcmpInEchoReplies].
1418 The number of ICMP Echo Reply messages sent
1419 by the interface per second [ipv6IfIcmpOutEchoReplies].
1424 The number of ICMPv6 Group Membership Query
1425 messages received by the interface per second
1426 [ipv6IfIcmpInGroupMembQueries].
1431 The number of ICMPv6 Group Membership Response messages
1432 received by the interface per second
1433 [ipv6IfIcmpInGroupMembResponses].
1438 The number of ICMPv6 Group Membership Response
1439 messages sent per second
1440 [ipv6IfIcmpOutGroupMembResponses].
1445 The number of ICMPv6 Group Membership Reduction messages
1446 received by the interface per second
1447 [ipv6IfIcmpInGroupMembReductions].
1452 The number of ICMPv6 Group Membership Reduction
1453 messages sent per second
1454 [ipv6IfIcmpOutGroupMembReductions].
1459 The number of ICMP Router Solicit messages
1460 received by the interface per second
1461 [ipv6IfIcmpInRouterSolicits].
1466 The number of ICMP Router Solicitation messages
1467 sent by the interface per second
1468 [ipv6IfIcmpOutRouterSolicits].
1473 The number of ICMP Router Advertisement messages
1474 received by the interface per second
1475 [ipv6IfIcmpInRouterAdvertisements].
1480 The number of ICMP Neighbor Solicit messages
1481 received by the interface per second
1482 [ipv6IfIcmpInNeighborSolicits].
1487 The number of ICMP Neighbor Solicitation
1488 messages sent by the interface per second
1489 [ipv6IfIcmpOutNeighborSolicits].
1494 The number of ICMP Neighbor Advertisement
1495 messages received by the interface per second
1496 [ipv6IfIcmpInNeighborAdvertisements].
1501 The number of ICMP Neighbor Advertisement
1502 messages sent by the interface per second
1503 [ipv6IfIcmpOutNeighborAdvertisements].
1508 keyword, statistics about ICMPv6 error messages are reported.
1509 Note that ICMPv6 statistics depend on sadc option "-S IPV6" to be
1511 The following values are displayed (formal SNMP names between
1516 The number of ICMP messages per second which the interface
1517 received but determined as having ICMP-specific
1518 errors (bad ICMP checksums, bad length, etc.)
1519 [ipv6IfIcmpInErrors]
1524 The number of ICMP Destination Unreachable
1525 messages received by the interface per second
1526 [ipv6IfIcmpInDestUnreachs].
1531 The number of ICMP Destination Unreachable
1532 messages sent by the interface per second
1533 [ipv6IfIcmpOutDestUnreachs].
1538 The number of ICMP Time Exceeded messages
1539 received by the interface per second
1540 [ipv6IfIcmpInTimeExcds].
1545 The number of ICMP Time Exceeded messages sent
1546 by the interface per second
1547 [ipv6IfIcmpOutTimeExcds].
1552 The number of ICMP Parameter Problem messages
1553 received by the interface per second
1554 [ipv6IfIcmpInParmProblems].
1559 The number of ICMP Parameter Problem messages
1560 sent by the interface per second
1561 [ipv6IfIcmpOutParmProblems].
1566 The number of Redirect messages received
1567 by the interface per second
1568 [ipv6IfIcmpInRedirects].
1573 The number of Redirect messages sent by
1574 the interface by second
1575 [ipv6IfIcmpOutRedirects].
1580 The number of ICMP Packet Too Big messages
1581 received by the interface per second
1582 [ipv6IfIcmpInPktTooBigs].
1587 The number of ICMP Packet Too Big messages sent
1588 by the interface per second
1589 [ipv6IfIcmpOutPktTooBigs].
1594 keyword, statistics about UDPv6 network traffic are reported.
1595 Note that UDPv6 statistics depend on sadc option "-S IPV6" to be
1597 The following values are displayed (formal SNMP names between
1602 The total number of UDP datagrams delivered per second to UDP users
1608 The total number of UDP datagrams sent per second from this
1609 entity [udpOutDatagrams].
