sar(1M)sar(1M)NAMEsar - system activity reporter
SYNOPSIS
psetid] file] t [n]
psetid] time] time] sec] file]
DESCRIPTION
In the first form above, samples cumulative activity counters in the
operating system at n intervals of t seconds. If the option is speci‐
fied, it saves the samples in file in binary format. The default value
of n is 1.
In the second form, with no sampling interval specified, extracts data
from a previously recorded file, either the one specified by option or,
by default, the standard system activity daily data file for the cur‐
rent day dd. The starting and ending times of the report can be
bounded via the and time arguments of the form hh[:mm[:ss]]. The
option selects records at sec-second intervals. Otherwise, all inter‐
vals found in the data file are reported.
Options
Subsets of data to be printed are specified by options:
Report use of file access system routines:
Number of file system
calls per second.
Number of file system
(pathname translation) calls per second.
Number of file system blocks read per second doing direc‐
tory lookup.
Report all data.
Equivalent to
Report buffer activity:
Number of physical reads per second from the disk
(or other block devices) to the buffer
cache.
Number of physical writes per second
from the buffer cache to the disk (or other
block device).
Number of reads per second from buffer cache.
Number of writes per second to buffer cache.
Buffer cache hit ratio for read requests,
for example, 1 − bread/lread.
Buffer cache hit ratio for write requests,
for example, 1 − bwrit/lwrit.
Number of reads per second from character device
using the (raw I/O) mechanism.
Number of writes per second to character device
using the (raw I/O) mechanism.
Report system calls:
Number of system calls of all types per second.
Number of and/or system calls per second.
Number of and/or system calls per second.
Number of and/or system calls per second.
Number of system calls per second.
Number of characters transferred by read system calls
block devices only) per second.
Number of characters transferred by write system calls
(block devices only) per second.
Report activity for each block device, for example, disk or tape
drive.
One line is printed for each device that had activity dur‐
ing the last interval. If no devices were active, a blank
line is printed. Each line contains the following data:
Logical name of the device and its corresponding instance.
Devices are categorized into the following
device types:
disk3 − SCSI and NIO FL disks
sdisk − SCSI disks.
Portion of time device was busy servicing a request.
Average number of requests outstanding for the device.
Number of data transfers per second
(read and writes) from and to the device.
Number of reads per second from the device. This informa‐
tion is
available only when used with the option.
Number of writes per second to the device. This informa‐
tion is
available only when used with the option.
Number of bytes transferred (in 512-byte units) from and
to the device.
Average time (in milliseconds)
that transfer requests waited idly on queue
for the device.
Average time (in milliseconds) to service each transfer
request
(includes seek, rotational latency, and
data transfer times) for the device.
Refer to the section for information about legacy LUN.
Uses file as the data source for
Default is the current daily data file
Report activity for each HBA device.
One line is printed for each HBA device that had activity
during the last interval. If no devices were active, a
blank line is printed. Each line contains the following
data:
Logical name of the HBA device and its corresponding
instance.
Portion of time device was busy servicing a request.
Number of bytes transferred (in MBs) from and to the
device.
Number of data transfers per second
(read and writes) from and to the device.
Number of reads per second from the device.
Number of writes per second to the device.
Number of bytes read (in MBs) from the device.
Number of bytes written (in MBs) to the device.
Average number of requests outstanding for the device.
Average time (in milliseconds)
that transfer requests waited idly on queue
for the device.
Average time (in milliseconds) to service each transfer
request
for the device.
Report activity for each active lunpath. One line is
printed for each lunpath that had activity during the last
interval. If no lunpath were active, a blank line is
printed.
NOTE: The lunpaths are not displayed in any specific order
and it is not guaranteed that the current order will be
maintained in future releases.
Each line contains the following data:
symbolic name of the lunpath of the form:
diskm_lunpathn, where m and n are the
instance numbers of LUN and lunpath respec‐
tively. These instance numbers are dis‐
played by using option for the LUN and lun‐
path entries. For more information on LUN
and lunpath hardware path, refer to
intro(7).
Portion of time lunpath was busy servicing a request.
Average number of requests outstanding for the lunpath.
Number of reads per second from the lunpath.
Number of writes per second to the lunpath.
Number of bytes transferred (in 512-byte units) from and
to
the LUN using this lunpath.
Average time (in milliseconds) that transfer requests
waited idly on queue for this lunpath.
Average time (in milliseconds) to service each transfer
request (includes seek, rotational latency,
and data transfer times) for this lunpath.
Report message and semaphore activities:
Number of System V
calls per second.
Number of System V
calls per second.
Number of System V
calls per second. This value will only be
reported if the option is also explicitly
specified.
Report the per-processor data on a multi-processor system when
used with
and/or options. If the option is not used on a multi-pro‐
cessor system, the output format of the and options is the
same as the uni-processor output format and the data
reported is the average value of all the active proces‐
sors.
Saves the samples in binary format.
Report the specified ProcessorSet (pset) information. This
option should
be used with the or option.
Report ProcessorSet (pset) information, mapping to the processor
in the system or the specified pset. This option can only
be used with and or and options. It can also be combined
with option to display information for a particular pset.
If system is not pset configured, it will display a warn‐
ing message.
