iopending - plot number of pending disk events. Uses DTrace. iopending [-c] [-d device] [-f filename] [-m mount_point] [interval [count]] This samples the number of disk events that are pending and plots a distribution graph. By doing this the "serialness" or "parallelness" of disk behaviour can be distinguished. A high occurance of a pending value of more than 1 is an indication of saturation. Since this uses DTrace, only the root user or users with the dtrace_kernel privilege can run this command. Solaris stable - needs the io provider.

-c clear screen

-d device instance name to snoop (eg, dad0)

-f filename full pathname of file to snoop

-m mount_point mountpoint for filesystem to snoop

Default output, print I/O summary every 1 second, # iopending

Print 10 second samples, # iopending 10

Print 12 x 5 second samples, # iopending 5 12

Snoop events on the root filesystem only, # iopending -m /

value number of pending events, 0 == idle

count number of samples @ 1000 Hz

load 1 min load average

disk_r total disk read Kb for sample

disk_w total disk write Kb for sample

Dr Rex di Bona

See the DTraceToolkit for further documentation under the Docs directory. The DTraceToolkit docs may include full worked examples with verbose descriptions explaining the output. iopending will run forever until Ctrl-C is hit, or the specified count is reached. Brendan Gregg [Sydney, Australia] iosnoop(1M), iotop(1M), dtrace(1M)