| 1 |
.\" sar manual page - (C) 1999-2023 Sebastien Godard (sysstat <at> orange.fr) |
| 2 |
.TH SAR 1 "NOVEMBER 2023" Linux "Linux User's Manual" -*- nroff -*- |
| 3 |
.SH NAME |
| 4 |
sar \- Collect, report, or save system activity information. |
| 5 |
|
| 6 |
.SH SYNOPSIS |
| 7 |
.B sar [ \-A ] [ \-B ] [ \-b ] [ \-C ] [ \-D ] [ \-d ] [ \-F [ MOUNT ] ] [ \-H ] [ \-h ] [ \-p ] |
| 8 |
.B [ \-r [ ALL ] ] [ \-S ] [ \-t ] [ \-u [ ALL ] ] [ \-V ] [ \-v ] [ \-W ] [ \-w ] [ \-x ] [ \-y ] [ \-z ] |
| 9 |
.B [ \-\-dec={ 0 | 1 | 2 } ] |
| 10 |
.BI "[ \-\-dev=" "dev_list " "] [ \-\-fs=" "fs_list " "] [ \-\-help ] [ \-\-human ] [ \-\-iface=" "iface_list" |
| 11 |
.BI "] [ \-\-int=" "int_list " "] [ \-\-pretty ] [ \-\-sadc ] [ \-I [ SUM | ALL ] ] [ \-P { " "cpu_list" |
| 12 |
.B | ALL } ] [ \-m { |
| 13 |
.IB "keyword" "[,...] | ALL } ] [ \-n { " "keyword" "[,...] | ALL } ] [ \-q [ " "keyword" "[,...] | ALL ] ]" |
| 14 |
.B [ \-j { SID | ID | LABEL | PATH | UUID | ... } ] |
| 15 |
.BI "[ \-f [ " "filename " "] | \-o [ " "filename " "] | \-[0\-9]+ ]" |
| 16 |
.BI "[ \-i " "interval " "] [ \-s [ " "start_time " |
| 17 |
.BI "] ] [ \-e [ " "end_time " "] ] ] [ " "interval " "[ " "count " "] ]" |
| 18 |
|
| 19 |
.SH DESCRIPTION |
| 20 |
.RB "The " "sar" |
| 21 |
command writes to standard output the contents of selected |
| 22 |
cumulative activity counters in the operating system. The accounting |
| 23 |
system, based on the values in the |
| 24 |
.IR "count " "and " "interval" |
| 25 |
parameters, writes information the specified number of times spaced |
| 26 |
at the specified intervals in seconds. |
| 27 |
If the |
| 28 |
.IR "interval " "parameter is set to zero, the" |
| 29 |
.B sar |
| 30 |
command displays the average statistics for the time |
| 31 |
since the system was started. If the |
| 32 |
.IR "interval " "parameter is specified without the " "count" |
| 33 |
parameter, then reports are generated continuously. |
| 34 |
The collected data can also |
| 35 |
be saved in the file specified by the |
| 36 |
.BI "\-o " "filename" |
| 37 |
flag, in addition to being displayed onto the screen. If |
| 38 |
.IR "filename " "is omitted," |
| 39 |
.B sar |
| 40 |
uses the standard system activity daily data file (see below). |
| 41 |
By default all the data available from the kernel are saved in the |
| 42 |
data file. |
| 43 |
.PP |
| 44 |
.RB "The " "sar" |
| 45 |
command extracts and writes to standard output records previously |
| 46 |
saved in a file. This file can be either the one specified by the |
| 47 |
.B \-f |
| 48 |
flag or, by default, the standard system activity daily data file. |
| 49 |
It is also possible to enter |
| 50 |
.BR "\-1" ", " "\-2 " "etc. as an argument to " "sar" |
| 51 |
to display data of that days ago. For example, |
| 52 |
.BR "\-1 " "will point at the standard system activity file of yesterday." |
| 53 |
.PP |
| 54 |
Standard system activity daily data files are named |
| 55 |
.IR "saDD " "or " "saYYYYMMDD" ", where" |
| 56 |
.IR "YYYY " "stands for the current year, " "MM " "for the current month and " "DD" |
| 57 |
for the current day. They are the default files used by |
| 58 |
.B sar |
| 59 |
only when no filename has been explicitly specified. |
| 60 |
When used to write data to files (with its option |
| 61 |
.BR "\-o" "), " "sar" |
| 62 |
will use |
| 63 |
.I saYYYYMMDD |
| 64 |
.RB "if option " "\-D" |
| 65 |
has also been specified, else it will use |
| 66 |
.IR "saDD" "." |
| 67 |
When used to display the records previously saved in a file, |
| 68 |
.B sar |
| 69 |
will look for the most recent of |
| 70 |
.IR "saDD " "and " "saYYYYMMDD" ", and use it." |
| 71 |
.PP |
| 72 |
Standard system activity daily data files are located in the |
| 73 |
.I /var/log/sysstat |
| 74 |
directory by default. Yet it is possible to specify an alternate |
| 75 |
location for them: If a directory (instead of a plain file) is used |
| 76 |
with options |
| 77 |
.BR "\-f " "or " "\-o" |
| 78 |
then it will be considered as the directory containing the data files. |
| 79 |
.PP |
| 80 |
.RB "Without the " "\-P " "flag, the " "sar" |
| 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 |
| 84 |
.BR "\-P " "flag is given, the " "sar" |
| 85 |
command reports activity which relates to the specified processor or |
| 86 |
processors. If |
| 87 |
.BR "\-P ALL " "is given, the " "sar" |
| 88 |
command reports statistics for each individual processor and global |
| 89 |
statistics among all processors. Offline processors are not displayed. |
| 90 |
.PP |
| 91 |
You can select information about specific system activities using |
| 92 |
flags. Not specifying any flags selects only CPU activity. |
| 93 |
Specifying the |
| 94 |
.B \-A |
| 95 |
flag selects all possible activities. |
| 96 |
.PP |
| 97 |
The default version of the |
| 98 |
.B sar |
| 99 |
command (CPU utilization report) might be one of the first facilities |
| 100 |
the user runs to begin system activity investigation, because it |
| 101 |
monitors major system resources. If CPU utilization is near 100 percent |
| 102 |
(user + nice + system), the workload sampled is CPU-bound. |
| 103 |
|
| 104 |
If multiple samples and multiple reports are desired, it is convenient |
| 105 |
to specify an output file for the |
| 106 |
.BR "sar " "command. Run the " "sar" |
| 107 |
command as a background process. The syntax for this is: |
| 108 |
|
| 109 |
.BI "sar \-o " "datafile interval count " ">/dev/null 2>&1 &" |
| 110 |
|
| 111 |
All data are captured in binary form and saved to a file |
| 112 |
.RI "(" "datafile" ")." |
| 113 |
The data can then be selectively displayed with the |
| 114 |
.BR "sar " "command using the " "\-f " "option. Set the" |
| 115 |
.IR "interval " "and " "count " "parameters to select " "count " "records at " "interval" |
| 116 |
.RI "second intervals. If the " "count" |
| 117 |
parameter is not set, all the records saved in the file will be selected. |
| 118 |
Collection of data in this manner is useful to characterize |
| 119 |
system usage over a period of time and determine peak usage hours. |
| 120 |
.PP |
| 121 |
.RB "Note: The " "sar" |
| 122 |
command only reports on local activities. |
| 123 |
|
| 124 |
.SH OPTIONS |
| 125 |
.TP |
| 126 |
.B \-A |
| 127 |
This is equivalent to specifying |
| 128 |
.BR "\-bBdFHISvwWy \-m ALL \-n ALL \-q ALL \-r ALL \-u ALL" "." |
| 129 |
This option also implies specifying |
| 130 |
.B \-I ALL \-P ALL |
| 131 |
unless these options are explicitly set on the command line. |
| 132 |
.TP |
| 133 |
.B \-B |
| 134 |
Report paging statistics. |
| 135 |
The following values are displayed: |
| 136 |
.RS |
| 137 |
.IP pgpgin/s |
| 138 |
Total number of kilobytes the system paged in from disk per second. |
| 139 |
.IP pgpgout/s |
| 140 |
Total number of kilobytes the system paged out to disk per second. |
| 141 |
.IP fault/s |
| 142 |
Number of page faults (major + minor) made by the system per second. |
| 143 |
This is not a count of page faults that generate I/O, because some page |
| 144 |
faults can be resolved without I/O. |
| 145 |
.IP majflt/s |
| 146 |
Number of major faults the system has made per second, those which |
| 147 |
have required loading a memory page from disk. |
| 148 |
.IP pgfree/s |
| 149 |
Number of pages placed on the free list by the system per second. |
| 150 |
.IP pgscank/s |
| 151 |
Number of pages scanned by the kswapd daemon per second. |
| 152 |
.IP pgscand/s |
| 153 |
Number of pages scanned directly per second. |
| 154 |
.IP pgsteal/s |
| 155 |
Number of pages the system has reclaimed from cache (pagecache and |
| 156 |
swapcache) per second to satisfy its memory demands. |
| 157 |
.IP pgprom/s |
| 158 |
Number of pages promoted (i.e. migrated from slow to fast memory types) |
| 159 |
by the system per second. |
| 160 |
.IP pgdem/s |
| 161 |
Number of pages demoted (i.e. migrated from fast to slow memory types) |
| 162 |
by the system per second. |
| 163 |
.RE |
| 164 |
.TP |
| 165 |
.B \-b |
| 166 |
Report I/O and transfer rate statistics. The following values are displayed: |
| 167 |
.RS |
| 168 |
.IP tps |
| 169 |
Total number of transfers per second that were issued to physical devices. |
| 170 |
A transfer is an I/O request to a physical device. Multiple logical |
| 171 |
requests can be combined into a single I/O request to the device. |
| 172 |
A transfer is of indeterminate size. |
| 173 |
.IP rtps |
| 174 |
Total number of read requests per second issued to physical devices. |
| 175 |
.IP wtps |
| 176 |
Total number of write requests per second issued to physical devices. |
| 177 |
.IP dtps |
| 178 |
Total number of discard requests per second issued to physical devices. |
| 179 |
.IP bread/s |
| 180 |
Total amount of data read from the devices in blocks per second. |
| 181 |
Blocks are equivalent to sectors and therefore have a size of 512 bytes. |
| 182 |
.IP bwrtn/s |
| 183 |
Total amount of data written to devices in blocks per second. |
| 184 |
.IP bdscd/s |
| 185 |
Total amount of data discarded for devices in blocks per second. |
| 186 |
.RE |
| 187 |
.TP |
| 188 |
.B \-C |
| 189 |
When reading data from a file, tell |
| 190 |
.BR "sar " "to display comments that have been inserted by " "sadc" "." |
| 191 |
.TP |
| 192 |
.B \-D |
| 193 |
.RI "Use " "saYYYYMMDD " "instead of " "saDD" |
| 194 |
as the standard system activity daily data file name. This option |
| 195 |
works only when used in conjunction with option |
| 196 |
.BR "\-o " "to save data to file." |
| 197 |
.TP |
| 198 |
.B \-d |
| 199 |
Report activity for each block device. |
| 200 |
When data are displayed, the device name is displayed as it |
| 201 |
(should) appear in |
| 202 |
.IR "/dev" "." |
| 203 |
.BR "sar " "uses data in" |
| 204 |
.IR "/sys " "to determine the device name based on its major and minor numbers." |
