high loadavarages in linux

[chaitanya1982]

Hi,

I am facing a high load averages in one of my linux server
Please find the output of the top command.
top - 13:00:01 up 28 days, 22:15, 64 users, load average: 39.97, 37.06, 37.63
Tasks: 655 total, 1 running, 643 sleeping, 9 stopped, 2 zombie
Cpu(s): 3.2% us, 3.8% sy, 6.1% ni, 86.8% id, 0.1% wa, 0.0% hi, 0.0% si
Mem: 49452332k total, 49311644k used, 140688k free, 1194364k buffers
Swap: 16779884k total, 39452k used, 16740432k free, 30620228k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
7253 sid 15 0 8868 1712 948 S 5.7 0.0 0:22.19 top
10974 root 15 0 8864 1608 864 R 3.8 0.0 0:00.02 top
17523 sumanku 16 0 9124 2028 956 S 3.8 0.0 100:04.65 top
4226 pgadepa 18 0 58448 9060 380 D 1.9 0.0 3:50.57 du
21587 srudra 15 0 748m 462m 49m D 1.9 1.0 104:57.77 xpsgui
22921 awdhesh 15 0 38908 27m 1892 S 1.9 0.1 8:58.47 Xvnc
1 root 16 0 4772 512 424 S 0.0 0.0 2:16.27 init
2 root RT 0 0 0 0 S 0.0 0.0 0:03.60 migration/0
3 root 34 19 0 0 0 S 0.0 0.0 0:00.68 ksoftirqd/0
4 root RT 0 0 0 0 S 0.0 0.0 0:00.75 migration/1
5 root 34 19 0 0 0 S 0.0 0.0 0:00.24 ksoftirqd/1
6 root RT 0 0 0 0 S 0.0 0.0 0:00.56 migration/2
7 root 34 19 0 0 0 S 0.0 0.0 0:00.14 ksoftirqd/2
8 root RT 0 0 0 0 S 0.0 0.0 0:00.39 migration/3
9 root 34 19 0 0 0 S 0.0 0.0 0:00.12 ksoftirqd/3
10 root RT 0 0 0 0 S 0.0 0.0 0:00.43 migration/4
11 root 34 19 0 0 0 S 0.0 0.0 0:00.06 ksoftirqd/4
12 root RT 0 0 0 0 S 0.0 0.0 0:00.30 migration/5
13 root 34 19 0 0 0 S 0.0 0.0 0:00.14 ksoftirqd/5
14 root RT 0 0 0 0 S 0.0 0.0 0:00.34 migration/6
15 root 34 19 0 0 0 S 0.0 0.0 0:00.24 ksoftirqd/6
16 root RT 0 0 0 0 S 0.0 0.0 0:00.22 migration/7
17 root 34 19 0 0 0 S 0.0 0.0 0:00.10 ksoftirqd/7
18 root RT 0 0 0 0 S 0.0 0.0 0:01.37 migration/8
19 root 34 19 0 0 0 S 0.0 0.0 0:00.60 ksoftirqd/8
20 root RT 0 0 0 0 S 0.0 0.0 0:00.34 migration/9
21 root 34 19 0 0 0 S 0.0 0.0 0:00.08 ksoftirqd/9
22 root RT 0 0 0 0 S 0.0 0.0 0:01.41 migration/10
23 root 34 19 0 0 0 S 0.0 0.0 0:00.08 ksoftirqd/10
24 root RT 0 0 0 0 S 0.0 0.0 0:00.22 migration/11.

How to reduce the load averages. Please help me out from this.

Regards,
Chaitanya

[matonb]

Load Averages are just an indicator. If I remember rightly the figure is based on the number of running and waiting processes.

I did notice in your top output that there are 643 sleeping processes, which probably has something to do with the high load average.

Please use the code tags when pasting command output, it makes it much easier to read, thanks.

