{ "name": "sar.sysstat(1)", "mode": "man", "parameter": "sar.sysstat", "section": "1", "url": "/phpMan.php/man/sar.sysstat/1/json", "generated": "2026-05-27T16:08:20Z", "synopsis": "", "sections": [ { "name": "NAME", "level": 1, "content": "sar - Collect, report, or save system activity information.\n\n", "subsections": [] }, { "name": "SYNOPSIS", "level": 1, "content": "", "subsections": [ { "name": "sar [ -A ] [ -B ] [ -b ] [ -C ] [ -D ] [ -d ] [ -F [ MOUNT ] ] [ -H ] [ -h ] [ -p ] [ -r [", "level": 2, "content": "" }, { "name": "ALL ] ] [ -S ] [ -t ] [ -u [ ALL ] ] [ -V ] [ -v ] [ -W ] [ -w ] [ -y ] [ -z ] [ --dec={ 0 |", "level": 2, "content": "1 | 2 } ] [ --dev=devlist ] [ --fs=fslist ] [ --help ] [ --human ] [ --iface=ifacelist ] [\n--pretty ] [ --sadc ] [ -I { intlist | SUM | ALL } ] [ -P { cpulist | ALL } ] [ -m { key‐\nword[,...] | ALL } ] [ -n { keyword[,...] | ALL } ] [ -q [ keyword[,...] | ALL ] ] [ -j { SID\n| ID | LABEL | PATH | UUID | ... } ] [ -f [ filename ] | -o [ filename ] | -[0-9]+ ] [ -i in‐\nterval ] [ -s [ hh:mm[:ss] ] ] [ -e [ hh:mm[:ss] ] ] [ interval [ count ] ]\n\n" } ] }, { "name": "DESCRIPTION", "level": 1, "content": "The sar command writes to standard output the contents of selected cumulative activity coun‐\nters in the operating system. The accounting system, based on the values in the count and in‐\nterval parameters, writes information the specified number of times spaced at the specified\nintervals in seconds. If the interval parameter is set to zero, the sar command displays the\naverage statistics for the time since the system was started. If the interval parameter is\nspecified without the count parameter, then reports are generated continuously. The col‐\nlected data can also be saved in the file specified by the -o filename flag, in addition to\nbeing displayed onto the screen. If filename is omitted, sar uses the standard system activ‐\nity daily data file (see below). By default all the data available from the kernel are saved\nin the data file.\n\nThe sar command extracts and writes to standard output records previously saved in a file.\nThis file can be either the one specified by the -f flag or, by default, the standard system\nactivity daily data file. It is also possible to enter -1, -2 etc. as an argument to sar to\ndisplay data of that days ago. For example, -1 will point at the standard system activity\nfile of yesterday.\n\nStandard system activity daily data files are named saDD or saYYYYMMDD, where YYYY stands for\nthe current year, MM for the current month and DD for the current day. They are the default\nfiles used by sar only when no filename has been explicitly specified. When used to write\ndata to files (with its option -o), sar will use saYYYYMMDD if option -D has also been speci‐\nfied, else it will use saDD. When used to display the records previously saved in a file,\nsar will look for the most recent of saDD and saYYYYMMDD, and use it.\n\nStandard system activity daily data files are located in the /var/log/sysstat directory by\ndefault. Yet it is possible to specify an alternate location for them: If a directory (in‐\nstead of a plain file) is used with options -f or -o then it will be considered as the direc‐\ntory containing the data files.\n\nWithout the -P flag, the sar command reports system-wide (global among all processors) sta‐\ntistics, which are calculated as averages for values expressed as percentages, and as sums\notherwise. If the -P flag is given, the sar command reports activity which relates to the\nspecified processor or processors. If -P ALL is given, the sar command reports statistics for\neach individual processor and global statistics among all processors. Offline processors are\nnot displayed.\n\nYou can select information about specific system activities using flags. Not specifying any\nflags selects only CPU activity. Specifying the -A flag selects all possible activities.\n\nThe default version of the sar command (CPU utilization report) might be one of the first fa‐\ncilities the user runs to begin system activity investigation, because it monitors major sys‐\ntem resources. If CPU utilization is near 100 percent (user + nice + system), the workload\nsampled is CPU-bound.\n\nIf multiple samples and multiple reports are desired, it is convenient to specify an output\nfile for the sar command. Run the sar command as a background process. The syntax for this\nis:\n\nsar -o datafile interval count >/dev/null 2>&1 &\n\nAll data are captured in binary form and saved to a file (datafile). The data can then be\nselectively displayed with the sar command using the -f option. Set the interval and count\nparameters to select count records at interval second intervals. If the count parameter is\nnot set, all the records saved in the file will be selected. Collection of data in this man‐\nner is useful to characterize system usage over a period of time and determine peak usage\nhours.\n\nNote: The sar command only reports on local activities.\n\n", "subsections": [] }, { "name": "OPTIONS", "level": 1, "content": "", "subsections": [ { "name": "-A -bBdFHSvwWy -I -m -n -q -r -u", "level": 2, "content": "ALL. This option also implies specifying -I ALL -P ALL unless these options are ex‐\nplicitly set on the command line.\n" }, { "name": "-B", "level": 2, "content": "pgpgin/s\nTotal number of kilobytes the system paged in from disk per second.\n\npgpgout/s\nTotal number of kilobytes the system paged out to disk per second.\n\nfault/s\nNumber of page faults (major + minor) made by the system per second. This is\nnot a count of page faults that generate I/O, because some page faults can be\nresolved without I/O.\n\nmajflt/s\nNumber of major faults the system has made per second, those which have re‐\nquired loading a memory page from disk.\n\npgfree/s\nNumber of pages placed on the free list by the system per second.\n\npgscank/s\nNumber of pages scanned by the kswapd daemon per second.\n\npgscand/s\nNumber of pages scanned directly per second.\n\npgsteal/s\nNumber of pages the system has reclaimed from cache (pagecache and swapcache)\nper second to satisfy its memory demands.\n\n%vmeff Calculated as pgsteal / pgscan, this is a metric of the efficiency of page re‐\nclaim. If it is near 100% then almost every page coming off the tail of the in‐\nactive list is being reaped. If it gets too low (e.g. less than 30%) then the\nvirtual memory is having some difficulty. This field is displayed as zero if\nno pages have been scanned during the interval of time.\n" }, { "name": "-b", "level": 2, "content": "tps Total number of transfers per second that were issued to physical devices. A\ntransfer is an I/O request to a physical device. Multiple logical requests can\nbe combined into a single I/O request to the device. A transfer is of indeter‐\nminate size.\n\nrtps Total number of read requests per second issued to physical devices.\n\nwtps Total number of write requests per second issued to physical devices.\n\ndtps Total number of discard requests per second issued to physical devices.\n\nbread/s\nTotal amount of data read from the devices in blocks per second. Blocks are\nequivalent to sectors and therefore have a size of 512 bytes.