1614 The total number of received UDP datagrams per second for which there
1615 was no application at the destination port [udpNoPorts].
1620 The number of received UDP datagrams per second that could not be
1621 delivered for reasons other than the lack of an application
1622 at the destination port [udpInErrors].
1627 keyword is equivalent to specifying all the keywords above and therefore all the network
1628 activities are reported.
1631 .IP "-o [ filename ]"
1632 Save the readings in the file in binary form. Each reading
1633 is in a separate record. The default value of the
1635 parameter is the current daily data file, the
1637 file. The -o option is exclusive of the -f option.
1638 All the data available from the kernel are saved in the file (in fact,
1640 calls its data collector
1642 with the option "-S ALL". See sadc(8) manual page).
1643 .IP "-P { cpu [,...] | ALL }"
1644 Report per-processor statistics for the specified processor or processors.
1647 keyword reports statistics for each individual processor, and globally for
1649 Note that processor 0 is the first processor.
1651 Pretty-print device names. Use this option in conjunction with option -d.
1652 By default names are printed as
1654 where m and n are the major and minor numbers for the device.
1655 Use of this option displays the names of the devices as they (should) appear
1656 in /dev. Name mappings are controlled by
1657 .B @SYSCONFIG_DIR@/sysstat.ioconf.
1659 Report queue length and load averages. The following values are displayed:
1664 Run queue length (number of tasks waiting for run time).
1669 Number of tasks in the task list.
1674 System load average for the last minute.
1675 The load average is calculated as the average number of runnable or
1676 running tasks (R state), and the number of tasks in uninterruptible
1677 sleep (D state) over the specified interval.
1682 System load average for the past 5 minutes.
1687 System load average for the past 15 minutes.
1692 Number of tasks currently blocked, waiting for I/O to complete.
1696 Report memory utilization statistics.
1697 The following values are displayed:
1702 Amount of free memory available in kilobytes.
1707 Amount of used memory in kilobytes. This does not take into account memory
1708 used by the kernel itself.
1713 Percentage of used memory.
1718 Amount of memory used as buffers by the kernel in kilobytes.
1723 Amount of memory used to cache data by the kernel in kilobytes.
1728 Amount of memory in kilobytes needed for current workload. This is an estimate of how much
1729 RAM/swap is needed to guarantee that there never is out of memory.
1734 Percentage of memory needed for current workload in relation to the total amount of memory (RAM+swap).
1735 This number may be greater than 100% because the kernel usually overcommits memory.
1740 Amount of active memory in kilobytes (memory that has been used more recently
1741 and usually not reclaimed unless absolutely necessary).
1746 Amount of inactive memory in kilobytes (memory which has been less recently
1747 used. It is more eligible to be reclaimed for other purposes).
1751 Report memory statistics. The following values are displayed:
1756 Number of memory pages freed by the system per second.
1757 A negative value represents a number of pages allocated by the system.
1758 Note that a page has a size of 4 kB or 8 kB according to the machine architecture.
1763 Number of additional memory pages used as buffers by the system per second.
1764 A negative value means fewer pages used as buffers by the system.
1769 Number of additional memory pages cached by the system per second.
1770 A negative value means fewer pages in the cache.
1773 .IP "-s [ hh:mm:ss ]"
1774 Set the starting time of the data, causing the
1776 command to extract records time-tagged at, or following, the time
1777 specified. The default starting time is 08:00:00.
1778 Hours must be given in 24-hour format. This option can be
1779 used only when data are read from a file (option
1783 Report swap space utilization statistics.
1784 The following values are displayed:
1789 Amount of free swap space in kilobytes.
1794 Amount of used swap space in kilobytes.
1799 Percentage of used swap space.
1804 Amount of cached swap memory in kilobytes.
1805 This is memory that once was swapped out, is swapped back in
1806 but still also is in the swap area (if memory is needed it doesn't need
1807 to be swapped out again because it is already in the swap area. This
1813 Percentage of cached swap memory in relation to the amount of used swap space.
1817 When reading data from a daily data file, indicate that
1819 should display the timestamps in the original locale time of
1820 the data file creator. Without this option, the
1822 command displays the timestamps in the user's locale time.