| 205 |
.RB "If this name resolution fails, " "sar " "will use name mapping controlled by" |
| 206 |
.IR "/etc/sysstat/sysstat.ioconf " "file." |
| 207 |
Persistent device names can also be printed if option |
| 208 |
.B \-j |
| 209 |
is used (see below). Statistics for all devices are displayed unless |
| 210 |
a restricted list is specified using option |
| 211 |
.B \-\-dev= |
| 212 |
(see corresponding option entry). |
| 213 |
Note that disk activity depends on |
| 214 |
.BR "sadc" "'s options " "\-S DISK " "and " "\-S XDISK" |
| 215 |
to be collected. The following values are displayed: |
| 216 |
.RS |
| 217 |
.IP tps |
| 218 |
Total number of transfers per second that were issued to physical devices. |
| 219 |
A transfer is an I/O request to a physical device. Multiple logical |
| 220 |
requests can be combined into a single I/O request to the device. |
| 221 |
A transfer is of indeterminate size. |
| 222 |
.IP rkB/s |
| 223 |
Number of kilobytes read from the device per second. |
| 224 |
.IP wkB/s |
| 225 |
Number of kilobytes written to the device per second. |
| 226 |
.IP dkB/s |
| 227 |
Number of kilobytes discarded for the device per second. |
| 228 |
.IP areq\-sz |
| 229 |
The average size (in kilobytes) of the I/O requests that were issued to the device. |
| 230 |
.br |
| 231 |
Note: In previous versions, this field was known as avgrq\-sz and was expressed in sectors. |
| 232 |
.IP aqu\-sz |
| 233 |
The average queue length of the requests that were issued to the device. |
| 234 |
.br |
| 235 |
Note: In previous versions, this field was known as avgqu\-sz. |
| 236 |
.IP await |
| 237 |
The average time (in milliseconds) for I/O requests issued to the device |
| 238 |
to be served. This includes the time spent by the requests in queue and |
| 239 |
the time spent servicing them. |
| 240 |
.IP %util |
| 241 |
Percentage of elapsed time during which I/O requests were issued to the device |
| 242 |
(bandwidth utilization for the device). Device saturation occurs when this |
| 243 |
value is close to 100% for devices serving requests serially. But for |
| 244 |
devices serving requests in parallel, such as RAID arrays and modern SSDs, |
| 245 |
this number does not reflect their performance limits. |
| 246 |
.RE |
| 247 |
.TP |
| 248 |
.B \-\-dec={ 0 | 1 | 2 } |
| 249 |
Specify the number of decimal places to use (0 to 2, default value is 2). |
| 250 |
.TP |
| 251 |
.BI "\-\-dev=" "dev_list" |
| 252 |
Specify the block devices for which statistics are to be displayed by |
| 253 |
.BR "sar" "." |
| 254 |
.IR "dev_list " "is a list of comma-separated device names." |
| 255 |
.RE |
| 256 |
.PP |
| 257 |
.BI "\-e [ " "hh" ":" "mm" "[:" "ss" "] ]" |
| 258 |
.br |
| 259 |
.BI "\-e [ " "seconds_since_the_epoch " "]" |
| 260 |
.RS |
| 261 |
Set the ending time of the report. The default ending time is |
| 262 |
18:00:00. Hours must be given in 24-hour format, or as the number of seconds |
| 263 |
since the epoch (given as a 10 digit number). |
| 264 |
This option can be used when data are read from |
| 265 |
or written to a file (options |
| 266 |
.BR "\-f " "or " "\-o" ")." |
| 267 |
.RE |
| 268 |
.TP |
| 269 |
.B \-F [ MOUNT ] |
| 270 |
Display statistics for currently mounted filesystems. Pseudo-filesystems are |
| 271 |
ignored. At the end of the report, |
| 272 |
.B sar |
| 273 |
will display a summary of all those filesystems. Use of the |
| 274 |
.B MOUNT |
| 275 |
parameter keyword indicates that mountpoint will be reported instead of |
| 276 |
filesystem device. Statistics for all filesystems are displayed unless |
| 277 |
a restricted list is specified using option |
| 278 |
.B \-\-fs= |
| 279 |
(see corresponding option entry). |
| 280 |
Note that filesystems statistics depend on |
| 281 |
.BR "sadc" "'s option " "\-S XDISK " |
| 282 |
to be collected. |
| 283 |
|
| 284 |
The following values are displayed: |
| 285 |
.RS |
| 286 |
.IP MBfsfree |
| 287 |
Total amount of free space in megabytes (including space available only to privileged user). |
| 288 |
.IP MBfsused |
| 289 |
Total amount of space used in megabytes. |
| 290 |
.IP %fsused |
| 291 |
Percentage of filesystem space used, as seen by a privileged user. |
| 292 |
.IP %ufsused |
| 293 |
Percentage of filesystem space used, as seen by an unprivileged user. |
| 294 |
.IP Ifree |
| 295 |
Total number of free file nodes in filesystem. |
| 296 |
.IP Iused |
| 297 |
Total number of file nodes used in filesystem. |
| 298 |
.IP %Iused |
| 299 |
Percentage of file nodes used in filesystem. |
| 300 |
.RE |
| 301 |
.TP |
| 302 |
.BI "\-f [ " "filename " "]" |
| 303 |
.RI "Extract records from " "filename " "(created by the" |
| 304 |
.BI "\-o " "filename" |
| 305 |
flag). The default value of the |
| 306 |
.I filename |
| 307 |
parameter is the current standard system activity daily data file. If |
| 308 |
.I filename |
| 309 |
is a directory instead of a plain file then it is considered as the |
| 310 |
directory where the standard system activity daily data files are |
| 311 |
located. Option |
| 312 |
.BR "\-f " "is exclusive of option " "\-o" "." |
| 313 |
.TP |
| 314 |
.BI "\-\-fs=" "fs_list" |
| 315 |
Specify the filesystems for which statistics are to be displayed by |
| 316 |
.BR "sar" "." |
| 317 |
.I fs_list |
| 318 |
is a list of comma-separated filesystem names or mountpoints. |
| 319 |
.TP |
| 320 |
.B \-H |
| 321 |
Report hugepages utilization statistics. |
| 322 |
The following values are displayed: |
| 323 |
.RS |
| 324 |
.IP kbhugfree |
| 325 |
Amount of hugepages memory in kilobytes that is not yet allocated. |
| 326 |
.IP kbhugused |
| 327 |
Amount of hugepages memory in kilobytes that has been allocated. |
| 328 |
.IP %hugused |
| 329 |
Percentage of total hugepages memory that has been allocated. |
| 330 |
.IP kbhugrsvd |
| 331 |
Amount of reserved hugepages memory in kilobytes. |
| 332 |
.IP kbhugsurp |
| 333 |
Amount of surplus hugepages memory in kilobytes. |
| 334 |
.RE |
| 335 |
.TP |
| 336 |
.B \-h |
| 337 |
This option is equivalent to specifying |
| 338 |
.BR "\-\-pretty \-\-human" "." |
| 339 |
.TP |
| 340 |
.B \-\-help |
| 341 |
Display a short help message then exit. |
| 342 |
.TP |
| 343 |
.B \-\-human |
| 344 |
Print sizes in human readable format (e.g. 1.0k, 1.2M, etc.) |
| 345 |
The units displayed with this option supersede any other default units (e.g. |
| 346 |
kilobytes, sectors...) associated with the metrics. |
| 347 |
.TP |
| 348 |
.BI "\-I [ SUM | ALL ]" |
| 349 |
Report statistics for interrupts. The values displayed are the number of interrupts |
| 350 |
per second for the given processor or among all processors. |
| 351 |
A list of interrupts can be specified using |
| 352 |
.B \-\-int= |
| 353 |
(see this option). The |
| 354 |
.B SUM |
| 355 |
keyword indicates that the total number of interrupts received per second |
| 356 |
is to be displayed. The |
| 357 |
.B ALL |
| 358 |
keyword indicates that statistics from all interrupts are to be reported |
| 359 |
(this is the default). |
| 360 |
Note that interrupts statistics depend on |
| 361 |
.BR "sadc" "'s option " "\-S INT" |
| 362 |
to be collected. |
| 363 |
.TP |
| 364 |
.BI "\-i " "interval" |
| 365 |
Select data records at seconds as close as possible to the number specified |
| 366 |
.RI "by the " "interval " "parameter." |
| 367 |
.TP |
| 368 |
.BI "\-\-iface=" "iface_list" |
| 369 |
Specify the network interfaces for which statistics are to be displayed by |
| 370 |
.BR "sar" "." |
| 371 |
.I iface_list |
| 372 |
is a list of comma-separated interface names. |
| 373 |
.TP |
| 374 |
.BI "\-\-int=" "int_list" |
| 375 |
Specify the interrupts names for which statistics are to be displayed by |
| 376 |
.BR "sar" "." |
| 377 |
.I int_list |
| 378 |
is a list of comma-separated values or range of values (e.g., |
| 379 |
.BR "0\-16,35,40\-" "). |
| 380 |
.TP |
| 381 |
.B \-j { SID | ID | LABEL | PATH | UUID | ... } |
| 382 |
Display persistent device names. Use this option in conjunction with option |
| 383 |
.BR "\-d" ". Keywords " "ID" ", " "LABEL" "," |
| 384 |
etc. specify the type of the persistent name. These keywords are not limited, |
| 385 |
only prerequisite is that directory with required persistent names is present in |
| 386 |
.IR "/dev/disk" "." |
| 387 |
.RB "Keyword " "SID" |
| 388 |
tries to get a stable identifier to use as the device name. A stable |
| 389 |
identifier won't change across reboots for the same physical device. If it exists, |
| 390 |
this identifier is normally the WWN (World Wide Name) of the device, as read from the |
| 391 |
.IR "/dev/disk/by\-id " "directory." |
| 392 |
.TP |
| 393 |
.BI "\-m { " "keyword" "[,...] | ALL }" |
| 394 |
Report power management statistics. |
| 395 |
Note that these statistics depend on |
| 396 |
.BR "sadc" "'s option " "\-S POWER " "to be collected." |
| 397 |
|
| 398 |
Possible keywords are |
| 399 |
.BR "BAT" ", " "CPU" ", " "FAN" ", " "FREQ" ", " "IN" ", " "TEMP " "and " "USB" "." |
| 400 |
|
| 401 |
.RB "With the " "BAT" |
| 402 |
keyword, statistics about batteries capacity are reported. |
| 403 |
The following values are displayed: |
| 404 |
.RS |
| 405 |
.IP %cap |
| 406 |
Battery capacity. |
| 407 |
.IP cap/min |
| 408 |
Capacity lost or gained per minute by the battery. |
| 409 |
.IP status |
| 410 |
Charging status of the battery: ↑ (full), ↗ (charging), → (not charging), ↘ (discharging), ? (unknown). |
| 411 |
.RE |
| 412 |
|
| 413 |
.IP |
| 414 |
.RB "With the " "CPU" |
| 415 |
keyword, statistics about CPU are reported. |
| 416 |
The following value is displayed: |
| 417 |
.RS |
| 418 |
.IP MHz |
| 419 |
Instantaneous CPU clock frequency in MHz. |
| 420 |
.RE |
| 421 |
|
| 422 |
.IP |
| 423 |
.RB "With the " "FAN" |
| 424 |
