On Saturday, February 12, 2011 12:43:53 pm M. Fioretti wrote:
On Sat, Feb 12, 2011 12:27:55 PM -0600, Rick Sewill (rsewill@gmail.com)
wrote:
Could you show the output of iostat -x 1, not iostat -x 1 | egrep -i 'device|sda' please?
Sure, sorry, here you go (this is with Firefox open, right now)
Linux 2.6.35.10-74.fc14.x86_64 (polaris.localdomain) 02/12/2011
_x86_64_ (2
CPU)
avg-cpu: %user %nice %system %iowait %steal %idle 28.93 0.00 3.23 0.69 0.00 67.15
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.76 12.23 1.72 2.07 96.94 111.76 54.97 0.10 26.58 4.13 1.57 dm-0 0.00 0.00 2.45 13.98 96.65 111.76 12.68 2.18 132.68 0.95 1.57 dm-1 0.00 0.00 0.01 0.00 0.09 0.00 8.00 0.00 5.45 3.18 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 48.76 0.00 0.50 0.00 0.00 50.75
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 16.58 0.00 1.01 0.00 0.00 82.41
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 0.00 0.00 19.00 0.00 152.00 8.00 0.01 0.79 0.11 0.20 dm-0 0.00 0.00 0.00 19.00 0.00 152.00 8.00 0.01 0.79 0.11 0.20 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 4.46 0.00 0.99 4.95 0.00 89.60
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 27.00 0.00 9.00 0.00 272.00 30.22 0.07 7.67 7.67 6.90 dm-0 0.00 0.00 0.00 34.00 0.00 272.00 8.00 0.10 2.82 2.03 6.90 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 4.50 0.00 1.00 0.00 0.00 94.50
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 12.87 0.00 0.99 0.00 0.00 86.14
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
avg-cpu: %user %nice %system %iowait %steal %idle 39.30 0.00 0.50 0.00 0.00 60.20
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Is there any correlation between avg-cpu %user and Device sda wsec/s writes?
Is there a burst of %user cpu activity followed by a burst of wsec/s writes?
If the system is doing so little, I'd expect less %user cpu activity. Since the system is 2 CPU, does 48% means one cpu ran solid for a second?
Someone help us...I know there is a command to show open files, lsof. Does that command include a way to find out disk activity per file or is there another command that can find out disk activity per file? I'm hoping, if we identify the file(s) with disk activity, we might identify the service/application/kernel feature that is hogging the cpu.