\n\nbwrtn/s\nTotal amount of data written to devices in blocks per second.\n\nbdscd/s\nTotal amount of data discarded for devices in blocks per second.\n" }, { "name": "-C", "level": 2, "content": "sadc.\n" }, { "name": "-D", "level": 2, "content": "This option works only when used in conjunction with option -o to save data to file.\n" }, { "name": "-d", "level": 2, "content": "displayed as it (should) appear in /dev. sar uses data in /sys to determine the de‐\nvice name based on its major and minor numbers. If this name resolution fails, sar\nwill use name mapping controlled by /etc/sysstat/sysstat.ioconf file. Persistent de‐\nvice names can also be printed if option -j is used (see below). Statistics for all\ndevices are displayed unless a restricted list is specified using option --dev= (see\ncorresponding option entry). Note that disk activity depends on sadc's options -S\nDISK and -S XDISK to be collected. The following values are displayed:\n\ntps Total number of transfers per second that were issued to physical devices. A\ntransfer is an I/O request to a physical device. Multiple logical requests can\nbe combined into a single I/O request to the device. A transfer is of indeter‐\nminate size.\n\nrkB/s Number of kilobytes read from the device per second.\n\nwkB/s Number of kilobytes written to the device per second.\n\ndkB/s Number of kilobytes discarded for the device per second.\n\nareq-sz\nThe average size (in kilobytes) of the I/O requests that were issued to the de‐\nvice.\nNote: In previous versions, this field was known as avgrq-sz and was expressed\nin sectors.\n\naqu-sz The average queue length of the requests that were issued to the device.\nNote: In previous versions, this field was known as avgqu-sz.\n\nawait The average time (in milliseconds) for I/O requests issued to the device to be\nserved. This includes the time spent by the requests in queue and the time\nspent servicing them.\n\n%util Percentage of elapsed time during which I/O requests were issued to the device\n(bandwidth utilization for the device). Device saturation occurs when this\nvalue is close to 100% for devices serving requests serially. But for devices\nserving requests in parallel, such as RAID arrays and modern SSDs, this number\ndoes not reflect their performance limits.\n" }, { "name": "--dec={ 0 | 1 | 2 }", "level": 2, "content": "Specify the number of decimal places to use (0 to 2, default value is 2).\n\n--dev=devlist\nSpecify the block devices for which statistics are to be displayed by sar. devlist\nis a list of comma-separated device names.\n" }, { "name": "-e", "level": 2, "content": "Set the ending time of the report. The default ending time is 18:00:00. Hours must be\ngiven in 24-hour format. This option can be used when data are read from or written\nto a file (options -f or -o).\n" }, { "name": "-F [ MOUNT ]", "level": 2, "content": "Display statistics for currently mounted filesystems. Pseudo-filesystems are ignored.\nAt the end of the report, sar will display a summary of all those filesystems. Use of\nthe MOUNT parameter keyword indicates that mountpoint will be reported instead of\nfilesystem device. Statistics for all filesystems are displayed unless a restricted\nlist is specified using option --fs= (see corresponding option entry). Note that\nfilesystems statistics depend on sadc's option -S XDISK to be collected.\n\nThe following values are displayed:\n\nMBfsfree\nTotal amount of free space in megabytes (including space available only to\nprivileged user).\n\nMBfsused\nTotal amount of space used in megabytes.\n\n%fsused\nPercentage of filesystem space used, as seen by a privileged user.\n\n%ufsused\nPercentage of filesystem space used, as seen by an unprivileged user.\n\nIfree Total number of free file nodes in filesystem.\n\nIused Total number of file nodes used in filesystem.\n\n%Iused Percentage of file nodes used in filesystem.\n" }, { "name": "-f", "level": 2, "content": "Extract records from filename (created by the -o filename flag). The default value of\nthe filename parameter is the current standard system activity daily data file. If\nfilename is a directory instead of a plain file then it is considered as the directory\nwhere the standard system activity daily data files are located. Option -f is exclu‐\nsive of option -o.\n\n--fs=fslist\nSpecify the filesystems for which statistics are to be displayed by sar. fslist is a\nlist of comma-separated filesystem names or mountpoints.\n" }, { "name": "-H", "level": 2, "content": "kbhugfree\nAmount of hugepages memory in kilobytes that is not yet allocated.\n\nkbhugused\nAmount of hugepages memory in kilobytes that has been allocated.\n\n%hugused\nPercentage of total hugepages memory that has been allocated.\n\nkbhugrsvd\nAmount of reserved hugepages memory in kilobytes.\n\nkbhugsurp\nAmount of surplus hugepages memory in kilobytes.\n" }, { "name": "-h --pretty --human", "level": 2, "content": "--help Display a short help message then exit.\n" }, { "name": "--human", "level": 2, "content": "Print sizes in human readable format (e.g. 1.0k, 1.2M, etc.) The units displayed with\nthis option supersede any other default units (e.g. kilobytes, sectors...) associated\nwith the metrics.\n" }, { "name": "-I", "level": 2, "content": "Report statistics for interrupts. intlist is a list of comma-separated values or\nrange of values (e.g., 0-16,35,400-). The SUM keyword indicates that the total number\nof interrupts received per second is to be displayed. The ALL keyword indicates that\nstatistics from all interrupts, including potential APIC interrupt sources, are to be\nreported. Note that interrupt statistics depend on sadc's option -S INT to be col‐\nlected.\n" }, { "name": "-i", "level": 2, "content": "Select data records at seconds as close as possible to the number specified by the in‐\nterval parameter.\n\n--iface=ifacelist\nSpecify the network interfaces for which statistics are to be displayed by sar.\nifacelist is a list of comma-separated interface names.\n" }, { "name": "-j { SID | ID | LABEL | PATH | UUID | ... }", "level": 2, "content": "Display persistent device names. Use this option in conjunction with option -d. Key‐\nwords ID, LABEL, etc. specify the type of the persistent name. These keywords are not\nlimited, only prerequisite is that directory with required persistent names is present\nin /dev/disk. Keyword SID tries to get a stable identifier to use as the device name.\nA stable identifier won't change across reboots for the same physical device. If it\nexists, this identifier is normally the WWN (World Wide Name) of the device, as read\nfrom the /dev/disk/by-id directory.\n" }, { "name": "-m", "level": 2, "content": "Report power management statistics. Note that these statistics depend on sadc's op‐\ntion -S POWER to be collected.\n\nPossible keywords are CPU, FAN, FREQ, IN, TEMP and USB.\n\nWith the CPU keyword, statistics about CPU are reported. The following value is dis‐\nplayed:\n\nMHz Instantaneous CPU clock frequency in MHz.\n\n\nWith the FAN keyword, statistics about fans speed are reported. The following values\nare displayed:\n\nrpm Fan speed expressed in revolutions per minute.