keyword, statistics about fans speed are reported. |
| 425 |
The following values are displayed: |
| 426 |
.RS |
| 427 |
.IP rpm |
| 428 |
Fan speed expressed in revolutions per minute. |
| 429 |
.IP drpm |
| 430 |
This field is calculated as the difference between current fan speed (rpm) |
| 431 |
and its low limit (fan_min). |
| 432 |
.IP DEVICE |
| 433 |
Sensor device name. |
| 434 |
.RE |
| 435 |
|
| 436 |
.IP |
| 437 |
.RB "With the " "FREQ" |
| 438 |
keyword, statistics about CPU clock frequency are reported. |
| 439 |
The following value is displayed: |
| 440 |
.RS |
| 441 |
.IP wghMHz |
| 442 |
Weighted average CPU clock frequency in MHz. |
| 443 |
Note that the cpufreq-stats driver must be compiled in the |
| 444 |
kernel for this option to work. |
| 445 |
.RE |
| 446 |
|
| 447 |
.IP |
| 448 |
.RB "With the " "IN" |
| 449 |
keyword, statistics about voltage inputs are reported. |
| 450 |
The following values are displayed: |
| 451 |
.RS |
| 452 |
.IP inV |
| 453 |
Voltage input expressed in Volts. |
| 454 |
.IP %in |
| 455 |
Relative input value. A value of 100% means that |
| 456 |
voltage input has reached its high limit (in_max) whereas |
| 457 |
a value of 0% means that it has reached its low limit (in_min). |
| 458 |
.IP DEVICE |
| 459 |
Sensor device name. |
| 460 |
.RE |
| 461 |
|
| 462 |
.IP |
| 463 |
.RB "With the " "TEMP" |
| 464 |
keyword, statistics about devices temperature are reported. |
| 465 |
The following values are displayed: |
| 466 |
.RS |
| 467 |
.IP degC |
| 468 |
Device temperature expressed in degrees Celsius. |
| 469 |
.IP %temp |
| 470 |
Relative device temperature. A value of 100% means that |
| 471 |
temperature has reached its high limit (temp_max). |
| 472 |
.IP DEVICE |
| 473 |
Sensor device name. |
| 474 |
.RE |
| 475 |
|
| 476 |
.IP |
| 477 |
.RB "With the " "USB " "keyword, the " "sar" |
| 478 |
command takes a snapshot of all the USB devices currently plugged into |
| 479 |
the system. At the end of the report, |
| 480 |
.B sar |
| 481 |
will display a summary of all those USB devices. |
| 482 |
The following values are displayed: |
| 483 |
.RS |
| 484 |
.IP BUS |
| 485 |
Root hub number of the USB device. |
| 486 |
.IP idvendor |
| 487 |
Vendor ID number (assigned by USB organization). |
| 488 |
.IP idprod |
| 489 |
Product ID number (assigned by Manufacturer). |
| 490 |
.IP maxpower |
| 491 |
Maximum power consumption of the device (expressed in mA). |
| 492 |
.IP manufact |
| 493 |
Manufacturer name. |
| 494 |
.IP product |
| 495 |
Product name. |
| 496 |
.RE |
| 497 |
|
| 498 |
.IP |
| 499 |
.RB "The " "ALL" |
| 500 |
keyword is equivalent to specifying all the keywords above and therefore all the power |
| 501 |
management statistics are reported. |
| 502 |
.TP |
| 503 |
.BI "\-n { " "keyword" "[,...] | ALL }" |
| 504 |
Report network statistics. |
| 505 |
|
| 506 |
Possible keywords are |
| 507 |
.BR "DEV" ", " "EDEV" ", " "FC" ", " "ICMP" ", " "EICMP" ", " "ICMP6" ", " "EICMP6" "," |
| 508 |
.BR "IP" ", " "EIP" ", " "IP6" ", " "EIP6" ", " "NFS" ", " "NFSD" ", " "SOCK" ", " "SOCK6" "," |
| 509 |
.BR "SOFT" ", " "TCP" ", " "ETCP" ", " "UDP " "and " "UDP6" "." |
| 510 |
|
| 511 |
.RB "With the " "DEV" |
| 512 |
keyword, statistics from the network devices are reported. |
| 513 |
Statistics for all network interfaces are displayed unless |
| 514 |
a restricted list is specified using option |
| 515 |
.B \-\-iface= |
| 516 |
(see corresponding option entry). |
| 517 |
The following values are displayed: |
| 518 |
.RS |
| 519 |
.IP IFACE |
| 520 |
Name of the network interface for which statistics are reported. |
| 521 |
.IP rxpck/s |
| 522 |
Total number of packets received per second. |
| 523 |
.IP txpck/s |
| 524 |
Total number of packets transmitted per second. |
| 525 |
.IP rxkB/s |
| 526 |
Total number of kilobytes received per second. |
| 527 |
.IP txkB/s |
| 528 |
Total number of kilobytes transmitted per second. |
| 529 |
.IP rxcmp/s |
| 530 |
Number of compressed packets received per second (for cslip etc.). |
| 531 |
.IP txcmp/s |
| 532 |
Number of compressed packets transmitted per second. |
| 533 |
.IP rxmcst/s |
| 534 |
Number of multicast packets received per second. |
| 535 |
.IP %ifutil |
| 536 |
Utilization percentage of the network interface. For half-duplex interfaces, |
| 537 |
utilization is calculated using the sum of rxkB/s and txkB/s as a percentage |
| 538 |
of the interface speed. For full-duplex, this is the greater of rxkB/S or txkB/s. |
| 539 |
.RE |
| 540 |
|
| 541 |
.IP |
| 542 |
.RB "With the " "EDEV" |
| 543 |
keyword, statistics on failures (errors) from the network devices are reported. |
| 544 |
Statistics for all network interfaces are displayed unless |
| 545 |
a restricted list is specified using option |
| 546 |
.B \-\-iface= |
| 547 |
(see corresponding option entry). |
| 548 |
The following values are displayed: |
| 549 |
.RS |
| 550 |
.IP IFACE |
| 551 |
Name of the network interface for which statistics are reported. |
| 552 |
.IP rxerr/s |
| 553 |
Total number of bad packets received per second. |
| 554 |
.IP txerr/s |
| 555 |
Total number of errors that happened per second while transmitting packets. |
| 556 |
.IP coll/s |
| 557 |
Number of collisions that happened per second while transmitting packets. |
| 558 |
.IP rxdrop/s |
| 559 |
Number of received packets dropped per second because of a lack of space in linux buffers. |
| 560 |
.IP txdrop/s |
| 561 |
Number of transmitted packets dropped per second because of a lack of space in linux buffers. |
| 562 |
.IP txcarr/s |
| 563 |
Number of carrier-errors that happened per second while transmitting packets. |
| 564 |
.IP rxfram/s |
| 565 |
Number of frame alignment errors that happened per second on received packets. |
| 566 |
.IP rxfifo/s |
| 567 |
Number of FIFO overrun errors that happened per second on received packets. |
| 568 |
.IP txfifo/s |
| 569 |
Number of FIFO overrun errors that happened per second on transmitted packets. |
| 570 |
.RE |
| 571 |
|
| 572 |
.IP |
| 573 |
.RB "With the " "FC" |
| 574 |
keyword, statistics about fibre channel traffic are reported. |
| 575 |
Note that fibre channel statistics depend on |
| 576 |
.BR "sadc" "'s option " "\-S DISK" |
| 577 |
to be collected. |
| 578 |
The following values are displayed: |
| 579 |
.RS |
| 580 |
.IP FCHOST |
| 581 |
Name of the fibre channel host bus adapter (HBA) interface for which statistics are reported. |
| 582 |
.IP fch_rxf/s |
| 583 |
The total number of frames received per second. |
| 584 |
.IP fch_txf/s |
| 585 |
The total number of frames transmitted per second. |
| 586 |
.IP fch_rxw/s |
| 587 |
The total number of transmission words received per second. |
| 588 |
.IP fch_txw/s |
| 589 |
The total number of transmission words transmitted per second. |
| 590 |
.RE |
| 591 |
|
| 592 |
.IP |
| 593 |
.RB "With the " "ICMP" |
| 594 |
keyword, statistics about ICMPv4 network traffic are reported. |
| 595 |
Note that ICMPv4 statistics depend on |
| 596 |
.BR "sadc" "'s option " "\-S SNMP" |
| 597 |
to be collected. |
| 598 |
The following values are displayed (formal SNMP names between |
| 599 |
square brackets): |
| 600 |
.RS |
| 601 |
.IP imsg/s |
| 602 |
The total number of ICMP messages which the entity |
| 603 |
received per second [icmpInMsgs]. |
| 604 |
Note that this counter includes all those counted by ierr/s. |
| 605 |
.IP omsg/s |
| 606 |
The total number of ICMP messages which this entity |
| 607 |
attempted to send per second [icmpOutMsgs]. |
| 608 |
Note that this counter includes all those counted by oerr/s. |
| 609 |
.IP iech/s |
| 610 |
The number of ICMP Echo (request) messages received per second [icmpInEchos]. |
| 611 |
.IP iechr/s |
| 612 |
The number of ICMP Echo Reply messages received per second [icmpInEchoReps]. |
| 613 |
.IP oech/s |
| 614 |
The number of ICMP Echo (request) messages sent per second [icmpOutEchos]. |
| 615 |
.IP oechr/s |
| 616 |
The number of ICMP Echo Reply messages sent per second [icmpOutEchoReps]. |
| 617 |
.IP itm/s |
| 618 |
The number of ICMP Timestamp (request) messages received per second [icmpInTimestamps]. |
| 619 |
.IP itmr/s |
| 620 |
The number of ICMP Timestamp Reply messages received per second [icmpInTimestampReps]. |
| 621 |
.IP otm/s |
| 622 |
The number of ICMP Timestamp (request) messages sent per second [icmpOutTimestamps]. |
| 623 |
.IP otmr/s |
| 624 |
The number of ICMP Timestamp Reply messages sent per second [icmpOutTimestampReps]. |
| 625 |
.IP iadrmk/s |
| 626 |
The number of ICMP Address Mask Request messages received per second [icmpInAddrMasks]. |
| 627 |
.IP iadrmkr/s |
| 628 |
The number of ICMP Address Mask Reply messages received per second [icmpInAddrMaskReps]. |
| 629 |
.IP oadrmk/s |
| 630 |
The number of ICMP Address Mask Request messages sent per second [icmpOutAddrMasks]. |
| 631 |
.IP oadrmkr/s |
| 632 |
The number of ICMP Address Mask Reply messages sent per second [icmpOutAddrMaskReps]. |
| 633 |
.RE |
| 634 |
|
| 635 |
.IP |
| 636 |
.RB "With the " "EICMP" |
| 637 |
keyword, statistics about ICMPv4 error messages are reported. |
| 638 |
Note that ICMPv4 statistics depend on |
| 639 |
.BR "sadc" "'s option " "\-S SNMP" |
| 640 |
to be collected. |
| 641 |
The following values are displayed (formal SNMP names between |
| 642 |
square brackets): |
| 643 |
.RS |
| 644 |
.IP ierr/s |
| 645 |
The number of ICMP messages per second which the entity received but |
| 646 |
determined as having ICMP-specific errors (bad ICMP |
| 647 |
checksums, bad length, etc.) [icmpInErrors]. |
| 648 |
.IP oerr/s |
| 649 |
The number of ICMP messages per second which this entity did not send |
| 650 |
due to problems discovered within ICMP such as a lack of buffers [icmpOutErrors]. |
| 651 |
.IP idstunr/s |
| 652 |
