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.
SEE ALSO
date(1),
sar(1),
time(2),
scsi_vhci(4D),
attributes(7),
mpstat(8),
sar(8),
vmstat(8)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)