\n\ndrpm This field is calculated as the difference between current fan speed (rpm) and\nits low limit (fanmin).\n\nDEVICE Sensor device name.\n\n\nWith the FREQ keyword, statistics about CPU clock frequency are reported. The follow‐\ning value is displayed:\n\nwghMHz Weighted average CPU clock frequency in MHz. Note that the cpufreq-stats\ndriver must be compiled in the kernel for this option to work.\n\n\nWith the IN keyword, statistics about voltage inputs are reported. The following val‐\nues are displayed:\n\ninV Voltage input expressed in Volts.\n\n%in Relative input value. A value of 100% means that voltage input has reached its\nhigh limit (inmax) whereas a value of 0% means that it has reached its low\nlimit (inmin).\n\nDEVICE Sensor device name.\n\n\nWith the TEMP keyword, statistics about devices temperature are reported. The follow‐\ning values are displayed:\n\ndegC Device temperature expressed in degrees Celsius.\n\n%temp Relative device temperature. A value of 100% means that temperature has reached\nits high limit (tempmax).\n\nDEVICE Sensor device name.\n\n\nWith the USB keyword, the sar command takes a snapshot of all the USB devices cur‐\nrently plugged into the system. At the end of the report, sar will display a summary\nof all those USB devices. The following values are displayed:\n\nBUS Root hub number of the USB device.\n\nidvendor\nVendor ID number (assigned by USB organization).\n\nidprod Product ID number (assigned by Manufacturer).\n\nmaxpower\nMaximum power consumption of the device (expressed in mA).\n\nmanufact\nManufacturer name.\n\nproduct\nProduct name.\n\n\nThe ALL keyword is equivalent to specifying all the keywords above and therefore all\nthe power management statistics are reported.\n" }, { "name": "-n", "level": 2, "content": "Report network statistics.\n\nPossible keywords are DEV, EDEV, FC, ICMP, EICMP, ICMP6, EICMP6, IP, EIP, IP6, EIP6,\nNFS, NFSD, SOCK, SOCK6, SOFT, TCP, ETCP, UDP and UDP6.\n\nWith the DEV keyword, statistics from the network devices are reported. Statistics\nfor all network interfaces are displayed unless a restricted list is specified using\noption --iface= (see corresponding option entry). The following values are displayed:\n\nIFACE Name of the network interface for which statistics are reported.\n\nrxpck/s\nTotal number of packets received per second.\n\ntxpck/s\nTotal number of packets transmitted per second.\n\nrxkB/s Total number of kilobytes received per second.\n\ntxkB/s Total number of kilobytes transmitted per second.\n\nrxcmp/s\nNumber of compressed packets received per second (for cslip etc.).\n\ntxcmp/s\nNumber of compressed packets transmitted per second.\n\nrxmcst/s\nNumber of multicast packets received per second.\n\n%ifutil\nUtilization percentage of the network interface. For half-duplex interfaces,\nutilization is calculated using the sum of rxkB/s and txkB/s as a percentage of\nthe interface speed. For full-duplex, this is the greater of rxkB/S or txkB/s.\n\n\nWith the EDEV keyword, statistics on failures (errors) from the network devices are\nreported. Statistics for all network interfaces are displayed unless a restricted\nlist is specified using option --iface= (see corresponding option entry). The follow‐\ning values are displayed:\n\nIFACE Name of the network interface for which statistics are reported.\n\nrxerr/s\nTotal number of bad packets received per second.\n\ntxerr/s\nTotal number of errors that happened per second while transmitting packets.\n\ncoll/s Number of collisions that happened per second while transmitting packets.\n\nrxdrop/s\nNumber of received packets dropped per second because of a lack of space in\nlinux buffers.\n\ntxdrop/s\nNumber of transmitted packets dropped per second because of a lack of space in\nlinux buffers.\n\ntxcarr/s\nNumber of carrier-errors that happened per second while transmitting packets.\n\nrxfram/s\nNumber of frame alignment errors that happened per second on received packets.\n\nrxfifo/s\nNumber of FIFO overrun errors that happened per second on received packets.\n\ntxfifo/s\nNumber of FIFO overrun errors that happened per second on transmitted packets.\n\n\nWith the FC keyword, statistics about fibre channel traffic are reported. Note that\nfibre channel statistics depend on sadc's option -S DISK to be collected. The follow‐\ning values are displayed:\n\nFCHOST Name of the fibre channel host bus adapter (HBA) interface for which statistics\nare reported.\n\nfchrxf/s\nThe total number of frames received per second.\n\nfchtxf/s\nThe total number of frames transmitted per second.\n\nfchrxw/s\nThe total number of transmission words received per second.\n\nfchtxw/s\nThe total number of transmission words transmitted per second.\n\n\nWith the ICMP keyword, statistics about ICMPv4 network traffic are reported. Note\nthat ICMPv4 statistics depend on sadc's option -S SNMP to be collected. The following\nvalues are displayed (formal SNMP names between square brackets):\n\nimsg/s The total number of ICMP messages which the entity received per second [icmpIn‐\nMsgs]. Note that this counter includes all those counted by ierr/s.\n\nomsg/s The total number of ICMP messages which this entity attempted to send per sec‐\nond [icmpOutMsgs]. Note that this counter includes all those counted by\noerr/s.\n\niech/s The number of ICMP Echo (request) messages received per second [icmpInEchos].\n\niechr/s\nThe number of ICMP Echo Reply messages received per second [icmpInEchoReps].\n\noech/s The number of ICMP Echo (request) messages sent per second [icmpOutEchos].\n\noechr/s\nThe number of ICMP Echo Reply messages sent per second [icmpOutEchoReps].\n\nitm/s The number of ICMP Timestamp (request) messages received per second [icmpIn‐\nTimestamps].\n\nitmr/s The number of ICMP Timestamp Reply messages received per second [icmpInTimes‐\ntampReps].\n\notm/s The number of ICMP Timestamp (request) messages sent per second [icmpOutTimes‐\ntamps].\n\notmr/s The number of ICMP Timestamp Reply messages sent per second [icmpOutTimestam‐\npReps].\n\niadrmk/s\nThe number of ICMP Address Mask Request messages received per second [icmpInAd‐\ndrMasks].\n\niadrmkr/s\nThe number of ICMP Address Mask Reply messages received per second [icmpInAddr‐\nMaskReps].\n\noadrmk/s\nThe number of ICMP Address Mask Request messages sent per second [icmpOutAddr‐\nMasks].\n\noadrmkr/s\nThe number of ICMP Address Mask Reply messages sent per second [icmpOutAddr‐\nMaskReps].\n\n\nWith the EICMP keyword, statistics about ICMPv4 error messages are reported. Note\nthat ICMPv4 statistics depend on sadc's option -S SNMP to be collected. The following\nvalues are displayed (formal SNMP names between square brackets):\n\nierr/s The number of ICMP messages per second which the entity received but determined\nas having ICMP-specific errors (bad ICMP checksums, bad length, etc.) [icmpIn‐\nErrors].\n\noerr/s The number of ICMP messages per second which this entity did not send due to\nproblems discovered within ICMP such as a lack of buffers [icmpOutErrors].\n\nidstunr/s\nThe number of ICMP Destination Unreachable messages received per second\n[icmpInDestUnreachs].\n\nodstunr/s\nThe number of ICMP Destination Unreachable messages sent per second [icmpOut‐\nDestUnreachs].