The number of ICMP Destination Unreachable messages |
| 653 |
received per second [icmpInDestUnreachs]. |
| 654 |
.IP odstunr/s |
| 655 |
The number of ICMP Destination Unreachable messages sent per second [icmpOutDestUnreachs]. |
| 656 |
.IP itmex/s |
| 657 |
The number of ICMP Time Exceeded messages received per second [icmpInTimeExcds]. |
| 658 |
.IP otmex/s |
| 659 |
The number of ICMP Time Exceeded messages sent per second [icmpOutTimeExcds]. |
| 660 |
.IP iparmpb/s |
| 661 |
The number of ICMP Parameter Problem messages received per second [icmpInParmProbs]. |
| 662 |
.IP oparmpb/s |
| 663 |
The number of ICMP Parameter Problem messages sent per second [icmpOutParmProbs]. |
| 664 |
.IP isrcq/s |
| 665 |
The number of ICMP Source Quench messages received per second [icmpInSrcQuenchs]. |
| 666 |
.IP osrcq/s |
| 667 |
The number of ICMP Source Quench messages sent per second [icmpOutSrcQuenchs]. |
| 668 |
.IP iredir/s |
| 669 |
The number of ICMP Redirect messages received per second [icmpInRedirects]. |
| 670 |
.IP oredir/s |
| 671 |
The number of ICMP Redirect messages sent per second [icmpOutRedirects]. |
| 672 |
.RE |
| 673 |
|
| 674 |
.IP |
| 675 |
.RB "With the " "ICMP6" |
| 676 |
keyword, statistics about ICMPv6 network traffic are reported. |
| 677 |
Note that ICMPv6 statistics depend on |
| 678 |
.BR "sadc" "'s option " "\-S IPV6" |
| 679 |
to be collected. |
| 680 |
The following values are displayed (formal SNMP names between |
| 681 |
square brackets): |
| 682 |
.RS |
| 683 |
.IP imsg6/s |
| 684 |
The total number of ICMP messages received |
| 685 |
by the interface per second which includes all those |
| 686 |
counted by ierr6/s [ipv6IfIcmpInMsgs]. |
| 687 |
.IP omsg6/s |
| 688 |
The total number of ICMP messages which this |
| 689 |
interface attempted to send per second [ipv6IfIcmpOutMsgs]. |
| 690 |
.IP iech6/s |
| 691 |
The number of ICMP Echo (request) messages |
| 692 |
received by the interface per second [ipv6IfIcmpInEchos]. |
| 693 |
.IP iechr6/s |
| 694 |
The number of ICMP Echo Reply messages received |
| 695 |
by the interface per second [ipv6IfIcmpInEchoReplies]. |
| 696 |
.IP oechr6/s |
| 697 |
The number of ICMP Echo Reply messages sent |
| 698 |
by the interface per second [ipv6IfIcmpOutEchoReplies]. |
| 699 |
.IP igmbq6/s |
| 700 |
The number of ICMPv6 Group Membership Query |
| 701 |
messages received by the interface per second |
| 702 |
[ipv6IfIcmpInGroupMembQueries]. |
| 703 |
.IP igmbr6/s |
| 704 |
The number of ICMPv6 Group Membership Response messages |
| 705 |
received by the interface per second |
| 706 |
[ipv6IfIcmpInGroupMembResponses]. |
| 707 |
.IP ogmbr6/s |
| 708 |
The number of ICMPv6 Group Membership Response |
| 709 |
messages sent per second |
| 710 |
[ipv6IfIcmpOutGroupMembResponses]. |
| 711 |
.IP igmbrd6/s |
| 712 |
The number of ICMPv6 Group Membership Reduction messages |
| 713 |
received by the interface per second |
| 714 |
[ipv6IfIcmpInGroupMembReductions]. |
| 715 |
.IP ogmbrd6/s |
| 716 |
The number of ICMPv6 Group Membership Reduction |
| 717 |
messages sent per second |
| 718 |
[ipv6IfIcmpOutGroupMembReductions]. |
| 719 |
.IP irtsol6/s |
| 720 |
The number of ICMP Router Solicit messages |
| 721 |
received by the interface per second |
| 722 |
[ipv6IfIcmpInRouterSolicits]. |
| 723 |
.IP ortsol6/s |
| 724 |
The number of ICMP Router Solicitation messages |
| 725 |
sent by the interface per second |
| 726 |
[ipv6IfIcmpOutRouterSolicits]. |
| 727 |
.IP irtad6/s |
| 728 |
The number of ICMP Router Advertisement messages |
| 729 |
received by the interface per second |
| 730 |
[ipv6IfIcmpInRouterAdvertisements]. |
| 731 |
.IP inbsol6/s |
| 732 |
The number of ICMP Neighbor Solicit messages |
| 733 |
received by the interface per second |
| 734 |
[ipv6IfIcmpInNeighborSolicits]. |
| 735 |
.IP onbsol6/s |
| 736 |
The number of ICMP Neighbor Solicitation |
| 737 |
messages sent by the interface per second |
| 738 |
[ipv6IfIcmpOutNeighborSolicits]. |
| 739 |
.IP inbad6/s |
| 740 |
The number of ICMP Neighbor Advertisement |
| 741 |
messages received by the interface per second |
| 742 |
[ipv6IfIcmpInNeighborAdvertisements]. |
| 743 |
.IP onbad6/s |
| 744 |
The number of ICMP Neighbor Advertisement |
| 745 |
messages sent by the interface per second |
| 746 |
[ipv6IfIcmpOutNeighborAdvertisements]. |
| 747 |
.RE |
| 748 |
|
| 749 |
.IP |
| 750 |
.RB "With the " "EICMP6" |
| 751 |
keyword, statistics about ICMPv6 error messages are reported. |
| 752 |
Note that ICMPv6 statistics depend on |
| 753 |
.BR "sadc" "'s option " "\-S IPV6" |
| 754 |
to be collected. |
| 755 |
The following values are displayed (formal SNMP names between |
| 756 |
square brackets): |
| 757 |
.RS |
| 758 |
.IP ierr6/s |
| 759 |
The number of ICMP messages per second which the interface |
| 760 |
received but determined as having ICMP-specific |
| 761 |
errors (bad ICMP checksums, bad length, etc.) |
| 762 |
[ipv6IfIcmpInErrors] |
| 763 |
.IP idtunr6/s |
| 764 |
The number of ICMP Destination Unreachable |
| 765 |
messages received by the interface per second |
| 766 |
[ipv6IfIcmpInDestUnreachs]. |
| 767 |
.IP odtunr6/s |
| 768 |
The number of ICMP Destination Unreachable |
| 769 |
messages sent by the interface per second |
| 770 |
[ipv6IfIcmpOutDestUnreachs]. |
| 771 |
.IP itmex6/s |
| 772 |
The number of ICMP Time Exceeded messages |
| 773 |
received by the interface per second |
| 774 |
[ipv6IfIcmpInTimeExcds]. |
| 775 |
.IP otmex6/s |
| 776 |
The number of ICMP Time Exceeded messages sent |
| 777 |
by the interface per second |
| 778 |
[ipv6IfIcmpOutTimeExcds]. |
| 779 |
.IP iprmpb6/s |
| 780 |
The number of ICMP Parameter Problem messages |
| 781 |
received by the interface per second |
| 782 |
[ipv6IfIcmpInParmProblems]. |
| 783 |
.IP oprmpb6/s |
| 784 |
The number of ICMP Parameter Problem messages |
| 785 |
sent by the interface per second |
| 786 |
[ipv6IfIcmpOutParmProblems]. |
| 787 |
.IP iredir6/s |
| 788 |
The number of Redirect messages received |
| 789 |
by the interface per second |
| 790 |
[ipv6IfIcmpInRedirects]. |
| 791 |
.IP oredir6/s |
| 792 |
The number of Redirect messages sent by |
| 793 |
the interface by second |
| 794 |
[ipv6IfIcmpOutRedirects]. |
| 795 |
.IP ipck2b6/s |
| 796 |
The number of ICMP Packet Too Big messages |
| 797 |
received by the interface per second |
| 798 |
[ipv6IfIcmpInPktTooBigs]. |
| 799 |
.IP opck2b6/s |
| 800 |
The number of ICMP Packet Too Big messages sent |
| 801 |
by the interface per second |
| 802 |
[ipv6IfIcmpOutPktTooBigs]. |
| 803 |
.RE |
| 804 |
|
| 805 |
.IP |
| 806 |
.RB "With the " "IP" |
| 807 |
keyword, statistics about IPv4 network traffic are reported. |
| 808 |
Note that IPv4 statistics depend on |
| 809 |
.BR "sadc" "'s option " "\-S SNMP" |
| 810 |
to be collected. |
| 811 |
The following values are displayed (formal SNMP names between |
| 812 |
square brackets): |
| 813 |
.RS |
| 814 |
.IP irec/s |
| 815 |
The total number of input datagrams received from interfaces |
| 816 |
per second, including those received in error [ipInReceives]. |
| 817 |
.IP fwddgm/s |
| 818 |
The number of input datagrams per second, for which this entity was not |
| 819 |
their final IP destination, as a result of which an attempt |
| 820 |
was made to find a route to forward them to that final |
| 821 |
destination [ipForwDatagrams]. |
| 822 |
.IP idel/s |
| 823 |
The total number of input datagrams successfully delivered per second |
| 824 |
to IP user-protocols (including ICMP) [ipInDelivers]. |
| 825 |
.IP orq/s |
| 826 |
The total number of IP datagrams which local IP user-protocols (including ICMP) |
| 827 |
supplied per second to IP in requests for transmission [ipOutRequests]. |
| 828 |
Note that this counter does not include any datagrams counted in fwddgm/s. |
| 829 |
.IP asmrq/s |
| 830 |
The number of IP fragments received per second which needed to be |
| 831 |
reassembled at this entity [ipReasmReqds]. |
| 832 |
.IP asmok/s |
| 833 |
The number of IP datagrams successfully re-assembled per second [ipReasmOKs]. |
| 834 |
.IP fragok/s |
| 835 |
The number of IP datagrams that have been successfully |
| 836 |
fragmented at this entity per second [ipFragOKs]. |
| 837 |
.IP fragcrt/s |
| 838 |
The number of IP datagram fragments that have been |
| 839 |
generated per second as a result of fragmentation at this entity [ipFragCreates]. |
| 840 |
.RE |
| 841 |
|
| 842 |
.IP |
| 843 |
.RB "With the " "EIP" |
| 844 |
keyword, statistics about IPv4 network errors are reported. |
| 845 |
Note that IPv4 statistics depend on |
| 846 |
.BR "sadc" "'s option " "\-S SNMP" |
| 847 |
to be collected. |
| 848 |
The following values are displayed (formal SNMP names between |
| 849 |
square brackets): |
| 850 |
.RS |
| 851 |
.IP ihdrerr/s |
| 852 |
The number of input datagrams discarded per second due to errors in |
| 853 |
their IP headers, including bad checksums, version number |
| 854 |
mismatch, other format errors, time-to-live exceeded, errors |
| 855 |
discovered in processing their IP options, etc. [ipInHdrErrors] |
| 856 |
.IP iadrerr/s |
| 857 |
The number of input datagrams discarded per second because the IP |
| 858 |
address in their IP header's destination field was not a |
| 859 |
valid address to be received at this entity. This count |
| 860 |
includes invalid addresses (e.g., 0.0.0.0) and addresses of |
| 861 |
unsupported Classes (e.g., Class E). For entities which are |
| 862 |
not IP routers and therefore do not forward datagrams, this |
| 863 |
counter includes datagrams discarded because the destination |
| 864 |
address was not a local address [ipInAddrErrors]. |
| 865 |
.IP iukwnpr/s |
| 866 |
The number of locally-addressed datagrams received |
| 867 |
successfully but discarded per second because of an unknown or |
| 868 |
