High load

fmen

Verified User
Joined
Mar 14, 2009
Messages
26
Hello
I have a server VPS with this load:

20:44:26 up 38 min, 2 users, load average: 14.10, 14.78, 11.49
[root@web001 ~]# uptime
20:44:40 up 38 min, 2 users, load average: 13.31, 14.57, 11.47

It is a high load, how can I check what the problem is?
 

Dravu

Verified User
Joined
Sep 9, 2007
Messages
324
Use the 'top' command and look at the load of the first few entries. Might be a run-away script.
 

fmen

Verified User
Joined
Mar 14, 2009
Messages
26
Code:
top - 21:01:31 up 55 min,  2 users,  load average: 20.15, 16.79, 14.25
Tasks: 135 total,  18 running, 114 sleeping,   0 stopped,   3 zombie
Cpu(s):  0.6%us,  0.4%sy,  0.0%ni, 98.3%id,  0.8%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   2097152k total,   422968k used,  1674184k free,        0k buffers
Swap:        0k total,        0k used,        0k free,        0k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 8061 apache    16   0 30968  16m 3796 D  0.3  0.8   0:00.08 httpd
11734 root      18   0  8872 2048 1640 R  0.3  0.1   0:00.01 exim
    1 root      18   0  2068  652  560 S  0.0  0.0   0:00.13 init
 2002 root      15   0  4524 1280 1120 S  0.0  0.1   0:00.00 mysqld_safe
 2044 mysql     15   0 35288  14m 3236 S  0.0  0.7   0:00.03 mysqld
 2047 mysql     15   0 35288  14m 3236 S  0.0  0.7   0:00.01 mysqld
 3072 mysql     20   0 35288  14m 3236 S  0.0  0.7   0:00.00 mysqld
 3073 mysql     18   0 35288  14m 3236 S  0.0  0.7   0:00.00 mysqld
 3074 mysql     16   0 35288  14m 3236 S  0.0  0.7   0:00.00 mysqld
 3076 mysql     23   0 35288  14m 3236 S  0.0  0.7   0:00.00 mysqld
 3081 mysql     18   0 35288  14m 3236 R  0.0  0.7   0:00.05 mysqld
 3082 mysql     18   0 35288  14m 3236 R  0.0  0.7   0:00.10 mysqld
 3083 mysql     18   0 35288  14m 3236 S  0.0  0.7   0:00.00 mysqld
 3085 mysql     24   0 35288  14m 3236 R  0.0  0.7   0:00.00 mysqld
 3107 root      16   0  3136  544  444 S  0.0  0.0   0:00.04 da-popb4smtp
 3159 nobody    15   0  8888 1340  948 S  0.0  0.1   0:00.34 directadmin
 3198 root      15   0 32516  25m  684 S  0.0  1.3   0:00.00 spamd
 3360 mail      18   0  8880 1036  620 S  0.0  0.0   0:00.00 exim
 3433 dovecot   18   0  4892 1828 1508 S  0.0  0.1   0:00.00 pop3-login
 3483 named     15   0 69444 3444 2060 S  0.0  0.2   0:00.08 named
 3485 named     18   0 69468 3464 2056 S  0.0  0.2   0:00.07 named
 3577 apache    15   0 34928  20m 4092 S  0.0  1.0   0:00.44 httpd
 3584 apache    18   0 30740  16m 4080 D  0.0  0.8   0:00.10 httpd
 3668 apache    15   0 31256  17m 4060 S  0.0  0.8   0:00.09 httpd
 3669 apache    15   0 47980  31m 4412 S  0.0  1.5   0:01.37 httpd
 3697 root      15   0 10140 2816 2276 S  0.0  0.1   0:00.09 sshd
 3962 dovecot   18   0  4888 1824 1508 S  0.0  0.1   0:00.00 pop3-login
 4031 root      15   0  4528 1484 1248 S  0.0  0.1   0:00.01 bash
 4079 dovecot   18   0  4888 1824 1508 S  0.0  0.1   0:00.00 pop3-login
 4081 dovecot   18   0  4892 1828 1508 S  0.0  0.1   0:00.00 pop3-login
 5242 root      15   0  1780  600  524 S  0.0  0.0   0:00.00 tload
 5659 root      18   0 32516  25m  600 S  0.0  1.2   0:00.00 spamd
 6036 apache    18   0 48408  31m 4116 D  0.0  1.5   0:00.55 httpd
 7239 root      18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7410 root      18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7452 root      18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7548 mailprox  18   0  5852 1500  872 S  0.0  0.1   0:00.00 crond
 7559 yourweb   18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7604 yourweb   18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7606 nathanie  18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7607 iwebhost  15   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 7664 duser     18   0  5852 1488  868 S  0.0  0.1   0:00.00 crond
 9242 apache    15   0 30220  16m 3988 S  0.0  0.8   0:00.04 httpd
 9285 apache    15   0 29424  15m 3992 D  0.0  0.7   0:00.03 httpd
 9293 apache    18   0 30744  16m 3660 S  0.0  0.8   0:00.08 httpd
 9459 dovecot   15   0     0    0    0 Z  0.0  0.0   0:00.00 pop3-login <defunct>
 9527 apache    18   0 30504  16m 3668 S  0.0  0.8   0:00.06 httpd
10056 root      15   0  2404  972  868 S  0.0  0.0   0:00.00 bash
10119 root      18   0  2400  972  864 S  0.0  0.0   0:00.00 sh
11626 nathanie  18   0 21460 8064 4112 D  0.0  0.4   0:00.03 php
11669 root      18   0  2404 1084  916 S  0.0  0.1   0:00.00 bfd
11671 yourweb   18   0  2400  980  868 S  0.0  0.0   0:00.00 sh
11731 duser     18   0   960  120   96 R  0.0  0.0   0:00.00 sh
11735 iwebhost  16   0  2404  980  868 S  0.0  0.0   0:00.00 sh
This is as far as I can get.
 