\n\nitmex/s\nThe number of ICMP Time Exceeded messages received per second [icmpInTimeEx‐\ncds].\n\notmex/s\nThe number of ICMP Time Exceeded messages sent per second [icmpOutTimeExcds].\n\niparmpb/s\nThe number of ICMP Parameter Problem messages received per second [icmpInParm‐\nProbs].\n\noparmpb/s\nThe number of ICMP Parameter Problem messages sent per second [icmpOutParm‐\nProbs].\n\nisrcq/s\nThe number of ICMP Source Quench messages received per second [icmpInSrc‐\nQuenchs].\n\nosrcq/s\nThe number of ICMP Source Quench messages sent per second [icmpOutSrcQuenchs].\n\niredir/s\nThe number of ICMP Redirect messages received per second [icmpInRedirects].\n\noredir/s\nThe number of ICMP Redirect messages sent per second [icmpOutRedirects].\n\n\nWith the ICMP6 keyword, statistics about ICMPv6 network traffic are reported. Note\nthat ICMPv6 statistics depend on sadc's option -S IPV6 to be collected. The following\nvalues are displayed (formal SNMP names between square brackets):\n\nimsg6/s\nThe total number of ICMP messages received by the interface per second which\nincludes all those counted by ierr6/s [ipv6IfIcmpInMsgs].\n\nomsg6/s\nThe total number of ICMP messages which this interface attempted to send per\nsecond [ipv6IfIcmpOutMsgs].\n\niech6/s\nThe number of ICMP Echo (request) messages received by the interface per second\n[ipv6IfIcmpInEchos].\n\niechr6/s\nThe number of ICMP Echo Reply messages received by the interface per second\n[ipv6IfIcmpInEchoReplies].\n\noechr6/s\nThe number of ICMP Echo Reply messages sent by the interface per second\n[ipv6IfIcmpOutEchoReplies].\n\nigmbq6/s\nThe number of ICMPv6 Group Membership Query messages received by the interface\nper second [ipv6IfIcmpInGroupMembQueries].\n\nigmbr6/s\nThe number of ICMPv6 Group Membership Response messages received by the inter‐\nface per second [ipv6IfIcmpInGroupMembResponses].\n\nogmbr6/s\nThe number of ICMPv6 Group Membership Response messages sent per second\n[ipv6IfIcmpOutGroupMembResponses].\n\nigmbrd6/s\nThe number of ICMPv6 Group Membership Reduction messages received by the inter‐\nface per second [ipv6IfIcmpInGroupMembReductions].\n\nogmbrd6/s\nThe number of ICMPv6 Group Membership Reduction messages sent per second\n[ipv6IfIcmpOutGroupMembReductions].\n\nirtsol6/s\nThe number of ICMP Router Solicit messages received by the interface per second\n[ipv6IfIcmpInRouterSolicits].\n\nortsol6/s\nThe number of ICMP Router Solicitation messages sent by the interface per sec‐\nond [ipv6IfIcmpOutRouterSolicits].\n\nirtad6/s\nThe number of ICMP Router Advertisement messages received by the interface per\nsecond [ipv6IfIcmpInRouterAdvertisements].\n\ninbsol6/s\nThe number of ICMP Neighbor Solicit messages received by the interface per sec‐\nond [ipv6IfIcmpInNeighborSolicits].\n\nonbsol6/s\nThe number of ICMP Neighbor Solicitation messages sent by the interface per\nsecond [ipv6IfIcmpOutNeighborSolicits].\n\ninbad6/s\nThe number of ICMP Neighbor Advertisement messages received by the interface\nper second [ipv6IfIcmpInNeighborAdvertisements].\n\nonbad6/s\nThe number of ICMP Neighbor Advertisement messages sent by the interface per\nsecond [ipv6IfIcmpOutNeighborAdvertisements].\n\n\nWith the EICMP6 keyword, statistics about ICMPv6 error messages are reported. Note\nthat ICMPv6 statistics depend on sadc's option -S IPV6 to be collected. The following\nvalues are displayed (formal SNMP names between square brackets):\n\nierr6/s\nThe number of ICMP messages per second which the interface received but deter‐\nmined as having ICMP-specific errors (bad ICMP checksums, bad length, etc.)\n[ipv6IfIcmpInErrors]\n\nidtunr6/s\nThe number of ICMP Destination Unreachable messages received by the interface\nper second [ipv6IfIcmpInDestUnreachs].\n\nodtunr6/s\nThe number of ICMP Destination Unreachable messages sent by the interface per\nsecond [ipv6IfIcmpOutDestUnreachs].\n\nitmex6/s\nThe number of ICMP Time Exceeded messages received by the interface per second\n[ipv6IfIcmpInTimeExcds].\n\notmex6/s\nThe number of ICMP Time Exceeded messages sent by the interface per second\n[ipv6IfIcmpOutTimeExcds].\n\niprmpb6/s\nThe number of ICMP Parameter Problem messages received by the interface per\nsecond [ipv6IfIcmpInParmProblems].\n\noprmpb6/s\nThe number of ICMP Parameter Problem messages sent by the interface per second\n[ipv6IfIcmpOutParmProblems].\n\niredir6/s\nThe number of Redirect messages received by the interface per second\n[ipv6IfIcmpInRedirects].\n\noredir6/s\nThe number of Redirect messages sent by the interface by second [ipv6IfIcm‐\npOutRedirects].\n\nipck2b6/s\nThe number of ICMP Packet Too Big messages received by the interface per second\n[ipv6IfIcmpInPktTooBigs].\n\nopck2b6/s\nThe number of ICMP Packet Too Big messages sent by the interface per second\n[ipv6IfIcmpOutPktTooBigs].\n\n\nWith the IP keyword, statistics about IPv4 network traffic are reported. Note that\nIPv4 statistics depend on sadc's option -S SNMP to be collected. The following values\nare displayed (formal SNMP names between square brackets):\n\nirec/s The total number of input datagrams received from interfaces per second, in‐\ncluding those received in error [ipInReceives].\n\nfwddgm/s\nThe number of input datagrams per second, for which this entity was not their\nfinal IP destination, as a result of which an attempt was made to find a route\nto forward them to that final destination [ipForwDatagrams].\n\nidel/s The total number of input datagrams successfully delivered per second to IP\nuser-protocols (including ICMP) [ipInDelivers].\n\norq/s The total number of IP datagrams which local IP user-protocols (including ICMP)\nsupplied per second to IP in requests for transmission [ipOutRequests]. Note\nthat this counter does not include any datagrams counted in fwddgm/s.\n\nasmrq/s\nThe number of IP fragments received per second which needed to be reassembled\nat this entity [ipReasmReqds].\n\nasmok/s\nThe number of IP datagrams successfully re-assembled per second [ipReasmOKs].\n\nfragok/s\nThe number of IP datagrams that have been successfully fragmented at this en‐\ntity per second [ipFragOKs].\n\nfragcrt/s\nThe number of IP datagram fragments that have been generated per second as a\nresult of fragmentation at this entity [ipFragCreates].\n\n\nWith the EIP keyword, statistics about IPv4 network errors are reported. Note that\nIPv4 statistics depend on sadc's option -S SNMP to be collected. The following values\nare displayed (formal SNMP names between square brackets):\n\nihdrerr/s\nThe number of input datagrams discarded per second due to errors in their IP\nheaders, including bad checksums, version number mismatch, other format errors,\ntime-to-live exceeded, errors discovered in processing their IP options, etc.\n[ipInHdrErrors]\n\niadrerr/s\nThe number of input datagrams discarded per second because the IP address in\ntheir IP header's destination field was not a valid address to be received at\nthis entity. This count includes invalid addresses (e.g., 0.0.0.0) and ad‐\ndresses of unsupported Classes (e.g., Class E). For entities which are not IP\nrouters and therefore do not forward datagrams, this counter includes datagrams\ndiscarded because the destination address was not a local address [ipInAddr‐\nErrors].