[chaitanya1982]

Hi
I found some process are in D state. How can I clear those process without reboot.


top -b -n 1 | awk '{if ($8 == "D"){count++;print}}END{print "No.of D state process are:"count}'
361 nitink 18 0 5384 672 540 D 0.0 0.0 0:00.72 kdesktop_lock
440 karthik 15 0 171m 65m 10m D 0.0 0.2 2:58.77 nautilus
533 seshadr 18 0 28720 1688 1204 D 0.0 0.0 0:00.73 login
563 vindhyan 15 0 194m 42m 10m D 0.0 0.1 1:13.35 nautilus
595 root 20 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
767 jignesh 15 0 109m 37m 9832 D 0.0 0.1 9:50.99 Xvnc
902 seshadr 16 0 103m 3188 2232 D 0.0 0.0 0:00.01 gdm-binary
1050 jignesh 15 0 224m 71m 10m D 0.0 0.2 5:06.00 nautilus
1164 pkum 15 0 55732 29m 4244 D 0.0 0.1 2:03.51 Xvnc
1841 sraman 15 0 96740 22m 4008 D 0.0 0.1 108:45.23 Xvnc
1867 root 20 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
1943 sraman 15 0 135m 13m 10m D 0.0 0.0 209:27.79 kdeinit
3157 mahmed 15 0 60140 32m 7832 D 0.0 0.1 18:05.36 Xvnc
4437 kapilu 18 0 100m 956 164 D 0.0 0.0 0:00.00 gnome-vfs-daemo
4603 kapilu 18 0 100m 956 164 D 0.0 0.0 0:00.00 gnome-vfs-daemo
4645 kapilu 18 0 100m 956 164 D 0.0 0.0 0:00.69 gnome-vfs-daemo
4669 kapilu 18 0 100m 956 164 D 0.0 0.0 0:00.04 gnome-vfs-daemo
4701 kapilu 18 0 100m 956 164 D 0.0 0.0 0:00.04 gnome-vfs-daemo
5345 nitink 15 0 49932 23m 3940 D 0.0 0.1 44:24.25 Xvnc
5440 nitink 15 0 73304 10m 8180 D 0.0 0.0 48:21.30 kdeinit
10090 rakeshh 15 0 40404 13m 5360 D 0.0 0.0 0:01.92 Xvnc
10314 badermd 15 0 98.7m 25m 7644 D 0.0 0.1 0:04.11 Xvnc
10481 karthik 15 0 63516 35m 6168 D 0.0 0.1 0:24.63 Xvnc
10954 karthik 15 0 172m 65m 10m D 0.0 0.2 3:27.89 nautilus
13151 rohitg 15 0 219m 66m 10m D 0.0 0.2 4:01.26 nautilus
13700 root 18 0 50032 132 48 D 0.0 0.0 0:00.07 lsof
13959 root 20 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
14233 root 21 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
14252 root 20 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
15323 root 18 0 49948 608 512 D 0.0 0.0 0:00.71 df
16964 nitink 15 0 0 0 0 D 0.0 0.0 0:01.57 kdeinit
23011 sraman 15 0 0 0 0 D 0.0 0.0 0:00.17 kdeinit
25145 preddyg 15 0 66348 39m 7016 D 0.0 0.1 2:28.08 Xvnc
26195 root 20 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
28029 root 18 0 50032 132 48 D 0.0 0.0 0:00.00 lsof
28481 sivaraj 15 0 58220 30m 7492 D 0.0 0.1 1:27.68 Xvnc
30750 seshadr 15 0 222m 69m 9.8m D 0.0 0.2 4:58.54 nautilus
31333 root 15 -4 29224 4520 184 D 0.0 0.0 0:00.83 automount
31651 root 18 0 49948 608 512 D 0.0 0.0 0:00.72 df
32545 vindhyan 15 0 115m 42m 8088 D 0.0 0.1 5:05.62 Xvnc
32741 karthik 15 0 63172 35m 6384 D 0.0 0.1 0:33.88 Xvnc
No.of D state process are:41

Thanks & regards,
Chaitanya

[matonb]

You can try killing the parent processes of the tasks in Uniterruptable Sleep, that may get rid of some of them.

However, you'll probably have to reboot to get rid of them all.

Please use the CODE tags # on the menu bar when posting output!
It's very difficult to read if it's unformatted.