vmstat

[David Willams]

kthr memory page disk faults cpu
r b w swap free re mf pi po fr de sr m0 m1 m5 m6 in sy cs us sy id
0 0 0 22182880 6451048 61 183 11 12 12 0 0 0 22 2 1 971 2316 743 1 1 99

Does the high number of faults in the vmstat really matter

[Shibin_2]

Yes. You need to find out, what processes are trying to utilize.

[David Willams]

but the system idle time is very high and mem. utilization is very very low...

faults
-------
in sys cs
1340 28765 4657


What does the above mean and how critical it could be.

What are the other factors that need checking.

[Hakki Aydin Ucar]

Hi,
if you have glance, you can narrow down problem easily.

OR you can go ;
http://forums13.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1288943634877+28353475&threadId=119443

and pick up html docs from JRF's post they are very useful.

[Shibin_2 · Accepted solution]

The section fault is not showing about your page faults ( memory ). But it shows about your VM system level faults.

in -> Device interrupts
sy -> System calls
cs -> CPU content switch rate.

Here, you have high number of system calls, and very low device interrupts and CPU switching.

You cannot simply depends on vmstat output. You may have to look for some other utilities to find out, what raised so much system calls. Whether it is from user generated or something else. You can use sar or glance to find out that.

Second it depends on your system load. I think, it is reasonable value. I have seen many production environment, sy is always higher.

[David Willams]

thanks