fmen

Verified User
Joined
Mar 14, 2009
Messages
26
My server just turned off.. I have rebooted it, but what can be the problem??
 

tillo

Verified User
Joined
Oct 28, 2007
Messages
862
Location
Switzerland
Try "htop" instead of "top". You can distinguish threads from processes, it works better and is far simpler to use.
A load of 15 means that in the last minute (and 5 and 15 minutes) 14 out of 15 processes/threads have in average been waiting for the CPU to be free instead of using it directly.
An ideal value for the load is less than 1, which means that in average all processes have had access to the CPU and there has been no CPU queue.
Now, if you don't see any process using 100% of the CPU even with htop, and given that linux adds to the load number also all the processes waiting for IO... the conclusion is that, very probably, you have a process using up all the IO resources of your disks or that you have a failing disk controller driver.
 

fmen

Verified User
Joined
Mar 14, 2009
Messages
26
Try "htop" instead of "top". You can distinguish threads from processes, it works better and is far simpler to use.
A load of 15 means that in the last minute (and 5 and 15 minutes) 14 out of 15 processes/threads have in average been waiting for the CPU to be free instead of using it directly.
An ideal value for the load is less than 1, which means that in average all processes have had access to the CPU and there has been no CPU queue.
Now, if you don't see any process using 100% of the CPU even with htop, and given that linux adds to the load number also all the processes waiting for IO... the conclusion is that, very probably, you have a process using up all the IO resources of your disks or that you have a failing disk controller driver.
I will install htop and try to run it. But what if there is not a 100% process?? It is on a VPS so could it be a problem from the hardware node?
 

fmen

Verified User
Joined
Mar 14, 2009
Messages
26
Thank you for your help but as I thought it is a problem on the host node. My isp told me that it was my VPS that was the problem, but is seems to be resolved now.

htop no 100% processes
 
Top