High CPU load but top processes in top all around 0%

Question

We have a server with unusual high load and cpu util, but we can't figure out why. When we run top all the procs seem to be very low cpu.

http://cl.ly/2d1g0K3q261r0R0K3e35

Is there a better way to look for what is causing this?

I should note I am on Ubuntu 10.04 – Ben May 22 '11 at 04:55 — Ben, May 22 '11 at 04:55

score 7 · Answer 1 · answered May 22 '11 at 06:40

7

Load is a measure of the workload a system has had on a 1, 5 and 15 minute basis.

The most common misconception is that Load Average is purely connected to the CPU usage of a system.
Load does however incorporate additional measurements such as CPU waiting for I/O which I think is your issue.

Based on the image I'm guessing you ran out of memory and started swapping data to disk.

A simple free -m will tell you how much RAM and swap is used.
The interesting column is the free column besides -/+ buffers/cache.
If it's close to zero you've run out of RAM and should act accordingly.

answered May 22 '11 at 06:40

Mark

740
5
5

1

+1 for swap. You could use `iotop` or `vmstat` to try and see if it is I/O. – Axel Knauf May 22 '11 at 09:32
1

-/+ buffers/cache: 2933 14586 – Ben May 22 '11 at 15:06

score 4 · Answer 2 · answered May 22 '11 at 06:25

Noticed that the load average is quite high (68, wow). Is it possible that there are a lot of processes which takes up a little bit of CPU, thus add up consuming all CPU time? Maybe, those processes just start and finish very quickly thus top cannot capture the existence of them, you may try to see if atop can see that or not.

score 4 · Answer 3 · edited May 22 '11 at 15:08

4

Try

iotop

IO was it for me most of the time.

edited May 22 '11 at 15:08

splattne

28,348
19
97
147

answered May 22 '11 at 14:29

MarcoHager

41
1

grs · Accepted Answer · 2011-05-22T16:55:53.183

4

I think this bug is your case. From what I see from the output, you have enough memory (note the cached 14 GB or so), no I/O issues, but you have xen-related processes running. This make me think it is a bug.

edited May 22 '11 at 16:55

answered May 22 '11 at 16:13

grs

2,235
6
28
36

It turned out to be a bug in Ubuntu 10.04 and the hardware that AWS was using – Ben Jun 03 '11 at 07:14

score 2 · Answer 5 · answered May 22 '11 at 04:49

2

Try using:

top -o cpu

The -o flag will force top to order the processes by CPU usage in descending order.

answered May 22 '11 at 04:49

Bandit

201
1
3

when i ran `top -o cpu` I got "top: unknown argument 'o'" – Ben May 22 '11 at 04:54
Ok, try running top and hitting `o` while it is running. It should ask you for a primary key. Type `cpu` and hit enter. – Bandit May 22 '11 at 06:10
1

even when sorted by cpu there is nothing at the top of the list above 1% and there are only a handful (2-3) at any given time. The rest are 0% – Ben May 22 '11 at 06:40
1

On Centos 7.2 at least, the correct command to do this is `top -o %CPU` – siliconrockstar Aug 15 '16 at 15:02

score 2 · Answer 6 · answered May 22 '11 at 14:21

2

It could be locked files on nfs or any other thing that locks a file that another process needs access to

could also be missed configured service with too many threads active

answered May 22 '11 at 14:21

John

391
1
4

score 1 · Answer 7 · edited Apr 13 '17 at 12:37

Looks like CPU usage is coming from a thread. top seems to not take this into account. I recently saw this on a mysql server. there are running INSERT statements but I was unable to get the new rows with SELECT because some thread of mysqld was updating the table index. top shows 100% user load on one core but every process including mysqld was an 0.0% CPU. hours later the same SELECT provided the expected result set.

High CPU load but top processes in top all around 0%

7 Answers7