illumos: manual page: iostat.8

IOSTAT(8) Maintenance Commands and Procedures IOSTAT(8)

NAME

iostat - report I/O statistics

SYNOPSIS

/usr/bin/iostat [-cCdDeEiImMnpPrstxXYz] [-l n] [-T u | d]
[disk]... [interval [count]]

DESCRIPTION

The iostat utility iteratively reports terminal, disk, and tape I/O
activity, as well as CPU utilization. The first line of output is for all
time since boot; each subsequent line is for the prior interval only.

To compute this information, the kernel maintains a number of counters.
For each disk, the kernel counts reads, writes, bytes read, and bytes
written. The kernel also takes hi-res time stamps at queue entry and exit
points, which allows it to keep track of the residence time and
cumulative residence-length product for each queue. Using these values,
iostat produces highly accurate measures of throughput, utilization,
queue lengths, transaction rates and service time. For terminals
collectively, the kernel simply counts the number of input and output
characters.

During execution of the kernel status command, the state of the system
can change. If relevant, a state change message is included in the iostat
output, in one of the following forms:

<<device added: sd0>>
<<device removed: sd0>>
<<partition added: sd0,a>>
<<partition removed: sd0,a>>
<<NFS mounted: nfs1>>
<<NFS unmounted: nfs1>>
<<multi-path added: ssd4>>
<<multi-path removed: ssd4>>
<<controller added: c1>>
<<controller removed: c1>>
<<processors added: 1, 3>>
<<processors removed: 1, 3>>

Note that the names printed in these state change messages are affected
by the -n and -m options as appropriate.

For more general system statistics, use sar(1), sar(8), or vmstat(8).

Output

The output of the iostat utility includes the following information.

device
name of the disk

r/s
reads per second

w/s
writes per second

kr/s
kilobytes read per second

The average I/O size during the interval can be computed from
kr/s divided by r/s.

kw/s
kilobytes written per second

The average I/O size during the interval can be computed from
kw/s divided by w/s.

wait
average number of transactions waiting for service (queue
length)

This is the number of I/O operations held in the device driver
queue waiting for acceptance by the device.

actv
average number of transactions actively being serviced (removed
from the queue but not yet completed)

This is the number of I/O operations accepted, but not yet
serviced, by the device.

svc_t
average response time of transactions, in milliseconds

The svc_t output reports the overall response time, rather than
the service time, of a device. The overall time includes the
time that transactions are in queue and the time that
transactions are being serviced. The time spent in queue is
shown with the -x option in the wsvc_t output column. The time
spent servicing transactions is the true service time. Service
time is also shown with the -x option and appears in the asvc_t
output column of the same report.

%w
percent of time there are transactions waiting for service
(queue non-empty)

%b
percent of time the disk is busy (transactions in progress)

wsvc_t
average service time in wait queue, in milliseconds

asvc_t
average service time of active transactions, in milliseconds

OPTIONS

The following options are supported:

-c
Report the percentage of time the system has spent in user
mode, in system mode, in dtrace probes, and idling. See the
NOTES section and mpstat(8) for more information.

-C
When the -x option is also selected, report extended disk
statistics aggregated by controller id.

-d
For each disk, report the number of kilobytes transferred per
second, the number of transfers per second, and the average
service time in milliseconds.

-D
For each disk, report the reads per second, writes per
second, and percentage disk utilization.

-e
Display device error summary statistics. The total errors,
hard errors, soft errors, and transport errors are displayed.

-E
Display all device error statistics.

-i
In -E output, display the Device ID instead of the Serial No.
The Device Id is a unique identifier registered by a driver
through ddi_devid_register(9F).

-I
Report the counts in each interval, rather than rates (where
applicable).

-l n
Limit the number of disks included in the report to n; the
disk limit defaults to 4 for -d and -D, and unlimited for -x.
Note: disks explicitly requested (see disk below) are not
subject to this disk limit.

-m
Report file system mount points. This option is most useful
if the -P or -p option is also specified or used in
conjunction with -Xn or -en. The -m option is useful only if
the mount point is actually listed in the output. This option
can only be used in conjunction with the -n option.

-M
Display data throughput in MB/sec instead of KB/sec.

-n
Display names in descriptive format. For example, cXtYdZ,
rmt/N, server:/export/path.

By default, disks are identified by instance names such as
ssd23 or md301. Combining the -n option with the -x option
causes disk names to display in the cXtYdZsN format which is
more easily associated with physical hardware
characteristics. The cXtYdZsN format is particularly useful
in FibreChannel (FC) environments where the FC World Wide
Name appears in the t field.

-p
For each disk, report per-partition statistics in addition to
per-device statistics.

-P
For each disk, report per-partition statistics only, no per-
device statistics.

-r
Display data in a comma-separated format.

-s
Suppress messages related to state changes.

-t
Report the number of characters read and written to terminals
per second.

-T u | d
Display a time stamp.

Specify u for a printed representation of the internal
representation of time. See time(2). Specify d for standard
date format. See date(1).

-X
For disks under scsi_vhci(4D) control, in addition to disk
lun statistics, also report statistics for lun.controller.

-x
Report extended disk statistics. By default, disks are
identified by instance names such as ssd23 or md301.
Combining the x option with the -n option causes disk names
to display in the cXtYdZsN format, more easily associated
with physical hardware characteristics. Using the cXtYdZsN
format is particularly helpful in the FibreChannel
environments where the FC World Wide Name appears in the t
field.