\n\niukwnpr/s\nThe number of locally-addressed datagrams received successfully but discarded\nper second because of an unknown or unsupported protocol [ipInUnknownProtos].\n\nidisc/s\nThe number of input IP datagrams per second for which no problems were encoun‐\ntered to prevent their continued processing, but which were discarded (e.g.,\nfor lack of buffer space) [ipInDiscards]. Note that this counter does not in‐\nclude any datagrams discarded while awaiting re-assembly.\n\nodisc/s\nThe number of output IP datagrams per second for which no problem was encoun‐\ntered to prevent their transmission to their destination, but which were dis‐\ncarded (e.g., for lack of buffer space) [ipOutDiscards]. Note that this\ncounter would include datagrams counted in fwddgm/s if any such packets met\nthis (discretionary) discard criterion.\n\nonort/s\nThe number of IP datagrams discarded per second because no route could be found\nto transmit them to their destination [ipOutNoRoutes]. Note that this counter\nincludes any packets counted in fwddgm/s which meet this 'no-route' criterion.\nNote that this includes any datagrams which a host cannot route because all of\nits default routers are down.\n\nasmf/s The number of failures detected per second by the IP re-assembly algorithm (for\nwhatever reason: timed out, errors, etc) [ipReasmFails]. Note that this is not\nnecessarily a count of discarded IP fragments since some algorithms can lose\ntrack of the number of fragments by combining them as they are received.\n\nfragf/s\nThe number of IP datagrams that have been discarded per second because they\nneeded to be fragmented at this entity but could not be, e.g., because their\nDon't Fragment flag was set [ipFragFails].\n\n\nWith the IP6 keyword, statistics about IPv6 network traffic are reported. Note that\nIPv6 statistics depend on sadc's option -S IPV6 to be collected. The following values\nare displayed (formal SNMP names between square brackets):\n\nirec6/s\nThe total number of input datagrams received from interfaces per second, in‐\ncluding those received in error [ipv6IfStatsInReceives].\n\nfwddgm6/s\nThe number of output datagrams per second which this entity received and for‐\nwarded to their final destinations [ipv6IfStatsOutForwDatagrams].\n\nidel6/s\nThe total number of datagrams successfully delivered per second to IPv6\nuser-protocols (including ICMP) [ipv6IfStatsInDelivers].\n\norq6/s The total number of IPv6 datagrams which local IPv6 user-protocols (including\nICMP) supplied per second to IPv6 in requests for transmission [ipv6IfStat‐\nsOutRequests]. Note that this counter does not include any datagrams counted\nin fwddgm6/s.\n\nasmrq6/s\nThe number of IPv6 fragments received per second which needed to be reassembled\nat this interface [ipv6IfStatsReasmReqds].\n\nasmok6/s\nThe number of IPv6 datagrams successfully reassembled per second [ipv6IfStat‐\nsReasmOKs].\n\nimcpck6/s\nThe number of multicast packets received per second by the interface [ipv6IfS‐\ntatsInMcastPkts].\n\nomcpck6/s\nThe number of multicast packets transmitted per second by the interface\n[ipv6IfStatsOutMcastPkts].\n\nfragok6/s\nThe number of IPv6 datagrams that have been successfully fragmented at this\noutput interface per second [ipv6IfStatsOutFragOKs].\n\nfragcr6/s\nThe number of output datagram fragments that have been generated per second as\na result of fragmentation at this output interface [ipv6IfStatsOutFragCreates].\n\n\nWith the EIP6 keyword, statistics about IPv6 network errors are reported. Note that\nIPv6 statistics depend on sadc's option -S IPV6 to be collected. The following values\nare displayed (formal SNMP names between square brackets):\n\nihdrer6/s\nThe number of input datagrams discarded per second due to errors in their IPv6\nheaders, including version number mismatch, other format errors, hop count ex‐\nceeded, errors discovered in processing their IPv6 options, etc. [ipv6IfS‐\ntatsInHdrErrors]\n\niadrer6/s\nThe number of input datagrams discarded per second because the IPv6 address in\ntheir IPv6 header's destination field was not a valid address to be received at\nthis entity. This count includes invalid addresses (e.g., ::0) and unsupported\naddresses (e.g., addresses with unallocated prefixes). For entities which are\nnot IPv6 routers and therefore do not forward datagrams, this counter includes\ndatagrams discarded because the destination address was not a local address\n[ipv6IfStatsInAddrErrors].\n\niukwnp6/s\nThe number of locally-addressed datagrams received successfully but discarded\nper second because of an unknown or unsupported protocol [ipv6IfStatsInUnknown‐\nProtos].\n\ni2big6/s\nThe number of input datagrams that could not be forwarded per second because\ntheir size exceeded the link MTU of outgoing interface [ipv6IfStatsInTooBigEr‐\nrors].\n\nidisc6/s\nThe number of input IPv6 datagrams per second for which no problems were en‐\ncountered to prevent their continued processing, but which were discarded\n(e.g., for lack of buffer space) [ipv6IfStatsInDiscards]. Note that this\ncounter does not include any datagrams discarded while awaiting re-assembly.\n\nodisc6/s\nThe number of output IPv6 datagrams per second for which no problem was encoun‐\ntered to prevent their transmission to their destination, but which were dis‐\ncarded (e.g., for lack of buffer space) [ipv6IfStatsOutDiscards]. Note that\nthis counter would include datagrams counted in fwddgm6/s if any such packets\nmet this (discretionary) discard criterion.\n\ninort6/s\nThe number of input datagrams discarded per second because no route could be\nfound to transmit them to their destination [ipv6IfStatsInNoRoutes].\n\nonort6/s\nThe number of locally generated IP datagrams discarded per second because no\nroute could be found to transmit them to their destination [unknown formal SNMP\nname].\n\nasmf6/s\nThe number of failures detected per second by the IPv6 re-assembly algorithm\n(for whatever reason: timed out, errors, etc.) [ipv6IfStatsReasmFails]. Note\nthat this is not necessarily a count of discarded IPv6 fragments since some al‐\ngorithms can lose track of the number of fragments by combining them as they\nare received.\n\nfragf6/s\nThe number of IPv6 datagrams that have been discarded per second because they\nneeded to be fragmented at this output interface but could not be [ipv6IfStat‐\nsOutFragFails].\n\nitrpck6/s\nThe number of input datagrams discarded per second because datagram frame\ndidn't carry enough data [ipv6IfStatsInTruncatedPkts].\n\n\nWith the NFS keyword, statistics about NFS client activity are reported. The follow‐\ning values are displayed:\n\ncall/s Number of RPC requests made per second.\n\nretrans/s\nNumber of RPC requests per second, those which needed to be retransmitted (for\nexample because of a server timeout).\n\nread/s Number of 'read' RPC calls made per second.\n\nwrite/s\nNumber of 'write' RPC calls made per second.\n\naccess/s\nNumber of 'access' RPC calls made per second.\n\ngetatt/s\nNumber of 'getattr' RPC calls made per second.\n\n\nWith the NFSD keyword, statistics about NFS server activity are reported. The follow‐\ning values are displayed:\n\nscall/s\nNumber of RPC requests received per second.