1824 Report CPU utilization. The
1826 keyword indicates that all the CPU fields should be displayed.
1827 The report may show the following fields:
1832 Percentage of CPU utilization that occurred while executing at the user
1833 level (application). Note that this field includes time spent running
1839 Percentage of CPU utilization that occurred while executing at the user
1840 level (application). Note that this field does NOT include time spent
1841 running virtual processors.
1846 Percentage of CPU utilization that occurred while executing at the user
1847 level with nice priority.
1852 Percentage of CPU utilization that occurred while executing at the system
1853 level (kernel). Note that this field includes time spent servicing
1854 hardware and software interrupts.
1859 Percentage of CPU utilization that occurred while executing at the system
1860 level (kernel). Note that this field does NOT include time spent servicing
1861 hardware or software interrupts.
1866 Percentage of time that the CPU or CPUs were idle during which
1867 the system had an outstanding disk I/O request.
1872 Percentage of time spent in involuntary wait by the virtual CPU
1873 or CPUs while the hypervisor was servicing another virtual processor.
1878 Percentage of time spent by the CPU or CPUs to service hardware interrupts.
1883 Percentage of time spent by the CPU or CPUs to service software interrupts.
1888 Percentage of time spent by the CPU or CPUs to run a virtual processor.
1893 Percentage of time that the CPU or CPUs were idle and the system
1894 did not have an outstanding disk I/O request.
1897 Note: On SMP machines a processor that does not have any activity at all
1898 (0.00 for every field) is a disabled (offline) processor.
1901 Report status of inode, file and other kernel tables.
1902 The following values are displayed:
1907 Number of unused cache entries in the directory cache.
1912 Number of file handles used by the system.
1917 Number of inode handlers used by the system.
1922 Number of pseudo-terminals used by the system.
1926 Print version number then exit.
1928 Report task creation and system switching activity.
1933 Total number of tasks created per second.
1938 Total number of context switches per second.
1942 Report swapping statistics. The following values are displayed:
1947 Total number of swap pages the system brought in per second.
1952 Total number of swap pages the system brought out per second.
1956 Report TTY device activity. The following values are displayed:
1961 Number of receive interrupts per second for current serial line. Serial line number
1962 is given in the TTY column.
1967 Number of transmit interrupts per second for current serial line.
1972 Number of frame errors per second for current serial line.
1977 Number of parity errors per second for current serial line.
1982 Number of breaks per second for current serial line.
1987 Number of overrun errors per second for current serial line.
1994 command takes into account the following environment variables:
1997 If this variable exists and its value is
1999 then the current locale will be ignored when printing the date in the report header.
2002 command will use the ISO 8601 format (YYYY-MM-DD) instead.
2005 If this variable exists and its value is
2009 will save its data in UTC time (data will still be displayed in local time).
2011 will also use UTC time instead of local time to determine the current daily
2012 data file located in the
2014 directory. This variable may be useful for servers with users located across
2019 Report CPU utilization for each 2 seconds. 5 lines are displayed.
2022 .B sar -I 14 -o int14.file 2 10
2024 Report statistics on IRQ 14 for each 2 seconds. 10 lines are displayed.
2025 Data are stored in a file called
2029 .B sar -r -n DEV -f @SA_DIR@/sa16
2031 Display memory and network statistics saved in daily data file 'sa16'.
2036 Display all the statistics saved in current daily data file.
2039 filesystem must be mounted for the
2043 All the statistics are not necessarily available, depending on the kernel version used.
2045 assumes that you are using at least a 2.6 kernel.
2049 Indicate the daily data file, where the
2051 parameter is a number representing the day of the month.
2055 contains various files with system statistics.
2057 Sebastien Godard (sysstat <at> orange.fr)
2068 .I http://pagesperso-orange.fr/sebastien.godard/