Netapp sysstat disk util exceed 100%

[FENDERAIYOH]

Hi all,

I would like to know why disk util will exceed 100%? Below is the  partial sysstat data that i have

CPU   NFS  CIFS  HTTP   Total    Net kB/s   Disk kB/s     Tape kB/s Cache Cache  CP   CP Disk    FCP iSCSI   FCP  kB/s iSCSI  kB/s

                                  in   out   read  write  read write   age   hit time  ty util                 in   out    in   out

  5%     0     0     0     651  9792  4706   4148      0     0     0     8s  35%   0%  -  100%      0   651     0     0  9191  4252

  5%     0     0     0     682  9557  4883   4280      0     0     0     8s  30%   0%  -  100%      0   682     0     0  8962  4424

  5%     0     0     0     677  9836  4696   4148      0     0     0     8s  31%   0%  -  100%      0   677     0     0  9232  4239

12%     0     0     0     659  8654  4487   4164   3080     0     0     8s  27%  14%  Tn 1008%      0   659     0     0  8110  4067

  8%     0     0     0     475  6398  3313   4168  26248     0     0     8   45% 100%  :f 100%      0   475     0     0  5997  3002

  7%     0     0     0     513  7189  3835   3544  31656     0     0     8   30% 100%  :f 100%      0   513     0     0  6738  3482

Regards,

fender

[Darkstar]

This is normal and has to do with the way the filer calculates the util% metric (it's not the average over all disks but only the disk with the highest utilization, for example). This is only a display problem and it is even documented (either in the man-page for the sysstat command, or in the admin guides, don't remember where)

You can ignore these bogus lines. However, having a disk utilization of 100% for longer periods is in itself a problem that you should look into. You might have a hotspot disk or some other problem

-Michael

[WICKEDSHARK]

Michael is correct sysstat only measures the highest number achieved on a single disk in the controller. This is not the average of all disks. Again as MIchael stated having 100% for long period of times is not a good thing. There is a better way to do disk monitoring for utilization of disks using advanced privilege sessions.