\n\nbadcall/s\nNumber of bad RPC requests received per second, those whose processing gener‐\nated an error.\n\npacket/s\nNumber of network packets received per second.\n\nudp/s Number of UDP packets received per second.\n\ntcp/s Number of TCP packets received per second.\n\nhit/s Number of reply cache hits per second.\n\nmiss/s Number of reply cache misses per second.\n\nsread/s\nNumber of 'read' RPC calls received per second.\n\nswrite/s\nNumber of 'write' RPC calls received per second.\n\nsaccess/s\nNumber of 'access' RPC calls received per second.\n\nsgetatt/s\nNumber of 'getattr' RPC calls received per second.\n\n\nWith the SOCK keyword, statistics on sockets in use are reported (IPv4). The follow‐\ning values are displayed:\n\ntotsck Total number of sockets used by the system.\n\ntcpsck Number of TCP sockets currently in use.\n\nudpsck Number of UDP sockets currently in use.\n\nrawsck Number of RAW sockets currently in use.\n\nip-frag\nNumber of IP fragments currently in queue.\n\ntcp-tw Number of TCP sockets in TIMEWAIT state.\n\n\nWith the SOCK6 keyword, statistics on sockets in use are reported (IPv6). Note that\nIPv6 statistics depend on sadc's option -S IPV6 to be collected. The following values\nare displayed:\n\ntcp6sck\nNumber of TCPv6 sockets currently in use.\n\nudp6sck\nNumber of UDPv6 sockets currently in use.\n\nraw6sck\nNumber of RAWv6 sockets currently in use.\n\nip6-frag\nNumber of IPv6 fragments currently in use.\n\n\nWith the SOFT keyword, statistics about software-based network processing are re‐\nported. The following values are displayed:\n\ntotal/s\nThe total number of network frames processed per second.\n\ndropd/s\nThe total number of network frames dropped per second because there was no room\non the processing queue.\n\nsqueezd/s\nThe number of times the softirq handler function terminated per second because\nits budget was consumed or the time limit was reached, but more work could have\nbeen done.\n\nrxrps/s\nThe number of times the CPU has been woken up per second to process packets via\nan inter-processor interrupt.\n\nflwlim/s\nThe number of times the flow limit has been reached per second. Flow limiting\nis an optional RPS feature that can be used to limit the number of packets\nqueued to the backlog for each flow to a certain amount. This can help ensure\nthat smaller flows are processed even though much larger flows are pushing\npackets in.\n\n\nWith the TCP keyword, statistics about TCPv4 network traffic are reported. Note that\nTCPv4 statistics depend on sadc's option -S SNMP to be collected. The following val‐\nues are displayed (formal SNMP names between square brackets):\n\nactive/s\nThe number of times TCP connections have made a direct transition to the\nSYN-SENT state from the CLOSED state per second [tcpActiveOpens].\n\npassive/s\nThe number of times TCP connections have made a direct transition to the\nSYN-RCVD state from the LISTEN state per second [tcpPassiveOpens].\n\niseg/s The total number of segments received per second, including those received in\nerror [tcpInSegs]. This count includes segments received on currently estab‐\nlished connections.\n\noseg/s The total number of segments sent per second, including those on current con‐\nnections but excluding those containing only retransmitted octets [tcpOutSegs].\n\n\nWith the ETCP keyword, statistics about TCPv4 network errors are reported. Note that\nTCPv4 statistics depend on sadc's option -S SNMP to be collected. The following val‐\nues are displayed (formal SNMP names between square brackets):\n\natmptf/s\nThe number of times per second TCP connections have made a direct transition to\nthe CLOSED state from either the SYN-SENT state or the SYN-RCVD state, plus the\nnumber of times per second TCP connections have made a direct transition to the\nLISTEN state from the SYN-RCVD state [tcpAttemptFails].\n\nestres/s\nThe number of times per second TCP connections have made a direct transition to\nthe CLOSED state from either the ESTABLISHED state or the CLOSE-WAIT state\n[tcpEstabResets].\n\nretrans/s\nThe total number of segments retransmitted per second - that is, the number of\nTCP segments transmitted containing one or more previously transmitted octets\n[tcpRetransSegs].\n\nisegerr/s\nThe total number of segments received in error (e.g., bad TCP checksums) per\nsecond [tcpInErrs].\n\norsts/s\nThe number of TCP segments sent per second containing the RST flag\n[tcpOutRsts].\n\n\nWith the UDP keyword, statistics about UDPv4 network traffic are reported. Note that\nUDPv4 statistics depend on sadc's option -S SNMP to be collected. The following val‐\nues are displayed (formal SNMP names between square brackets):\n\nidgm/s The total number of UDP datagrams delivered per second to UDP users [udpInData‐\ngrams].\n\nodgm/s The total number of UDP datagrams sent per second from this entity [udpOutData‐\ngrams].\n\nnoport/s\nThe total number of received UDP datagrams per second for which there was no\napplication at the destination port [udpNoPorts].\n\nidgmerr/s\nThe number of received UDP datagrams per second that could not be delivered for\nreasons other than the lack of an application at the destination port [udpIn‐\nErrors].\n\n\nWith the UDP6 keyword, statistics about UDPv6 network traffic are reported. Note that\nUDPv6 statistics depend on sadc's option -S IPV6 to be collected. The following val‐\nues are displayed (formal SNMP names between square brackets):\n\nidgm6/s\nThe total number of UDP datagrams delivered per second to UDP users [udpInData‐\ngrams].\n\nodgm6/s\nThe total number of UDP datagrams sent per second from this entity [udpOutData‐\ngrams].\n\nnoport6/s\nThe total number of received UDP datagrams per second for which there was no\napplication at the destination port [udpNoPorts].\n\nidgmer6/s\nThe number of received UDP datagrams per second that could not be delivered for\nreasons other than the lack of an application at the destination port [udpIn‐\nErrors].\n\n\nThe ALL keyword is equivalent to specifying all the keywords above and therefore all\nthe network activities are reported.\n" }, { "name": "-o", "level": 2, "content": "Save the readings in the file in binary form. Each reading is in a separate record.\nThe default value of the filename parameter is the current standard system activity\ndaily data file. If filename is a directory instead of a plain file then it is consid‐\nered as the directory where the standard system activity daily data files are located.\nOption -o is exclusive of option -f. All the data available from the kernel are saved\nin the file (in fact, sar calls its data collector sadc with the option -S ALL. See\nsadc(8) manual page).\n" }, { "name": "-P", "level": 2, "content": "Report per-processor statistics for the specified processor or processors. cpulist\nis a list of comma-separated values or range of values (e.g., 0,2,4-7,12-). Note that\nprocessor 0 is the first processor, and processor all is the global average among all\nprocessors. Specifying the ALL keyword reports statistics for each individual proces‐\nsor, and globally for all processors. Offline processors are not displayed.