HOSTSTAT

Section: Maintenance Commands (8)
Updated: November 1991
Index Return to Main Contents

NAME

hoststat - report performance statistics for several hosts

SYNOPSIS

hoststat host1 host2 ... [ interval ]

DESCRIPTION

hoststat uses the rstatd(8) to gather system statistics from the specified hosts. It is typically used to watch a number of processors running distributed applications. Other uses include monitoring a network of machines for overloading, balance, etc.

Upon start up, hoststat displays a one-line summary of the activity on each system over the last second. This information is repeated every interval seconds (default: 5 seconds).

For example, the following command displays a summary of what the system is doing every five seconds.

example% hoststat sunbox0 sunbox1 sunbox2 sunbox3
   host load ipkt opkt err d0 d1 d2 d3 ipg opg intr    cs usr sys idl
sunbox0  1.4  717  768   2 40 88  0  0  59  13 2643  3730  34 154  16
sunbox1  0.4   72   73   0  0  0  0  0   6   0  216   144  67  38   0
sunbox2  0.4    5    5   0  0  0  0  0   2   0   34    31  99   2   0
sunbox3  0.6    0    1   0  0  0  0  0   0   0   14    19  87  14   0
   host load ipkt opkt err d0 d1 d2 d3 ipg opg intr    cs usr sys idl
sunbox0  0.5  453  514   2  9 14  0  0  23  18 1500  2220  13  87   1
sunbox1  0.1   54   46   0  0  0  0  0   3   0  141   100  78  22   0
sunbox2  0.1   35   32   0  0  0  0  0   1   0   96    77  84  16   0
sunbox3  0.1    0    0   0  0  0  0  0   0   0    1    15  90  10   0
^C
example%

All of the fields of hoststat are averaged over the interval specified. The fields are:

load: The 1 minute load average.
ipkt: The number of input packets (summation of all network interfaces).
opkt: The number of output packets (as above).
err: The number of packets considered errors, both output and collusion errors (as above).
d0: Disk activity on the ``first'' disk. The names are useless but you can usually figure out which is which.
d1: Disk activity on the ``second'' disk.
d2: Disk activity on the ``third'' disk.
d3: Disk activity on the ``fourth'' disk.
ipg: Pages faulted in.
opg: Pages paged out (by pageout daemon or freebehind of sequential I/O).
intr: Device interrupts.
cs: Context switches.
usr: Percentage of CPU time spent executing user applications.
sys: Percentage of CPU time spent in the kernel.
idl: Percentage of CPU time spent idle.

BUGS

The rstat()ing hosts effect the timing enough that sometimes the numbers come out wrong. They are usually too large. Use larger intervals if you start getting bogus numbers.

Silently ignores unreachable hosts. Crashing hosts crashes the program.

Rstatd(8) doesn't dig out all the info I'd like. It would be cool if we could implement vmstat(8) and xperfmon(8) with rstatd(8).

(I'd refer you to rstat(3) to figure out the fields but that seems to be undocumented. Sigh.)

AUTHOR

Larry McVoy, Stanford University

This document was created by man2html, using the manual pages.
Time: 21:51:34 GMT, February 02, 2023