unsupported protocol [ipInUnknownProtos]. |
| 869 |
.IP idisc/s |
| 870 |
The number of input IP datagrams per second for which no problems were |
| 871 |
encountered to prevent their continued processing, but which |
| 872 |
were discarded (e.g., for lack of buffer space) [ipInDiscards]. |
| 873 |
Note that this counter does not include any datagrams discarded while |
| 874 |
awaiting re-assembly. |
| 875 |
.IP odisc/s |
| 876 |
The number of output IP datagrams per second for which no problem was |
| 877 |
encountered to prevent their transmission to their |
| 878 |
destination, but which were discarded (e.g., for lack of |
| 879 |
buffer space) [ipOutDiscards]. |
| 880 |
Note that this counter would include |
| 881 |
datagrams counted in fwddgm/s if any such packets met |
| 882 |
this (discretionary) discard criterion. |
| 883 |
.IP onort/s |
| 884 |
The number of IP datagrams discarded per second because no route could |
| 885 |
be found to transmit them to their destination [ipOutNoRoutes]. |
| 886 |
Note that this counter includes any packets counted in fwddgm/s |
| 887 |
which meet this 'no-route' criterion. |
| 888 |
Note that this includes any datagrams which a host cannot route because all |
| 889 |
of its default routers are down. |
| 890 |
.IP asmf/s |
| 891 |
The number of failures detected per second by the IP re-assembly |
| 892 |
algorithm (for whatever reason: timed out, errors, etc) [ipReasmFails]. |
| 893 |
Note that this is not necessarily a count of discarded IP |
| 894 |
fragments since some algorithms can lose track of the number of |
| 895 |
fragments by combining them as they are received. |
| 896 |
.IP fragf/s |
| 897 |
The number of IP datagrams that have been discarded per second because |
| 898 |
they needed to be fragmented at this entity but could not |
| 899 |
be, e.g., because their Don't Fragment flag was set [ipFragFails]. |
| 900 |
.RE |
| 901 |
|
| 902 |
.IP |
| 903 |
.RB "With the " "IP6" |
| 904 |
keyword, statistics about IPv6 network traffic are reported. |
| 905 |
Note that IPv6 statistics depend on |
| 906 |
.BR "sadc" "'s option " "\-S IPV6" |
| 907 |
to be collected. |
| 908 |
The following values are displayed (formal SNMP names between |
| 909 |
square brackets): |
| 910 |
.RS |
| 911 |
.IP irec6/s |
| 912 |
The total number of input datagrams received from |
| 913 |
interfaces per second, including those received in error |
| 914 |
[ipv6IfStatsInReceives]. |
| 915 |
.IP fwddgm6/s |
| 916 |
The number of output datagrams per second which this |
| 917 |
entity received and forwarded to their final |
| 918 |
destinations [ipv6IfStatsOutForwDatagrams]. |
| 919 |
.IP idel6/s |
| 920 |
The total number of datagrams successfully |
| 921 |
delivered per second to IPv6 user-protocols (including ICMP) |
| 922 |
[ipv6IfStatsInDelivers]. |
| 923 |
.IP orq6/s |
| 924 |
The total number of IPv6 datagrams which local IPv6 |
| 925 |
user-protocols (including ICMP) supplied per second to IPv6 in |
| 926 |
requests for transmission [ipv6IfStatsOutRequests]. |
| 927 |
Note that this counter |
| 928 |
does not include any datagrams counted in fwddgm6/s. |
| 929 |
.IP asmrq6/s |
| 930 |
The number of IPv6 fragments received per second which needed |
| 931 |
to be reassembled at this interface [ipv6IfStatsReasmReqds]. |
| 932 |
.IP asmok6/s |
| 933 |
The number of IPv6 datagrams successfully |
| 934 |
reassembled per second [ipv6IfStatsReasmOKs]. |
| 935 |
.IP imcpck6/s |
| 936 |
The number of multicast packets received per second |
| 937 |
by the interface [ipv6IfStatsInMcastPkts]. |
| 938 |
.IP omcpck6/s |
| 939 |
The number of multicast packets transmitted per second |
| 940 |
by the interface [ipv6IfStatsOutMcastPkts]. |
| 941 |
.IP fragok6/s |
| 942 |
The number of IPv6 datagrams that have been |
| 943 |
successfully fragmented at this output interface per second |
| 944 |
[ipv6IfStatsOutFragOKs]. |
| 945 |
.IP fragcr6/s |
| 946 |
The number of output datagram fragments that have |
| 947 |
been generated per second as a result of fragmentation at |
| 948 |
this output interface [ipv6IfStatsOutFragCreates]. |
| 949 |
.RE |
| 950 |
|
| 951 |
.IP |
| 952 |
.RB "With the " "EIP6" |
| 953 |
keyword, statistics about IPv6 network errors are reported. |
| 954 |
Note that IPv6 statistics depend on |
| 955 |
.BR "sadc" "'s option " "\-S IPV6" |
| 956 |
to be collected. |
| 957 |
The following values are displayed (formal SNMP names between |
| 958 |
square brackets): |
| 959 |
.RS |
| 960 |
.IP ihdrer6/s |
| 961 |
The number of input datagrams discarded per second due to |
| 962 |
errors in their IPv6 headers, including version |
| 963 |
number mismatch, other format errors, hop count |
| 964 |
exceeded, errors discovered in processing their |
| 965 |
IPv6 options, etc. [ipv6IfStatsInHdrErrors] |
| 966 |
.IP iadrer6/s |
| 967 |
The number of input datagrams discarded per second because |
| 968 |
the IPv6 address in their IPv6 header's destination |
| 969 |
field was not a valid address to be received at |
| 970 |
this entity. This count includes invalid |
| 971 |
addresses (e.g., ::0) and unsupported addresses |
| 972 |
(e.g., addresses with unallocated prefixes). For |
| 973 |
entities which are not IPv6 routers and therefore |
| 974 |
do not forward datagrams, this counter includes |
| 975 |
datagrams discarded because the destination address |
| 976 |
was not a local address [ipv6IfStatsInAddrErrors]. |
| 977 |
.IP iukwnp6/s |
| 978 |
The number of locally-addressed datagrams |
| 979 |
received successfully but discarded per second because of an |
| 980 |
unknown or unsupported protocol [ipv6IfStatsInUnknownProtos]. |
| 981 |
.IP i2big6/s |
| 982 |
The number of input datagrams that could not be |
| 983 |
forwarded per second because their size exceeded the link MTU |
| 984 |
of outgoing interface [ipv6IfStatsInTooBigErrors]. |
| 985 |
.IP idisc6/s |
| 986 |
The number of input IPv6 datagrams per second for which no |
| 987 |
problems were encountered to prevent their |
| 988 |
continued processing, but which were discarded |
| 989 |
(e.g., for lack of buffer space) |
| 990 |
[ipv6IfStatsInDiscards]. Note that this |
| 991 |
counter does not include any datagrams discarded |
| 992 |
while awaiting re-assembly. |
| 993 |
.IP odisc6/s |
| 994 |
The number of output IPv6 datagrams per second for which no |
| 995 |
problem was encountered to prevent their |
| 996 |
transmission to their destination, but which were |
| 997 |
discarded (e.g., for lack of buffer space) |
| 998 |
[ipv6IfStatsOutDiscards]. Note |
| 999 |
that this counter would include datagrams counted |
| 1000 |
in fwddgm6/s if any such packets |
| 1001 |
met this (discretionary) discard criterion. |
| 1002 |
.IP inort6/s |
| 1003 |
The number of input datagrams discarded per second because no |
| 1004 |
route could be found to transmit them to their |
| 1005 |
destination [ipv6IfStatsInNoRoutes]. |
| 1006 |
.IP onort6/s |
| 1007 |
The number of locally generated IP datagrams discarded per second |
| 1008 |
because no route could be found to transmit them to their |
| 1009 |
destination [unknown formal SNMP name]. |
| 1010 |
.IP asmf6/s |
| 1011 |
The number of failures detected per second by the IPv6 |
| 1012 |
re-assembly algorithm (for whatever reason: timed |
| 1013 |
out, errors, etc.) [ipv6IfStatsReasmFails]. |
| 1014 |
Note that this is not necessarily a count of discarded |
| 1015 |
IPv6 fragments since some algorithms |
| 1016 |
can lose track of the number of fragments |
| 1017 |
by combining them as they are received. |
| 1018 |
.IP fragf6/s |
| 1019 |
The number of IPv6 datagrams that have been |
| 1020 |
discarded per second because they needed to be fragmented |
| 1021 |
at this output interface but could not be |
| 1022 |
[ipv6IfStatsOutFragFails]. |
| 1023 |
.IP itrpck6/s |
| 1024 |
The number of input datagrams discarded per second because |
| 1025 |
datagram frame didn't carry enough data |
| 1026 |
[ipv6IfStatsInTruncatedPkts]. |
| 1027 |
.RE |
| 1028 |
|
| 1029 |
.IP |
| 1030 |
.RB "With the " "NFS" |
| 1031 |
keyword, statistics about NFS client activity are reported. |
| 1032 |
The following values are displayed: |
| 1033 |
.RS |
| 1034 |
.IP call/s |
| 1035 |
Number of RPC requests made per second. |
| 1036 |
.IP retrans/s |
| 1037 |
Number of RPC requests per second, those which needed to be retransmitted |
| 1038 |
(for example because of a server timeout). |
| 1039 |
.IP read/s |
| 1040 |
Number of 'read' RPC calls made per second. |
| 1041 |
.IP write/s |
| 1042 |
Number of 'write' RPC calls made per second. |
| 1043 |
.IP access/s |
| 1044 |
Number of 'access' RPC calls made per second. |
| 1045 |
.IP getatt/s |
| 1046 |
Number of 'getattr' RPC calls made per second. |
| 1047 |
.RE |
| 1048 |
|
| 1049 |
.IP |
| 1050 |
.RB "With the " "NFSD" |
| 1051 |
keyword, statistics about NFS server activity are reported. |
| 1052 |
The following values are displayed: |
| 1053 |
.RS |
| 1054 |
.IP scall/s |
| 1055 |
Number of RPC requests received per second. |
| 1056 |
.IP badcall/s |
| 1057 |
Number of bad RPC requests received per second, those whose |
| 1058 |
processing generated an error. |
| 1059 |
.IP packet/s |
| 1060 |
Number of network packets received per second. |
| 1061 |
.IP udp/s |
| 1062 |
Number of UDP packets received per second. |
| 1063 |
.IP tcp/s |
| 1064 |
Number of TCP packets received per second. |
| 1065 |
.IP hit/s |
| 1066 |
Number of reply cache hits per second. |
| 1067 |
.IP miss/s |
| 1068 |
Number of reply cache misses per second. |
| 1069 |
.IP sread/s |
| 1070 |
Number of 'read' RPC calls received per second. |
| 1071 |
.IP swrite/s |
| 1072 |
Number of 'write' RPC calls received per second. |
| 1073 |
.IP saccess/s |
| 1074 |
Number of 'access' RPC calls received per second. |
| 1075 |
