incorrect value on vmstat

[Lewis Chang]

Hello,

I have a system with eight CPUs and I'm wondering how are the values of the cpu columns be calculated. Here is my output.

psrv3.lewis:/mes3/lewis> vmstat 3
procs memory page faults cpu
r b w avm free re at pi po fr de sr in sy cs us sy id
8 0 0 473332 1974293 0 0 0 0 0 0 0 0 58 1 100 0 100
8 0 0 473332 1974175 29 3 1 0 0 0 0 6539 5815 1106 100 0 100
8 0 0 471795 1974462 16 0 0 0 0 0 0 6158 5322 1078 100 0 100
8 0 0 471795 1974457 8 0 0 0 0 0 0 6912 5314 1130 100 0 100

Here is another output for each cpu.

hpsrv3.lewis:/mes3/lewis> vmstat -n 3
VM
memory page faults
avm free re at pi po fr de sr in sy cs
471459 1974482 0 0 0 0 0 0 0 0 58 1
CPU
cpu procs
us sy id r b w
44 3 53 8 0 0
42 3 55
41 4 55
42 3 55
43 2 54
47 2 50
47 2 51
46 2 52
471459 1974433 4 0 0 0 0 0 0 2007 4035 511
87 -0 13 8 0 0
55 3 41
98 1 1
57 1 42
105 0 -5
95 0 4
103 0 -3
99 0 1
470033 1974433 7 0 1 0 0 0 0 2046 4040 510
98 -0 3 8 0 0
63 5 33
100 -0 1
82 4 14
79 1 20
81 0 19
94 0 6
98 0 2

Can any one tell me how does vmstat get 100 on column us and id? Many thanks.

[Dietmar Konermann]

Hi!

Since you seem to get even values >100 I would suggest you install PHCO_25979 (assuming you are on 11.11).

The vmstat(1) derives the metrics from pstat() system call. Should be the same as for top(1).

Regards...
Dietmar.