vmstat

[Monah Baki <mbaki_(_at_)_infoa_(_dot_)_com>]

Hi all,

I'm trying to troubleshoot a OBSD 3.1 running squid on sparc64 (784MB 
RAM with 3x18 scsi HDD). After 2 days, cache response becomes too slow 
and I need to run vmstat too see where the problem lies. First timme 
using vmstat for troubleshooting. Server is supporting 2000 users.

So far, the server responds with this:

procs   memory        page                    disks     faults     cpu
 r b w    avm    fre   flt  re  pi  po  fr  sr sd0 sd1   in    sy   cs 
us sy id
 1 1 0 421976 124464     6   0   0   0   0   0  15   6  835  1066   89 
19 22 59
 2 2 0 421976 124448    75   0   0   0   0   0  11  12  875   994   85 
18 22 60
 1 1 0 421976 124448     5   0   0   0   0   0  20   7  741   852   70 
17 24 59

What do I need to look for?
How do I tell if it's good or bad?
How do I correct it?

Thank you.