\n" }, { "name": "-p --pretty", "level": 2, "content": "Make reports easier to read by a human. This option may be especially useful when\ndisplaying e.g., network interfaces or block devices statistics.\n" }, { "name": "-q", "level": 2, "content": "Report system load and pressure-stall statistics.\n\nPossible keywords are CPU, IO, LOAD, MEM and PSI\".\n\nWith the CPU keyword, CPU pressure statistics are reported. The following values are\ndisplayed:\n\n%scpu-10\nPercentage of the time that at least some runnable tasks were delayed because\nthe CPU was unavailable to them, over the last 10 second window.\n\n%scpu-60\nPercentage of the time that at least some runnable tasks were delayed because\nthe CPU was unavailable to them, over the last 60 second window.\n\n%scpu-300\nPercentage of the time that at least some runnable tasks were delayed because\nthe CPU was unavailable to them, over the last 300 second window.\n\n%scpu Percentage of the time that at least some runnable tasks were delayed because\nthe CPU was unavailable to them, over the last time interval.\n\n\nWith the IO keyword, I/O pressure statistics are reported. The following values are\ndisplayed:\n\n%sio-10\nPercentage of the time that at least some tasks lost waiting for I/O, over the\nlast 10 second window.\n\n%sio-60\nPercentage of the time that at least some tasks lost waiting for I/O, over the\nlast 60 second window.\n\n%sio-300\nPercentage of the time that at least some tasks lost waiting for I/O, over the\nlast 300 second window.\n\n%sio Percentage of the time that at least some tasks lost waiting for I/O, over the\nlast time interval.\n\n%fio-10\nPercentage of the time during which all non-idle tasks were stalled waiting for\nI/O, over the last 10 second window.\n\n%fio-60\nPercentage of the time during which all non-idle tasks were stalled waiting for\nI/O, over the last 60 second window.\n\n%fio-300\nPercentage of the time during which all non-idle tasks were stalled waiting for\nI/O, over the last 300 second window.\n\n%fio Percentage of the time during which all non-idle tasks were stalled waiting for\nI/O, over the last time interval.\n\n\nWith the LOAD keyword, queue length and load averages statistics are reported. The\nfollowing values are displayed:\n\nrunq-sz\nRun queue length (number of tasks running or waiting for run time).\n\nplist-sz\nNumber of tasks in the task list.\n\nldavg-1\nSystem load average for the last minute. The load average is calculated as the\naverage number of runnable or running tasks (R state), and the number of tasks\nin uninterruptible sleep (D state) over the specified interval.\n\nldavg-5\nSystem load average for the past 5 minutes.\n\nldavg-15\nSystem load average for the past 15 minutes.\n\nblocked\nNumber of tasks currently blocked, waiting for I/O to complete.\n\n\nWith the MEM keyword, memory pressure statistics are reported. The following values\nare displayed:\n\n%smem-10\nPercentage of the time during which at least some tasks were waiting for memory\nresources, over the last 10 second window.\n\n%smem-60\nPercentage of the time during which at least some tasks were waiting for memory\nresources, over the last 60 second window.\n\n%smem-300\nPercentage of the time during which at least some tasks were waiting for memory\nresources, over the last 300 second window.\n\n%smem Percentage of the time during which at least some tasks were waiting for memory\nresources, over the last time interval.\n\n%fmem-10\nPercentage of the time during which all non-idle tasks were stalled waiting for\nmemory resources, over the last 10 second window.\n\n%fmem-60\nPercentage of the time during which all non-idle tasks were stalled waiting for\nmemory resources, over the last 60 second window.\n\n%fmem-300\nPercentage of the time during which all non-idle tasks were stalled waiting for\nmemory resources, over the last 300 second window.\n\n%fmem Percentage of the time during which all non-idle tasks were stalled waiting for\nmemory resources, over the last time interval.\n\n\nThe PSI keyword is equivalent to specifying CPU, IO and MEM keywords together and\ntherefore all the pressure-stall statistics are reported.\n\nThe ALL keyword is equivalent to specifying all the keywords above and therefore all\nthe statistics are reported.\n" }, { "name": "-r [ ALL ]", "level": 2, "content": "Report memory utilization statistics. The ALL keyword indicates that all the memory\nfields should be displayed. The following values may be displayed:\n\nkbmemfree\nAmount of free memory available in kilobytes.\n\nkbavail\nEstimate of how much memory in kilobytes is available for starting new applica‐\ntions, without swapping. The estimate takes into account that the system needs\nsome page cache to function well, and that not all reclaimable slab will be re‐\nclaimable, due to items being in use. The impact of those factors will vary\nfrom system to system.\n\nkbmemused\nAmount of used memory in kilobytes (calculated as total installed memory - kb‐\nmemfree - kbbuffers - kbcached - kbslab).\n\n%memused\nPercentage of used memory.\n\nkbbuffers\nAmount of memory used as buffers by the kernel in kilobytes.\n\nkbcached\nAmount of memory used to cache data by the kernel in kilobytes.\n\nkbcommit\nAmount of memory in kilobytes needed for current workload. This is an estimate\nof how much RAM/swap is needed to guarantee that there never is out of memory.\n\n%commit\nPercentage of memory needed for current workload in relation to the total\namount of memory (RAM+swap). This number may be greater than 100% because the\nkernel usually overcommits memory.\n\nkbactive\nAmount of active memory in kilobytes (memory that has been used more recently\nand usually not reclaimed unless absolutely necessary).\n\nkbinact\nAmount of inactive memory in kilobytes (memory which has been less recently\nused. It is more eligible to be reclaimed for other purposes).\n\nkbdirty\nAmount of memory in kilobytes waiting to get written back to the disk.\n\nkbanonpg\nAmount of non-file backed pages in kilobytes mapped into userspace page tables.\n\nkbslab Amount of memory in kilobytes used by the kernel to cache data structures for\nits own use.\n\nkbkstack\nAmount of memory in kilobytes used for kernel stack space.\n\nkbpgtbl\nAmount of memory in kilobytes dedicated to the lowest level of page tables.\n\nkbvmused\nAmount of memory in kilobytes of used virtual address space.\n" }, { "name": "-S", "level": 2, "content": "kbswpfree\nAmount of free swap space in kilobytes.\n\nkbswpused\nAmount of used swap space in kilobytes.\n\n%swpused\nPercentage of used swap space.\n\nkbswpcad\nAmount of cached swap memory in kilobytes. This is memory that once was\nswapped out, is swapped back in but still also is in the swap area (if memory\nis needed it doesn't need to be swapped out again because it is already in the\nswap area. This saves I/O).\n\n%swpcad\nPercentage of cached swap memory in relation to the amount of used swap space.\n" }, { "name": "-s", "level": 2, "content": "Set the starting time of the data, causing the sar command to extract records\ntime-tagged at, or following, the time specified. The default starting time is\n08:00:00. Hours must be given in 24-hour format. This option can be used only when\ndata are read from a file (option -f).