.IP sgetatt/s |
| 1076 |
Number of 'getattr' RPC calls received per second. |
| 1077 |
.RE |
| 1078 |
|
| 1079 |
.IP |
| 1080 |
.RB "With the " "SOCK" |
| 1081 |
keyword, statistics on sockets in use are reported (IPv4). |
| 1082 |
The following values are displayed: |
| 1083 |
.RS |
| 1084 |
.IP totsck |
| 1085 |
Total number of sockets used by the system. |
| 1086 |
.IP tcpsck |
| 1087 |
Number of TCP sockets currently in use. |
| 1088 |
.IP udpsck |
| 1089 |
Number of UDP sockets currently in use. |
| 1090 |
.IP rawsck |
| 1091 |
Number of RAW sockets currently in use. |
| 1092 |
.IP ip\-frag |
| 1093 |
Number of IP fragments currently in queue. |
| 1094 |
.IP tcp\-tw |
| 1095 |
Number of TCP sockets in TIME_WAIT state. |
| 1096 |
.RE |
| 1097 |
|
| 1098 |
.IP |
| 1099 |
.RB "With the " "SOCK6" |
| 1100 |
keyword, statistics on sockets in use are reported (IPv6). |
| 1101 |
Note that IPv6 statistics depend on |
| 1102 |
.BR "sadc" "'s option " "\-S IPV6" |
| 1103 |
to be collected. |
| 1104 |
The following values are displayed: |
| 1105 |
.RS |
| 1106 |
.IP tcp6sck |
| 1107 |
Number of TCPv6 sockets currently in use. |
| 1108 |
.IP udp6sck |
| 1109 |
Number of UDPv6 sockets currently in use. |
| 1110 |
.IP raw6sck |
| 1111 |
Number of RAWv6 sockets currently in use. |
| 1112 |
.IP ip6\-frag |
| 1113 |
Number of IPv6 fragments currently in use. |
| 1114 |
.RE |
| 1115 |
|
| 1116 |
.IP |
| 1117 |
.RB "With the " "SOFT" |
| 1118 |
keyword, statistics about software-based network processing are reported. |
| 1119 |
The following values are displayed: |
| 1120 |
.RS |
| 1121 |
.IP total/s |
| 1122 |
The total number of network frames processed per second. |
| 1123 |
.IP dropd/s |
| 1124 |
The total number of network frames dropped per second because there |
| 1125 |
was no room on the processing queue. |
| 1126 |
.IP squeezd/s |
| 1127 |
The number of times the softirq handler function terminated per second |
| 1128 |
because its budget was consumed or the time limit was reached, but more |
| 1129 |
work could have been done. |
| 1130 |
.IP rx_rps/s |
| 1131 |
The number of times the CPU has been woken up per second |
| 1132 |
to process packets via an inter-processor interrupt. |
| 1133 |
.IP flw_lim/s |
| 1134 |
The number of times the flow limit has been reached per second. |
| 1135 |
Flow limiting is an optional RPS feature that can be used to limit the number of |
| 1136 |
packets queued to the backlog for each flow to a certain amount. |
| 1137 |
This can help ensure that smaller flows are processed even though |
| 1138 |
much larger flows are pushing packets in. |
| 1139 |
.IP blg_len |
| 1140 |
The length of the network backlog. |
| 1141 |
.RE |
| 1142 |
|
| 1143 |
.IP |
| 1144 |
.RB "With the " "TCP" |
| 1145 |
keyword, statistics about TCPv4 network traffic are reported. |
| 1146 |
Note that TCPv4 statistics depend on |
| 1147 |
.BR "sadc" "'s option " "\-S SNMP" |
| 1148 |
to be collected. |
| 1149 |
The following values are displayed (formal SNMP names between |
| 1150 |
square brackets): |
| 1151 |
.RS |
| 1152 |
.IP active/s |
| 1153 |
The number of times TCP connections have made a direct |
| 1154 |
transition to the SYN\-SENT state from the CLOSED state per second [tcpActiveOpens]. |
| 1155 |
.IP passive/s |
| 1156 |
The number of times TCP connections have made a direct |
| 1157 |
transition to the SYN\-RCVD state from the LISTEN state per second [tcpPassiveOpens]. |
| 1158 |
.IP iseg/s |
| 1159 |
The total number of segments received per second, including those |
| 1160 |
received in error [tcpInSegs]. This count includes segments received on |
| 1161 |
currently established connections. |
| 1162 |
.IP oseg/s |
| 1163 |
The total number of segments sent per second, including those on |
| 1164 |
current connections but excluding those containing only |
| 1165 |
retransmitted octets [tcpOutSegs]. |
| 1166 |
.RE |
| 1167 |
|
| 1168 |
.IP |
| 1169 |
.RB "With the " "ETCP" |
| 1170 |
keyword, statistics about TCPv4 network errors are reported. |
| 1171 |
Note that TCPv4 statistics depend on |
| 1172 |
.BR "sadc" "'s option " "\-S SNMP" |
| 1173 |
to be collected. |
| 1174 |
The following values are displayed (formal SNMP names between |
| 1175 |
square brackets): |
| 1176 |
.RS |
| 1177 |
.IP atmptf/s |
| 1178 |
The number of times per second TCP connections have made a direct |
| 1179 |
transition to the CLOSED state from either the SYN\-SENT |
| 1180 |
state or the SYN\-RCVD state, plus the number of times per second TCP |
| 1181 |
connections have made a direct transition to the LISTEN |
| 1182 |
state from the SYN\-RCVD state [tcpAttemptFails]. |
| 1183 |
.IP estres/s |
| 1184 |
The number of times per second TCP connections have made a direct |
| 1185 |
transition to the CLOSED state from either the ESTABLISHED |
| 1186 |
state or the CLOSE\-WAIT state [tcpEstabResets]. |
| 1187 |
.IP retrans/s |
| 1188 |
The total number of segments retransmitted per second - that is, the |
| 1189 |
number of TCP segments transmitted containing one or more |
| 1190 |
previously transmitted octets [tcpRetransSegs]. |
| 1191 |
.IP isegerr/s |
| 1192 |
The total number of segments received in error (e.g., bad |
| 1193 |
TCP checksums) per second [tcpInErrs]. |
| 1194 |
.IP orsts/s |
| 1195 |
The number of TCP segments sent per second containing the RST flag [tcpOutRsts]. |
| 1196 |
.RE |
| 1197 |
|
| 1198 |
.IP |
| 1199 |
.RB "With the " "UDP" |
| 1200 |
keyword, statistics about UDPv4 network traffic are reported. |
| 1201 |
Note that UDPv4 statistics depend on |
| 1202 |
.BR "sadc" "'s option " "\-S SNMP" |
| 1203 |
to be collected. |
| 1204 |
The following values are displayed (formal SNMP names between |
| 1205 |
square brackets): |
| 1206 |
.RS |
| 1207 |
.IP idgm/s |
| 1208 |
The total number of UDP datagrams delivered per second to UDP users [udpInDatagrams]. |
| 1209 |
.IP odgm/s |
| 1210 |
The total number of UDP datagrams sent per second from this entity [udpOutDatagrams]. |
| 1211 |
.IP noport/s |
| 1212 |
The total number of received UDP datagrams per second for which there |
| 1213 |
was no application at the destination port [udpNoPorts]. |
| 1214 |
.IP idgmerr/s |
| 1215 |
The number of received UDP datagrams per second that could not be |
| 1216 |
delivered for reasons other than the lack of an application |
| 1217 |
at the destination port [udpInErrors]. |
| 1218 |
.RE |
| 1219 |
|
| 1220 |
.IP |
| 1221 |
.RB "With the " "UDP6" |
| 1222 |
keyword, statistics about UDPv6 network traffic are reported. |
| 1223 |
Note that UDPv6 statistics depend on |
| 1224 |
.BR "sadc" "'s option " "\-S IPV6" |
| 1225 |
to be collected. |
| 1226 |
The following values are displayed (formal SNMP names between |
| 1227 |
square brackets): |
| 1228 |
.RS |
| 1229 |
.IP idgm6/s |
| 1230 |
The total number of UDP datagrams delivered per second to UDP users |
| 1231 |
[udpInDatagrams]. |
| 1232 |
.IP odgm6/s |
| 1233 |
The total number of UDP datagrams sent per second from this |
| 1234 |
entity [udpOutDatagrams]. |
| 1235 |
.IP noport6/s |
| 1236 |
The total number of received UDP datagrams per second for which there |
| 1237 |
was no application at the destination port [udpNoPorts]. |
| 1238 |
.IP idgmer6/s |
| 1239 |
The number of received UDP datagrams per second that could not be |
| 1240 |
delivered for reasons other than the lack of an application |
| 1241 |
at the destination port [udpInErrors]. |
| 1242 |
.RE |
| 1243 |
|
| 1244 |
.IP |
| 1245 |
.RB "The " "ALL" |
| 1246 |
keyword is equivalent to specifying all the keywords above and therefore all the network |
| 1247 |
activities are reported. |
| 1248 |
.TP |
| 1249 |
.BI "\-o [ " "filename " "]" |
| 1250 |
Save the readings in the file in binary form. Each reading |
| 1251 |
is in a separate record. The default value of the |
| 1252 |
.I filename |
| 1253 |
parameter is the current standard system activity daily data file. If |
| 1254 |
.I filename |
| 1255 |
is a directory instead of a plain file then it is considered as the directory |
| 1256 |
where the standard system activity daily data files are located. Option |
| 1257 |
.BR "\-o " "is exclusive of option " "\-f" "." |
| 1258 |
All the data available from the kernel are saved in the file (in fact, |
| 1259 |
.BR "sar " "calls its data collector " "sadc " "with the option " "\-S ALL" "." |
| 1260 |
.RB "See " "sadc" "(8) manual page)." |
| 1261 |
.TP |
| 1262 |
.BI "\-P { " "cpu_list " "| ALL }" |
| 1263 |
Report per-processor statistics for the specified processor or processors. |
| 1264 |
.I cpu_list |
| 1265 |
is a list of comma-separated values or range of values (e.g., |
| 1266 |
.BR "0,2,4\-7,12\-" ")." |
| 1267 |
Note that processor 0 is the first processor, and processor |
| 1268 |
.B all |
| 1269 |
is the global average among all processors. |
| 1270 |
Specifying the |
| 1271 |
.B ALL |
| 1272 |
keyword reports statistics for each individual processor, and globally for |
| 1273 |
all processors. Offline processors are not displayed. |
| 1274 |
.TP |
| 1275 |
.BR "\-p" ", " "\-\-pretty" |
| 1276 |
Make reports easier to read by a human. |
| 1277 |
This option may be especially useful when displaying e.g., network interfaces |
| 1278 |
or block devices statistics. |
| 1279 |
.TP |
| 1280 |
.BI "\-q [ " "keyword" "[,...] | ALL ]" |
| 1281 |
Report system load and pressure-stall statistics. |
| 1282 |
|
| 1283 |
Possible keywords are |
| 1284 |
.BR "CPU" ", " "IO" ", " "LOAD" ", " "MEM " "and " "PSI" "." |
| 1285 |
|
| 1286 |
.RB "With the " "CPU" |
| 1287 |
keyword, CPU pressure statistics are reported. |
| 1288 |
The following values are displayed: |
| 1289 |
.RS |
| 1290 |
.IP %scpu\-10 |
| 1291 |
Percentage of the time that at least some runnable tasks were delayed because the CPU |
| 1292 |
