VMSTAT

[robert3975]

HI guys

just doing a good check around my box is when I done a vmstat

procs memory page disk faults cpu
r b w swap free re mf pi po fr de sr m0 m1 m2 m3 in sy cs us sy id
0 0 0 4954160 966440 0 0 0 0 0 0 0 0 0 0 0 4294967196 0 0 -40 -7 -29

There is a rather large figure in faults device interupt as u can see however if I do vmstat 5

it goes down to 500 ish

Is this anything to worry about and if so how do I veiw this in more detail and what do I do ?

Big thanks

 

[alray]

The first line output from vmstat, iostat etc. is an average since the box was last booted, subsequent lines are "as now".

HTH

Ray

 

[robert3975]

Yes I'm stupid thanks ray

but what I meant to say was what is considered to be an unacceptable level and how would you go about fixing it?

 

[marrow]

Hello Robert,

This is just FYI and sometimes it helps to look at other systems. We're using Sun_fire V240 & V210's...

I think maybe you need to look at your CPU usage. On one of our busier systems if CPU drops to well under 50% available and Run Queue starts to have occasional jobs waiting then the faults in turn increase..

r b w swap free re mf pi po fr de sr s0 s1 s2 s3 in sy cs us sy id
0 0 0 1355152 583008 23 1913 68 12 12 0 0 2 0 13 0 466 819232 471 37 6 57
0 0 0 1360448 583136 11 1454 0 24 24 0 0 3 0 2 0 389 708035 428 29 6 66
4 0 0 1331536 581944 49 2921 0 8 8 0 0 3 0 2 0 402 1251997 449 49 12 40
0 0 0 1360448 583128 0 0 0 0 0 0 0 1 0 0 0 317 2358 360 0 0 100

When system clears faults drop.

On a quiet system...looks like this....
r b w swap free re mf pi po fr de sr m0 m3 m4 m5 in sy cs us sy id
0 0 0 2537944 1739648 0 0 0 0 0 0 0 0 0 0 0 305 335 294 0 0 100
0 0 0 2537944 1739648 28 28 0 0 0 0 0 0 0 0 0 296 564 302 0 0 100
0 0 0 2537944 1739648 0 0 0 0 0 0 0 0 0 0 0 291 491 309 0 1 99

Above systems are for testing only. As I said it's useful to look at other systems, in our case I don't think we've ever worried about "faults" but have been more interested in rbw swap/free & sr columns to highlight memory/cpu shortages.

Bye

Marrow