Report average queue length while occupied,
and percent of time occupied. On a multi-processor
machine, if the option is used together with the option,
the per-CPU run queue as well as the average run queue of
all the active processors are reported. If the option is
not used, only the average run queue information of all
the active processors is reported. In a multi-processor
pset configured system, if the option is used with the
option, then the pset column will be displayed before the
cpu column:
pset id (only on a multi-processor and pset configured
system,
used with and option).
cpu number (only on a multi-processor system and used with
the
option).
Average length of the run queue(s) of processes (in memory
and runnable).
The percentage of time the run queue(s) were occupied by
processes
(in memory and runnable).
Average length of the swap queue of runnable processes
(processes swapped out but ready to run).
This column will not be displayed when
option is used to display a particular Pro‐
cessorSet (pset) information in a pset con‐
figured system.
The percentage of time the swap queue of runnable pro‐
cesses
(processes swapped out but ready to run)
was occupied. This column will not be dis‐
played when option is used to display a
particular ProcessorSet (pset) information
in a pset configured system.
This option is used with the
option to report the number of reads and writes per second
to the device. Refer to the description of the option for
details.
Report the number of System V
calls per second. This option is valid only if the option
is specified.
Report activity for each tape device.
One line is printed for each tape device that had activity
during the last interval. If no devices were active, a
blank line is printed. Each line contains the following
data:
Logical name of the tape device and its corresponding
instance.
Portion of time the device was busy servicing a request.
Number of reads per second from the device.
Number of writes per second to the device.
Number of bytes read (in MBs) from the device.
Number of bytes written (in MBs) to the device.
Average time (in milliseconds) to service each transfer
request
(includes seek, rotational latency, and
data transfer times) for the device.
Report CPU
utilization (the default); portion of time running in one
of several modes. On a multi-processor system, if the
option is used together with the option, per-CPU utiliza‐
tion as well as the average CPU utilization of all the
active processors are reported. If the option is not
used, only the average CPU utilization of all the active
processors is reported. On a multi-processor ProcessorSet
(pset) configured system, if the option is used together
with the and options, the column for pset mapping to pro‐
cessor will be displayed before the cpu column.
pset id (only on a multi-processor pset configured system
with the
and option).
cpu number (only on a multi-processor system with the
option).
user mode.
system mode.
idle with some process waiting for I/O (only block I/O,
raw I/O, or VM pageins/swapins indicated).
otherwise idle.
Report status of text, process, inode and file tables:
(Not Applicable).
The current-size and maximum-size of the process table.
The current-size and maximum-size of the inode table
(inode cache).
The current-size and maximum-size of the system file ta‐
ble.
(Not Applicable).
The number of times the process table overflowed
(number of times the kernel could not find
any available process table entries)
between sample points.
The number of times the inode table (inode cache) over‐
flowed
(number of times the kernel could not find
any available inode table entries) between
sample points.
The number of times the system file table overflowed
(number of times the kernel could not find
any available file table entries) between
sample points.
Report system swapping and switching activity:
Number of process swapins per second.
Number of process swapouts per second.
Number of 512-byte units transferred for swapins per sec‐
ond.
Number of 512-byte units transferred for swapouts per sec‐
ond.
Number of process context switches per second.
Report tty device activity:
Raw input characters per second.
Input characters per second processed by
Output characters per second.
Receive incoming character interrupts per second.
Transmit outgoing character interrupts per second.
Modem interrupt rate (not supported; always 0).
Report Memory File System (MemFS) data.
will recognize the option when HP's optional product is
installed on the system; it will display the number of
MemFS blocks in memory (blkcnt) and the number of MemFS
blocks in swap (swpcnt). The number of MemFS blocks is
expressed in terms of 4 KB only.
Note that the option does not record MemFS data into a
file. However, if the and options are used together, then
the data for all the options except the MemFS option is
recorded into a file. You can then query this data.
It is also not possible to use the option with the option.
Using the option and the option together will not display
the MemFS data. However, using the option with t and n
will display the MemFS data.
RETURN VALUE
This command returns the following values upon completion:
0 Completed successfully.
>0 An error occurred.
EXAMPLES
Watch CPU activity evolve for 5 seconds:
Watch CPU activity evolve for 10 minutes and save data:
If the product is installed, watch MemFS activity evolve for 10 minutes
and save data; all data other than MemFS data is stored in file
Review disk and tape activity from that period later:
Review cpu utilization on a multi-processor system later:
Watch lunpath activity evolve for 5 seconds:
Watch tape activity evolve for 5 seconds:
Watch HBAs activity evolve for 5 seconds:
Watch disk reads and writes per second for 5 seconds:
WARNINGS
Users of must not rely on the exact field widths and spacing of its
output, as these will vary depending on the system, the release of HP-
UX, and the data to be displayed.
The output of is unpredictable if a corrupted or wrong data file is
given as argument with the option.
For devices that are opened and closed frequently (such as tapes), data
might not be reported for the activity in the prior interval. The data
might not be reported, because the device might be closed at the
instant when data is collected. The recommendation for such devices is
to use shorter intervals (t) of seconds.
When multi-pathing is enabled and I/O is issued on a legacy LUN, may
not report I/O that is redirected through other paths to the agile
device.
The behavior of the and scripts and the command remain unaltered with
the product; these scripts and command do not record data reported by
the option.
FILES
daily data file, where dd is two digits representing the day of the
month.
SEE ALSOioscan(1M), sa1(1M), pstat(2), intro(7).
STANDARDS CONFORMANCEsar(1M)
