[Users] wierd case of high load averages

[Aleksandar Ivanisevic]

Hi,

I really have no idea what to make of this. Although HN's load average
is almost zero, most of the containers have LA in high 3 digits range.

To make it even more interesting, LA fluctuates from sngle to triple
digits within a minute, below are three samples taken every 30
seconds. Note i.e. container 1078 that has a load of 0.06 and 30
seconds later 857.53.

Everything else is normal, containers are responsive, vmstat is all
zeros with 100% idle.

Any ideas?


# uptime
 16:09:18 up 12 days, 16:43,  1 user,  load average: 0.02, 0.05, 0.16
# for vz in $(vzlist -Ho veid); do printf "%20s\t%s\n" $vz "$(vzctl exec $vz uptime)"; done
                1043     16:09:20 up 12 days, 16:40,  0 users,  load average: 350.64, 11.94, 95.60
                1045     16:09:20 up 12 days, 16:40,  0 users,  load average: 660.90, 14.86, 97.17
                1047     16:09:20 up 12 days, 16:39,  0 users,  load average: 1021.99, 21.37, 99.81
                1049     16:09:20 up 12 days, 16:38,  0 users,  load average: 42.89, 31.91, 106.02
                1050     15:09:20 up 44 min,  0 users,  load average: 171.15, 0.16, 43.92
                1067     16:09:20 up 12 days, 16:38,  2 users,  load average: 57.28, 30.88, 108.88
                1068     16:09:20 up 12 days, 16:37,  0 users,  load average: 320.34, 67.89, 112.53
                1070     16:09:20 up 12 days, 16:37,  0 users,  load average: 505.05, 96.16, 126.54
                1072     16:09:20 up 7 days,  6:11,  0 users,  load average: 605.07, 7.20, 313.08
                1077     16:09:20 up 12 days, 16:36,  1 user,  load average: 553.71, 91.94, 121.71
                1078     16:09:20 up 12 days, 16:36,  0 users,  load average: 0.06, 182.17, 289.51
                1079     16:09:20 up 12 days, 16:35,  0 users,  load average: 787.56, 100.48, 126.54
                1080     16:09:20 up 12 days, 16:35,  0 users,  load average: 7.75, 114.44, 130.67
                1081     16:09:20 up 12 days, 16:35,  0 users,  load average: 0.97, 109.61, 130.68
                1181     16:08:56 up  1:25,  0 users,  load average: 41.10, 22.78, 344.78
# for vz in $(vzlist -Ho veid); do printf "%20s\t%s\n" $vz "$(vzctl exec $vz uptime)"; done
                1043     16:09:36 up 12 days, 16:41,  0 users,  load average: 263.67, 9.29, 93.39
                1045     16:09:37 up 12 days, 16:40,  0 users,  load average: 505.65, 12.17, 94.97
                1047     16:09:37 up 12 days, 16:40,  0 users,  load average: 786.75, 18.37, 97.57
                1049     16:09:37 up 12 days, 16:39,  0 users,  load average: 24.53, 28.39, 103.68
                1050     15:09:37 up 44 min,  0 users,  load average: 126.82, 989.02, 42.75
                1067     16:09:37 up 12 days, 16:38,  2 users,  load average: 35.73, 27.41, 106.50
                1068     16:09:37 up 12 days, 16:38,  0 users,  load average: 240.52, 62.61, 110.09
                1070     16:09:37 up 12 days, 16:37,  0 users,  load average: 384.31, 89.49, 123.87
                1072     16:09:37 up 7 days,  6:12,  0 users,  load average: 462.18, 4.89, 307.43
                1077     16:09:37 up 12 days, 16:36,  1 user,  load average: 422.19, 85.48, 119.12
                1078     16:09:37 up 12 days, 16:36,  0 users,  load average: 857.53, 171.23, 284.21
                1079     16:09:37 up 12 days, 16:36,  0 users,  load average: 604.25, 93.60, 123.87
                1080     16:09:37 up 12 days, 16:35,  0 users,  load average: 1021.17, 106.87, 127.93
                1081     16:09:37 up 12 days, 16:35,  0 users,  load average: 858.46, 102.28, 127.94
                1181     16:09:13 up  1:25,  0 users,  load average: 24.24, 19.95, 338.70
# for vz in $(vzlist -Ho veid); do printf "%20s\t%s\n" $vz "$(vzctl exec $vz uptime)"; done
                1043     16:10:01 up 12 days, 16:41,  0 users,  load average: 159.03, 5.09, 89.77
                1045     16:10:01 up 12 days, 16:40,  0 users,  load average: 319.13, 7.90, 91.36
                1047     16:10:01 up 12 days, 16:40,  0 users,  load average: 504.32, 13.60, 93.89
                1049     16:10:01 up 12 days, 16:39,  0 users,  load average: 2.17, 22.82, 99.84
                1050     15:10:01 up 45 min,  0 users,  load average: 73.31, 907.19, 40.82
                1067     16:10:01 up 12 days, 16:39,  2 users,  load average: 9.55, 21.92, 102.58
                1068     16:10:01 up 12 days, 16:38,  0 users,  load average: 144.46, 54.29, 106.07
                1070     16:10:01 up 12 days, 16:38,  0 users,  load average: 239.20, 79.01, 119.49
                1072     16:10:01 up 7 days,  6:12,  0 users,  load average: 290.49, 1.20, 298.17
                1077     16:10:01 up 12 days, 16:37,  1 user,  load average: 264.15, 75.32, 114.86
                1078     16:10:01 up 12 days, 16:37,  0 users,  load average: 550.61, 154.11, 275.55
                1079     16:10:02 up 12 days, 16:36,  0 users,  load average: 384.09, 82.79, 119.49
                1080     16:10:02 up 12 days, 16:36,  0 users,  load average: 658.76, 95.00, 123.44
                1081     16:10:02 up 12 days, 16:36,  0 users,  load average: 551.57, 90.78, 123.45
                1181     16:09:38 up  1:26,  0 users,  load average: 3.68, 15.45, 328.74


-- 
Ivanisevicu, pokusaj se za promjenu drzati teme i replicirati na ono o 
cemu sugovornik pise. Osim sto je tvoje pracenje "koji me vrag zapravo 
muci" off-topic na Linux grupi, radi se i o zabadanju njuske u moje 
privatne stvari. Vec si pokazao nedostatak dobrog odgoja no ovo granici 
sa sociopatijom. -- Mladen Gogala, hr.comp.os.linux