was unavailable to them, over the last 10 second window. |
| 1293 |
.IP %scpu\-60 |
| 1294 |
Percentage of the time that at least some runnable tasks were delayed because the CPU |
| 1295 |
was unavailable to them, over the last 60 second window. |
| 1296 |
.IP %scpu\-300 |
| 1297 |
Percentage of the time that at least some runnable tasks were delayed because the CPU |
| 1298 |
was unavailable to them, over the last 300 second window. |
| 1299 |
.IP %scpu |
| 1300 |
Percentage of the time that at least some runnable tasks were delayed because the CPU |
| 1301 |
was unavailable to them, over the last time interval. |
| 1302 |
.RE |
| 1303 |
|
| 1304 |
.IP |
| 1305 |
.RB "With the " "IO" |
| 1306 |
keyword, I/O pressure statistics are reported. |
| 1307 |
The following values are displayed: |
| 1308 |
.RS |
| 1309 |
.IP %sio\-10 |
| 1310 |
Percentage of the time that at least some tasks lost waiting for I/O, |
| 1311 |
over the last 10 second window. |
| 1312 |
.IP %sio\-60 |
| 1313 |
Percentage of the time that at least some tasks lost waiting for I/O, |
| 1314 |
over the last 60 second window. |
| 1315 |
.IP %sio\-300 |
| 1316 |
Percentage of the time that at least some tasks lost waiting for I/O, |
| 1317 |
over the last 300 second window. |
| 1318 |
.IP %sio |
| 1319 |
Percentage of the time that at least some tasks lost waiting for I/O, |
| 1320 |
over the last time interval. |
| 1321 |
.IP %fio\-10 |
| 1322 |
Percentage of the time during which all non-idle tasks were stalled |
| 1323 |
waiting for I/O, over the last 10 second window. |
| 1324 |
.IP %fio\-60 |
| 1325 |
Percentage of the time during which all non-idle tasks were stalled |
| 1326 |
waiting for I/O, over the last 60 second window. |
| 1327 |
.IP %fio\-300 |
| 1328 |
Percentage of the time during which all non-idle tasks were stalled |
| 1329 |
waiting for I/O, over the last 300 second window. |
| 1330 |
.IP %fio |
| 1331 |
Percentage of the time during which all non-idle tasks were stalled |
| 1332 |
waiting for I/O, over the last time interval. |
| 1333 |
.RE |
| 1334 |
|
| 1335 |
.IP |
| 1336 |
.RB "With the " "LOAD" |
| 1337 |
keyword, queue length and load averages statistics are reported. |
| 1338 |
The following values are displayed: |
| 1339 |
.RS |
| 1340 |
.IP runq\-sz |
| 1341 |
Run queue length (number of tasks running or waiting for run time). |
| 1342 |
.IP plist\-sz |
| 1343 |
Number of tasks in the task list. |
| 1344 |
.IP ldavg\-1 |
| 1345 |
System load average for the last minute. |
| 1346 |
The load average is calculated as the average number of runnable or |
| 1347 |
running tasks (R state), and the number of tasks in uninterruptible |
| 1348 |
sleep (D state) over the specified interval. |
| 1349 |
.IP ldavg\-5 |
| 1350 |
System load average for the past 5 minutes. |
| 1351 |
.IP ldavg\-15 |
| 1352 |
System load average for the past 15 minutes. |
| 1353 |
.IP blocked |
| 1354 |
Number of tasks currently blocked, waiting for I/O to complete. |
| 1355 |
.RE |
| 1356 |
|
| 1357 |
.IP |
| 1358 |
.RB "With the " "MEM" |
| 1359 |
keyword, memory pressure statistics are reported. |
| 1360 |
The following values are displayed: |
| 1361 |
.RS |
| 1362 |
.IP %smem\-10 |
| 1363 |
Percentage of the time during which at least some tasks were waiting |
| 1364 |
for memory resources, over the last 10 second window. |
| 1365 |
.IP %smem\-60 |
| 1366 |
Percentage of the time during which at least some tasks were waiting |
| 1367 |
for memory resources, over the last 60 second window. |
| 1368 |
.IP %smem\-300 |
| 1369 |
Percentage of the time during which at least some tasks were waiting |
| 1370 |
for memory resources, over the last 300 second window. |
| 1371 |
.IP %smem |
| 1372 |
Percentage of the time during which at least some tasks were waiting |
| 1373 |
for memory resources, over the last time interval. |
| 1374 |
.IP %fmem\-10 |
| 1375 |
Percentage of the time during which all non-idle tasks were stalled |
| 1376 |
waiting for memory resources, over the last 10 second window. |
| 1377 |
.IP %fmem\-60 |
| 1378 |
Percentage of the time during which all non-idle tasks were stalled |
| 1379 |
waiting for memory resources, over the last 60 second window. |
| 1380 |
.IP %fmem\-300 |
| 1381 |
Percentage of the time during which all non-idle tasks were stalled |
| 1382 |
waiting for memory resources, over the last 300 second window. |
| 1383 |
.IP %fmem |
| 1384 |
Percentage of the time during which all non-idle tasks were stalled |
| 1385 |
waiting for memory resources, over the last time interval. |
| 1386 |
.RE |
| 1387 |
|
| 1388 |
.IP |
| 1389 |
.RB "The " "PSI" |
| 1390 |
keyword is equivalent to specifying CPU, IO and MEM keywords together |
| 1391 |
and therefore all the pressure-stall statistics are reported. |
| 1392 |
|
| 1393 |
.RB "The " "ALL" |
| 1394 |
keyword is equivalent to specifying all the keywords above |
| 1395 |
and therefore all the statistics are reported. |
| 1396 |
.TP |
| 1397 |
.B \-r [ ALL ] |
| 1398 |
Report memory utilization statistics. The |
| 1399 |
.B ALL |
| 1400 |
keyword indicates that all the memory fields should be displayed. |
| 1401 |
The following values may be displayed: |
| 1402 |
.RS |
| 1403 |
.IP kbmemfree |
| 1404 |
Amount of free memory available in kilobytes. |
| 1405 |
.IP kbavail |
| 1406 |
Estimate of how much memory in kilobytes is available for starting new |
| 1407 |
applications, without swapping. |
| 1408 |
The estimate takes into account that the system needs some page cache to |
| 1409 |
function well, and that not all reclaimable slab will be reclaimable, |
| 1410 |
due to items being in use. The impact of those factors will vary from |
| 1411 |
system to system. |
| 1412 |
.IP kbmemused |
| 1413 |
Amount of used memory in kilobytes (calculated as total installed memory \- |
| 1414 |
kbmemfree \- kbbuffers \- kbcached \- kbslab). |
| 1415 |
.IP %memused |
| 1416 |
Percentage of used memory. |
| 1417 |
.IP kbbuffers |
| 1418 |
Amount of memory used as buffers by the kernel in kilobytes. |
| 1419 |
.IP kbcached |
| 1420 |
Amount of memory used to cache data by the kernel in kilobytes. |
| 1421 |
.IP kbcommit |
| 1422 |
Amount of memory in kilobytes needed for current workload. |
| 1423 |
This is an estimate of how much |
| 1424 |
RAM/swap is needed to guarantee that there never is out of memory. |
| 1425 |
.IP %commit |
| 1426 |
Percentage of memory needed for current workload in relation to the |
| 1427 |
total amount of memory (RAM+swap). This number may be greater |
| 1428 |
than 100% because the kernel usually overcommits memory. |
| 1429 |
.IP kbactive |
| 1430 |
Amount of active memory in kilobytes (memory that has been used more recently |
| 1431 |
and usually not reclaimed unless absolutely necessary). |
| 1432 |
.IP kbinact |
| 1433 |
Amount of inactive memory in kilobytes (memory which has been less recently |
| 1434 |
used. It is more eligible to be reclaimed for other purposes). |
| 1435 |
.IP kbdirty |
| 1436 |
Amount of memory in kilobytes waiting to get written back to the disk. |
| 1437 |
.IP kbanonpg |
| 1438 |
Amount of non-file backed pages in kilobytes mapped into userspace page tables. |
| 1439 |
.IP kbslab |
| 1440 |
Amount of memory in kilobytes used by the kernel to cache data structures |
| 1441 |
for its own use. |
| 1442 |
.IP kbkstack |
| 1443 |
Amount of memory in kilobytes used for kernel stack space. |
| 1444 |
.IP kbpgtbl |
| 1445 |
Amount of memory in kilobytes dedicated to the lowest level of page tables. |
| 1446 |
.IP kbvmused |
| 1447 |
Amount of memory in kilobytes of used virtual address space. |
| 1448 |
.RE |
| 1449 |
.TP |
| 1450 |
.B \-S |
| 1451 |
Report swap space utilization statistics. |
| 1452 |
The following values are displayed: |
| 1453 |
.RS |
| 1454 |
.IP kbswpfree |
| 1455 |
Amount of free swap space in kilobytes. |
| 1456 |
.IP kbswpused |
| 1457 |
Amount of used swap space in kilobytes. |
| 1458 |
.IP %swpused |
| 1459 |
Percentage of used swap space. |
| 1460 |
.IP kbswpcad |
| 1461 |
Amount of cached swap memory in kilobytes. |
| 1462 |
This is memory that once was swapped out, is swapped back in but still also |
| 1463 |
is in the swap area (if memory is needed it doesn't need to be swapped out |
| 1464 |
again because it is already in the swap area. This saves I/O). |
| 1465 |
.IP %swpcad |
| 1466 |
Percentage of cached swap memory in relation to the amount of used swap space. |
| 1467 |
.RE |
| 1468 |
.PP |
| 1469 |
.BI "\-s [ " "hh" ":" "mm" "[:" "ss" "] ]" |
| 1470 |
.br |
| 1471 |
.BI "\-s [ " "seconds_since_the_epoch " "]" |
| 1472 |
.RS |
| 1473 |
Set the starting time of the data, causing the |
| 1474 |
.B sar |
| 1475 |
command to extract records time-tagged at, or following, the time |
| 1476 |
specified. The default starting time is 08:00:00. |
| 1477 |
Hours must be given in 24-hour format, or as the number of seconds |
| 1478 |
since the epoch (given as a 10 digit number). This option can be |
| 1479 |
used only when data are read from a file (option |
| 1480 |
.BR "\-f" ")." |
| 1481 |
.RE |
| 1482 |
.TP |
| 1483 |
.B \-\-sadc |
| 1484 |
Indicate which data collector is called by |
| 1485 |
.BR "sar" "." |
| 1486 |
If the data collector is sought in |
| 1487 |
.B PATH |
| 1488 |
then enter "which sadc" to know where it is located. |
| 1489 |
.TP |
| 1490 |
.B \-t |
| 1491 |
When reading data from a daily data file, indicate that |
| 1492 |
.B sar |
| 1493 |
should display the timestamps in the original local time of |
| 1494 |
the data file creator. Without this option, the |
| 1495 |
.B sar |
| 1496 |
command displays the timestamps in the user's local time. |
| 1497 |
.TP |
| 1498 |
.B \-u [ ALL ] |
| 1499 |
Report CPU utilization. The |
| 1500 |
.B ALL |
| 1501 |
keyword indicates that all the CPU fields should be displayed. |
