2

This is the result of iostat

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          13.65    0.00    9.98    3.86    2.22   70.29

Device:         rrqm/s   wrqm/s     r/s     w/s    rkB/s    wkB/s avgrq-sz avgqu-sz   await  svctm  %util
xvdap1            0.00     9.00    0.00   14.40     0.00   140.00    19.44    27.48 5991.25  46.32  66.70
xvdap2            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
           4.19    0.00    3.93    6.69    0.03   85.15

Device:         rrqm/s   wrqm/s     r/s     w/s    rkB/s    wkB/s avgrq-sz avgqu-sz   await  svctm  %util
xvdap1            0.00     3.00    0.00    0.00     0.00     0.00     0.00    19.08    0.00   0.00  45.44
xvdap2            0.00     0.00    0.00    0.00     0.00     0.00     0.00     0.00    0.00   0.00   0.00

iostat await sucks time is 7000. Does that mean process need to wait for 7 seconds before writing?

  • just so we are on the same page, you are talking about the 5.9 seconds, not 7, right? – Thorsten Staerk Apr 20 '14 at 9:00
0

Yes. At least one process has waited for 5.9s. I saw the same thing (5.3 seconds). It seemed to be a broken fibrechannel path that was not detected as broken; it worked (aka the symptom went away) after disabling one port on the fibrechannel switch: https://serverfault.com/questions/575447/server-hangs-randomly-without-log-entries

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.