Hard drive performance question

Jason Lieurance jason at vipersystems.biz
Tue Aug 5 13:40:07 PDT 2003


Hello,

I have FreeBSD 4.7 server running a qmail-imap-squirrelmail email server and a
apache 1.3.27 web server hosting 6 virtual domains that don't get a lot of
traffic.

Hardware is:

P3 933MHz CPU
512 DDR ram
Fujitsu MAN3184MP 18.2 GB U160 10K rpm 8MB buffer 2.9 ms access Hard drive Adaptec
29160LP U160 SCSI controller

When I run vmstat, I get:

procs      memory      page                    disks     faults      CPU
 r b w     avm    fre  flt  re  pi  po  fr  sr da0 md0   in   sy  cs us sy id
 0 14 0  184016  40900   26   0   0   0  31   0   0   0  245  839  28  1  1 98
 0 14 0   86940  40896    7   0   0   0  15   0  12   0  242  115  21  0  0 99
 0 14 0   81460  40896    6   0   0   0  12   0  10   0  239   98  18  0  0 100
 0 14 0   85172  40896    7   0   0   0  15   0  14   0  243  116  21  0  1 99
 0 14 0  190092  40896   32   0   0   0  37   0  10   0  250  220  35  0  0 100
 0 14 0  190092  40896    7   0   0   0  17   0  15   0  244  114  21  0  0 100
 0 14 0  189720  40896    7   0   0   0  15   0  12   0  241  116  21  0  0 100
 0 14 0  189720  40896    6   0   0   0  12   0  10   0  239   96  18  0  0 100
 0 14 0   86860  40896    7   0   0   0  15   0  12   0  241  117  21  0  0 99
 1 14 0   86860  40912    6   0   0   0  13   0  11   0  241  100  18  0  0 100

I've read this is a disk access problem where process are waiting on the disk. As
you can see this is a constant but I thought the disk system I have was sufficient.
Any thoughts, analysis, or other is appreciated, thanks.

--
Jason





More information about the freebsd-questions mailing list