\n\n--sadc Indicate which data collector is called by sar. If the data collector is sought in\nPATH then enter \"which sadc\" to know where it is located.\n" }, { "name": "-t", "level": 2, "content": "stamps in the original local time of the data file creator. Without this option, the\nsar command displays the timestamps in the user's locale time.\n" }, { "name": "-u [ ALL ]", "level": 2, "content": "Report CPU utilization. The ALL keyword indicates that all the CPU fields should be\ndisplayed. The report may show the following fields:\n\n%user Percentage of CPU utilization that occurred while executing at the user level\n(application). Note that this field includes time spent running virtual proces‐\nsors.\n\n%usr Percentage of CPU utilization that occurred while executing at the user level\n(application). Note that this field does NOT include time spent running virtual\nprocessors.\n\n%nice Percentage of CPU utilization that occurred while executing at the user level\nwith nice priority.\n\n%system\nPercentage of CPU utilization that occurred while executing at the system level\n(kernel). Note that this field includes time spent servicing hardware and soft‐\nware interrupts.\n\n%sys Percentage of CPU utilization that occurred while executing at the system level\n(kernel). Note that this field does NOT include time spent servicing hardware\nor software interrupts.\n\n%iowait\nPercentage of time that the CPU or CPUs were idle during which the system had\nan outstanding disk I/O request.\n\n%steal Percentage of time spent in involuntary wait by the virtual CPU or CPUs while\nthe hypervisor was servicing another virtual processor.\n\n%irq Percentage of time spent by the CPU or CPUs to service hardware interrupts.\n\n%soft Percentage of time spent by the CPU or CPUs to service software interrupts.\n\n%guest Percentage of time spent by the CPU or CPUs to run a virtual processor.\n\n%gnice Percentage of time spent by the CPU or CPUs to run a niced guest.\n\n%idle Percentage of time that the CPU or CPUs were idle and the system did not have\nan outstanding disk I/O request.\n" }, { "name": "-V", "level": 2, "content": "" }, { "name": "-v", "level": 2, "content": "played:\n\ndentunusd\nNumber of unused cache entries in the directory cache.\n\nfile-nr\nNumber of file handles used by the system.\n\ninode-nr\nNumber of inode handlers used by the system.\n\npty-nr Number of pseudo-terminals used by the system.\n" }, { "name": "-W", "level": 2, "content": "pswpin/s\nTotal number of swap pages the system brought in per second.\n\npswpout/s\nTotal number of swap pages the system brought out per second.\n" }, { "name": "-w", "level": 2, "content": "played:\n\nproc/s Total number of tasks created per second.\n\ncswch/s\nTotal number of context switches per second.\n" }, { "name": "-y", "level": 2, "content": "rcvin/s\nNumber of receive interrupts per second for current serial line. Serial line\nnumber is given in the TTY column.\n\nxmtin/s\nNumber of transmit interrupts per second for current serial line.\n\nframerr/s\nNumber of frame errors per second for current serial line.\n\nprtyerr/s\nNumber of parity errors per second for current serial line.\n\nbrk/s Number of breaks per second for current serial line.\n\novrun/s\nNumber of overrun errors per second for current serial line.\n" }, { "name": "-z", "level": 2, "content": "sample period.\n\n" } ] }, { "name": "ENVIRONMENT", "level": 1, "content": "The sar command takes into account the following environment variables:\n\nSCOLORS\nBy default statistics are displayed in color when the output is connected to a termi‐\nnal. Use this variable to change the settings. Possible values for this variable are\nnever, always or auto (the latter is equivalent to the default settings).\nPlease note that the color (being red, yellow, or some other color) used to display a\nvalue is not indicative of any kind of issue simply because of the color. It only in‐\ndicates different ranges of values.\n\nSCOLORSSGR\nSpecify the colors and other attributes used to display statistics on the terminal.\nIts value is a colon-separated list of capabilities that defaults to\nC=33;22:H=31;1:I=32;22:M=35;1:N=34;1:R=31;22:Z=34;22. Supported capabilities are:\n\nC= SGR (Select Graphic Rendition) substring for comments inserted in the binary\ndaily data files.\n\nH= SGR substring for percentage values greater than or equal to 75%.\n\nI= SGR substring for item names or values (eg. network interfaces, CPU number...)\n\nM= SGR substring for percentage values in the range from 50% to 75%.\n\nN= SGR substring for non-zero statistics values.\n\nR= SGR substring for restart messages.\n\nZ= SGR substring for zero values.\n\nSTIMEDEFTIME\nIf this variable exists and its value is UTC then sar will save its data in UTC time\n(data will still be displayed in local time). sar will also use UTC time instead of\nlocal time to determine the current daily data file located in the /var/log/sysstat\ndirectory. This variable may be useful for servers with users located across several\ntimezones.\n\nSTIMEFORMAT\nIf this variable exists and its value is ISO then the current locale will be ignored\nwhen printing the date in the report header. The sar command will use the ISO 8601\nformat (YYYY-MM-DD) instead. The timestamp will also be compliant with ISO 8601 for‐\nmat.\n\n", "subsections": [] }, { "name": "EXAMPLES", "level": 1, "content": "", "subsections": [ { "name": "sar -u 2 5", "level": 2, "content": "Report CPU utilization for each 2 seconds. 5 lines are displayed.\n" }, { "name": "sar -I 14 -o int14.file 2 10", "level": 2, "content": "Report statistics on IRQ 14 for each 2 seconds. 10 lines are displayed. Data are\nstored in a file called int14.file.\n" }, { "name": "sar -r -n DEV -f /var/log/sysstat/sa16", "level": 2, "content": "Display memory and network statistics saved in daily data file sa16.\n\nsar -A Display all the statistics saved in current daily data file.\n\n" } ] }, { "name": "BUGS", "level": 1, "content": "/proc filesystem must be mounted for the sar command to work.\n\nAll the statistics are not necessarily available, depending on the kernel version used. sar\nassumes that you are using at least a 2.6 kernel.\n\nAlthough sar speaks of kilobytes (kB), megabytes (MB)..., it actually uses kibibytes (kiB),\nmebibytes (MiB)... A kibibyte is equal to 1024 bytes, and a mebibyte is equal to 1024\nkibibytes.\n\n", "subsections": [] }, { "name": "FILES", "level": 1, "content": "/var/log/sysstat/saDD\n/var/log/sysstat/saYYYYMMDD\nThe standard system activity daily data files and their default location. YYYY stands\nfor the current year, MM for the current month and DD for the current day.\n\n/proc and /sys contain various files with system statistics.\n\n", "subsections": [] }, { "name": "AUTHOR", "level": 1, "content": "Sebastien Godard (sysstat orange.fr)\n\n", "subsections": [ { "name": "SEE ALSO", "level": 2, "content": "sadc(8), sa1(8), sa2(8), sadf(1), sysstat(5), pidstat(1), mpstat(1), iostat(1), vmstat(8)\n\nhttps://github.com/sysstat/sysstat\nhttp://pagesperso-orange.fr/sebastien.godard/\n\n\n" } ] }, { "name": "Linux AUGUST 2020 SAR(1)", "level": 1, "content": "", "subsections": [] } ] }