If no output display is requested (no -x, -e, -E), -x is
implied.

-Y
For disks under scsi_vhci(4D) control, in addition to disk
lun statistics, also report statistics for lun.targetport and
lun.targetport.controller.

In -n (descriptive) mode the targetport is shown in using the
target-port property of the path. Without -n the targetport
is shown using the shorter port-id. All target ports with the
same target-port property value share the same port-id. The
target-port-to-port-id association does not persist across
reboot.

If no output display is requested (no -x, -e, -E), -x is
implied.

-z
Do not print lines whose underlying data values are all
zeros.

The option set -xcnCXTdz interval is particularly useful for determining
whether disk I/O problems exist and for identifying problems.

OPERANDS

The following operands are supported:

count
Display only count reports.

disk
Explicitly specify the disks to be reported; in addition to
any explicit disks, any active disks up to the disk limit
(see -l above) will also be reported.

interval
Report once each interval seconds.

EXAMPLES

Example 1: Using iostat to Generate User and System Operation Statistics

The following command displays two reports of extended device statistics,
aggregated by controller id, for user (us) and system (sy) operations.
Because the -n option is used with the -x option, devices are identified
by controller names.

example% iostat -xcnCXTdz 5

Mon Nov 24 14:58:36 2003
cpu
us sy dt id
14 31 0 20
extended device statistics
r/s w/s kr/s kw wait actv wsvc_t asvc_t %w %b device
3.8 29.9 145.8 44.0 0.0 0.2 0.1 6.4 0 5 c0
666.3 814.8 12577.6 17591.1 91.3 82.3 61.6 55.6 0 2 c12
180.0 234.6 4401.1 5712.6 0.0 147.7 0.0 356.3 0 98 d10

Mon Nov 24 14:58:41 2003
cpu
us sy dt id
11 31 1 22
extended device statistics
r/s w/s kr/s kw wait actv wsvc_t asvc_t %w %b device
0.8 41.0 5.2 20.5 0.0 0.2 0.2 4.4 0 6 c0
565.3 581.7 8573.2 10458.9 0.0 26.6 0.0 23.2 0 3 c12
106.5 81.3 3393.2 1948.6 0.0 5.7 0.0 30.1 0 99 d10

Example 2: Using iostat to Generate TTY Statistics

The following command displays two reports on the activity of five disks
in different modes of operation. Because the -x option is used, disks are
identified by instance names.

example% iostat -x tc 5 2

extended device statistics tty cpu
device r/s w/s kr/s kw/s wait actv svc_t %w %b tin tout us sy dt id
sd0 0.4 0.3 10.4 8.0 0.0 0.0 36.9 0 1 0 10 0 0 0 99
sd1 0.0 0.0 0.3 0.4 0.0 0.0 35.0 0 0
sd6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0
nfs1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0
nfs2 0.0 0.0 0.0 0.1 0.0 0.0 35.6 0 0
extended device statistics tty cpu
device r/s w/s kr/s kw/s wait actv svc_t %w %b tin tout us sy dt id
sd0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0 155 0 0 0 100
sd1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0
sd6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0
nfs1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0
nfs2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0

Example 3: Using iostat to Generate Partition and Device Statistics

The following command generates partition and device statistics for each
disk. Because the -n option is used with the -x option, disks are
identified by controller names.

example% iostat -xnp

extended device statistics
r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device
0.4 0.3 10.4 7.9 0.0 0.0 0.0 36.9 0 1 c0t0d0
0.3 0.3 9.0 7.3 0.0 0.0 0.0 37.2 0 1 c0t0d0s0
0.0 0.0 0.1 0.5 0.0 0.0 0.0 34.0 0 0 c0t0d0s1
0.0 0.0 0.0 0.1 0.0 0.0 0.6 35.0 0 0 fuji:/export/home/user3

Example 4: Show Translation from Instance Name to Descriptive Name

The following example illustrates the use of iostat to translate a
specific instance name to a descriptive name.

example% iostat -xn sd1
extended device statistics
r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c8t1d0

Example 5: Show Target Port and Controller Activity for a Specific Disk

In the following example, there are four controllers, all connected to
the same target port.

# iostat -Y ssd22
extended device statistics
device r/s w/s kr/s kw/s wait actv svc_t %w %b
ssd22 0.2 0.0 1.5 0.0 0.0 0.0 0.7 0 0
ssd22.t2 0.2 0.0 1.5 0.0 0.0 0.0 0.0 0 0
ssd22.t2.fp0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0 0
ssd22.t2.fp1 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0 0
ssd22.t2.fp2 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0 0
ssd22.t2.fp3 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0 0

ATTRIBUTES

See attributes(7) for descriptions of the following attributes:

+--------------------+-----------------+
| ATTRIBUTE TYPE | ATTRIBUTE VALUE |
+--------------------+-----------------+
|Interface Stability | See below. |
+--------------------+-----------------+

Invocation is evolving. Human readable output is unstable.

NOTES

The sum of CPU utilization might vary slightly from 100 because of
rounding errors in the production of a percentage figure.

The svc_t response time is not particularly significant when the I/0
(r/s+w/s) rates are under 0.5 per second. Harmless spikes are fairly
normal in such cases.

The mpstat utility reports the same dt, usr, and sys statistics. See
mpstat(8) for more information.

When executed in a zone and if the pools facility is active, iostat(8)
will only provide information for those processors in the processor set
of the pool to which the zone is bound.

March 23, 2009 IOSTAT(8)