| 1502 |
The report may show the following fields: |
| 1503 |
.RS |
| 1504 |
.IP %user |
| 1505 |
Percentage of CPU utilization that occurred while executing at the user |
| 1506 |
level (application). Note that this field includes time spent running |
| 1507 |
virtual processors. |
| 1508 |
.IP %usr |
| 1509 |
Percentage of CPU utilization that occurred while executing at the user |
| 1510 |
level (application). Note that this field does NOT include time spent |
| 1511 |
running virtual processors. |
| 1512 |
.IP %nice |
| 1513 |
Percentage of CPU utilization that occurred while executing at the user |
| 1514 |
level with nice priority. |
| 1515 |
.IP %system |
| 1516 |
Percentage of CPU utilization that occurred while executing at the system |
| 1517 |
level (kernel). Note that this field includes time spent servicing |
| 1518 |
hardware and software interrupts. |
| 1519 |
.IP %sys |
| 1520 |
Percentage of CPU utilization that occurred while executing at the system |
| 1521 |
level (kernel). Note that this field does NOT include time spent servicing |
| 1522 |
hardware or software interrupts. |
| 1523 |
.IP %iowait |
| 1524 |
Percentage of time that the CPU or CPUs were idle during which |
| 1525 |
the system had an outstanding disk I/O request. |
| 1526 |
.IP %steal |
| 1527 |
Percentage of time spent in involuntary wait by the virtual CPU |
| 1528 |
or CPUs while the hypervisor was servicing another virtual processor. |
| 1529 |
.IP %irq |
| 1530 |
Percentage of time spent by the CPU or CPUs to service hardware interrupts. |
| 1531 |
.IP %soft |
| 1532 |
Percentage of time spent by the CPU or CPUs to service software interrupts. |
| 1533 |
.IP %guest |
| 1534 |
Percentage of time spent by the CPU or CPUs to run a virtual processor. |
| 1535 |
.IP %gnice |
| 1536 |
Percentage of time spent by the CPU or CPUs to run a niced guest. |
| 1537 |
.IP %idle |
| 1538 |
Percentage of time that the CPU or CPUs were idle and the system |
| 1539 |
did not have an outstanding disk I/O request. |
| 1540 |
.RE |
| 1541 |
.TP |
| 1542 |
.B \-V |
| 1543 |
Print version number then exit. |
| 1544 |
.TP |
| 1545 |
.B \-v |
| 1546 |
Report status of inode, file and other kernel tables. |
| 1547 |
The following values are displayed: |
| 1548 |
.RS |
| 1549 |
.IP dentunusd |
| 1550 |
Number of unused cache entries in the directory cache. |
| 1551 |
.IP file\-nr |
| 1552 |
Number of file handles used by the system. |
| 1553 |
.IP inode\-nr |
| 1554 |
Number of inode handlers used by the system. |
| 1555 |
.IP pty\-nr |
| 1556 |
Number of pseudo-terminals used by the system. |
| 1557 |
.RE |
| 1558 |
.TP |
| 1559 |
.B \-W |
| 1560 |
Report swapping statistics. The following values are displayed: |
| 1561 |
.RS |
| 1562 |
.IP pswpin/s |
| 1563 |
Total number of swap pages the system brought in per second. |
| 1564 |
.IP pswpout/s |
| 1565 |
Total number of swap pages the system brought out per second. |
| 1566 |
.RE |
| 1567 |
.TP |
| 1568 |
.B \-w |
| 1569 |
Report task creation and system switching activity. |
| 1570 |
The following values are displayed: |
| 1571 |
.RS |
| 1572 |
.IP proc/s |
| 1573 |
Total number of tasks created per second. |
| 1574 |
.IP cswch/s |
| 1575 |
Total number of context switches per second. |
| 1576 |
.RE |
| 1577 |
.TP |
| 1578 |
.B \-x |
| 1579 |
Extended reports: Display minimum and maximum values in addition to |
| 1580 |
average ones at the end of the report. |
| 1581 |
.TP |
| 1582 |
.B \-y |
| 1583 |
Report TTY devices activity. The following values are displayed: |
| 1584 |
.RS |
| 1585 |
.IP rcvin/s |
| 1586 |
Number of receive interrupts per second for current serial line. |
| 1587 |
Serial line number is given in the TTY column. |
| 1588 |
.IP xmtin/s |
| 1589 |
Number of transmit interrupts per second for current serial line. |
| 1590 |
.IP framerr/s |
| 1591 |
Number of frame errors per second for current serial line. |
| 1592 |
.IP prtyerr/s |
| 1593 |
Number of parity errors per second for current serial line. |
| 1594 |
.IP brk/s |
| 1595 |
Number of breaks per second for current serial line. |
| 1596 |
.IP ovrun/s |
| 1597 |
Number of overrun errors per second for current serial line. |
| 1598 |
.RE |
| 1599 |
.TP |
| 1600 |
.B \-z |
| 1601 |
.RB "Tell " "sar" |
| 1602 |
to omit output for any devices for which there was no activity during the |
| 1603 |
sample period. |
| 1604 |
|
| 1605 |
.SH ENVIRONMENT |
| 1606 |
The |
| 1607 |
.B sar |
| 1608 |
command takes into account the following environment variables: |
| 1609 |
.TP |
| 1610 |
.B S_COLORS |
| 1611 |
By default statistics are displayed in color when the output is connected to a terminal. |
| 1612 |
Use this variable to change the settings. Possible values for this variable are |
| 1613 |
.IR "never" ", " "always " "or " "auto" |
| 1614 |
(the latter is equivalent to the default settings). |
| 1615 |
.br |
| 1616 |
Please note that the color (being red, yellow, or some other color) used to display a value |
| 1617 |
is not indicative of any kind of issue simply because of the color. It only indicates different |
| 1618 |
ranges of values. |
| 1619 |
.TP |
| 1620 |
.B S_COLORS_SGR |
| 1621 |
Specify the colors and other attributes used to display statistics on the terminal. |
| 1622 |
Its value is a colon-separated list of capabilities that defaults to |
| 1623 |
.BR "C=33;22:I=32;22:N=34;1:R=31;22:W=35;1:X=31;1:Z=34;22" "." |
| 1624 |
Supported capabilities are: |
| 1625 |
.RS |
| 1626 |
.TP |
| 1627 |
.B C= |
| 1628 |
SGR (Select Graphic Rendition) substring for comments inserted in the binary daily |
| 1629 |
data files. |
| 1630 |
.TP |
| 1631 |
.B I= |
| 1632 |
SGR substring for item names or values (eg. network interfaces, CPU number...) |
| 1633 |
.TP |
| 1634 |
.B N= |
| 1635 |
SGR substring for non-zero statistics values. |
| 1636 |
.TP |
| 1637 |
.B R= |
| 1638 |
SGR substring for restart messages. |
| 1639 |
.TP |
| 1640 |
.BR "W=" " (or " "M=" ")" |
| 1641 |
SGR substring for percentage values in the range from 75% to 90% (or in the range 10% to 25% |
| 1642 |
depending on the metric's meaning). |
| 1643 |
It is also used for negative values in the range from \-10 to \-5. |
| 1644 |
.TP |
| 1645 |
.BR "X=" " (or " "H=" ")" |
| 1646 |
SGR substring for percentage values greater than or equal to 90% (or lower than or equal to 10% |
| 1647 |
depending on the metric's meaning). |
| 1648 |
It is also used for negative values lower than or equal to \-10. |
| 1649 |
.TP |
| 1650 |
.B Z= |
| 1651 |
SGR substring for zero values. |
| 1652 |
.RE |
| 1653 |
.TP |
| 1654 |
.B S_REPEAT_HEADER |
| 1655 |
This variable contains the maximum number of lines after which a header has to be |
| 1656 |
displayed by |
| 1657 |
.B sar |
| 1658 |
when the output is not a terminal. |
| 1659 |
.TP |
| 1660 |
.B S_TIME_DEF_TIME |
| 1661 |
If this variable exists and its value is |
| 1662 |
.BR "UTC " "then " "sar" |
| 1663 |
will save its data in UTC time (data will still be displayed in local time). |
| 1664 |
.B sar |
| 1665 |
will also use UTC time instead of local time to determine the current daily |
| 1666 |
data file located in the |
| 1667 |
.I /var/log/sysstat |
| 1668 |
directory. This variable may be useful for servers with users located across |
| 1669 |
several timezones. |
| 1670 |
.TP |
| 1671 |
.B S_TIME_FORMAT |
| 1672 |
If this variable exists and its value is |
| 1673 |
.B ISO |
| 1674 |
then the current locale will be ignored when printing the date in the report header. |
| 1675 |
.RB "The " "sar" |
| 1676 |
command will use the ISO 8601 format (YYYY-MM-DD) instead. |
| 1677 |
The timestamp will also be compliant with ISO 8601 format. |
| 1678 |
|
| 1679 |
.SH EXAMPLES |
| 1680 |
.TP |
| 1681 |
.B sar \-u 2 5 |
| 1682 |
Report CPU utilization for each 2 seconds. 5 lines are displayed. |
| 1683 |
.TP |
| 1684 |
.B sar \-I \-\-int=14 \-o int14.file 2 10 |
| 1685 |
Report statistics on IRQ 14 for each 2 seconds. 10 lines are displayed. |
| 1686 |
Data are stored in a file called |
| 1687 |
.IR "int14.file" "." |
| 1688 |
.TP |
| 1689 |
.B sar \-r \-n DEV \-f /var/log/sysstat/sa16 |
| 1690 |
.RI "Display memory and network statistics saved in daily data file " "sa16" "." |
| 1691 |
.TP |
| 1692 |
.B sar \-A |
| 1693 |
Display all the statistics saved in current daily data file. |
| 1694 |
|
| 1695 |
.SH BUGS |
| 1696 |
.IR "/proc " "filesystem must be mounted for the |
| 1697 |
.BR "sar " "command to work." |
| 1698 |
.PP |
| 1699 |
All the statistics are not necessarily available, depending on the kernel version used. |
| 1700 |
.B sar |
| 1701 |
assumes that you are using at least a 2.6 kernel. |
| 1702 |
.PP |
| 1703 |
.RB "Although " "sar" |
| 1704 |
speaks of kilobytes (kB), megabytes (MB)..., it actually uses kibibytes (kiB), mebibytes (MiB)... |
| 1705 |
A kibibyte is equal to 1024 bytes, and a mebibyte is equal to 1024 kibibytes. |
| 1706 |
|
| 1707 |
.SH FILES |
| 1708 |
.I /var/log/sysstat/saDD |
| 1709 |
.br |
| 1710 |
.I /var/log/sysstat/saYYYYMMDD |
| 1711 |
.RS |
| 1712 |
The standard system activity daily data files and their default location. |
| 1713 |
.IR "YYYY " "stands for the current year, " "MM " "for the current month and " "DD" |
| 1714 |
for the current day. |
| 1715 |
.RE |
| 1716 |
|
| 1717 |
.IR "/proc " "and " "/sys " "contain various files with system statistics." |
| 1718 |
|
| 1719 |
.SH AUTHOR |
| 1720 |
Sebastien Godard (sysstat <at> orange.fr) |
| 1721 |
|
| 1722 |
.SH SEE ALSO |
| 1723 |
.BR "sadc" "(8), " "sa1" "(8), " "sa2" "(8), " "sadf" "(1), " "sysstat" "(5), " "pidstat" "(1)," |
| 1724 |
.BR "mpstat" "(1), " "iostat" "(1), " "vmstat" "(8)" |
| 1725 |
.PP |
| 1726 |
.I https://github.com/sysstat/sysstat |
| 1727 |
.br |
| 1728 |
.I https://sysstat.github.io/ |