Monitoring server load

[erick_paper]

Hi. I am getting consistent 2.8 kind of server loads, with all of my 8 CPUs occupied all the time.

Is this too high for high traffic websites or should I let it be? What's an "alarm" level of server load? Is there a way to send myself an email when the server reaches a threshold or perhaps restart Apache automagically etc?

Would love to hear some tips from people who run high traffic sites. Thanks!

 

[cPanelDavidG]

Hi. I am getting consistent 2.8 kind of server loads, with all of my 8 CPUs occupied all the time.

Is this too high for high traffic websites or should I let it be? What's an "alarm" level of server load? Is there a way to send myself an email when the server reaches a threshold or perhaps restart Apache automagically etc?

Would love to hear some tips from people who run high traffic sites. Thanks!

A general rule of thumb is that server load should be less than the number of CPUs you have. So if you have 8 CPUs, your server load should be less than 8. So 2.8 isn't neccessarily bad for a server with 8 CPUs.

From there, different people use different criteria for what's tolerable vs. alarming. I'm sure others on these forums will have advice to share with regards to this and running high-traffic sites .

 

[AndyReed]

Is this too high for high traffic websites or should I let it be? What's an "alarm" level of server load?

This article might help: http://servertune.com/kbase/entry/67/

 

[Batmunkh]

my server load is higher than 50

my server load is higher than 50. How can i optimize it? what is the max load point? my server has 4 CPUs. always load higher than 40 is ok?? is it bad?

 

[ChrisRHS]

my server load is higher than 50. How can i optimize it? what is the max load point? my server has 4 CPUs. always load higher than 40 is ok?? is it bad?

That is very high, and needs to be investigated as to what is causing the load to be so high.

 

[Batmunkh]

That is very high, and needs to be investigated as to what is causing the load to be so high.

what should i check... when i stop mysql service it gets lower and normal. but i can't find mysql log files...

 

[ChrisRHS]

Watch your top command in ssh, see which processes are taking up the resources. It sounds like you may have high mysql usage, so I would suggest you start tweaking the my.cnf, and see if you can lower the load.

 

[Batmunkh]

i have re-compiled easyapache3. its average is now 10. any tips and tweaks please... also thank you for replying ....

top command result is:

top - 23:23:22 up 2 days, 5:24, 3 users, load average: 15.42, 14.51, 19.85
Tasks: 319 total, 18 running, 287 sleeping, 11 stopped, 3 zombie
Cpu(s): 73.3%us, 19.5%sy, 0.0%ni, 6.3%id, 0.1%wa, 0.0%hi, 0.8%si, 0.0%st
Mem: 6105748k total, 4961296k used, 1144452k free, 155824k buffers
Swap: 12289716k total, 158632k used, 12131084k free, 3847840k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
17084 mysql 15 0 576m 79m 4172 S 24 1.3 26:32.32 /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/cp.alguu.net.pid --skip-external-locking
20679 azmn 17 0 0 0 0 Z 4 0.0 0:00.13 [php5] <defunct>
20737 azmn 16 0 132m 24m 6392 R 4 0.4 0:00.12 php5
20732 azmn 16 0 130m 22m 6196 R 4 0.4 0:00.11 php5
20734 azmn 16 0 130m 22m 6196 S 4 0.4 0:00.11 php5
20733 azmn 17 0 128m 20m 5980 R 3 0.3 0:00.10 php5
20735 azmn 16 0 127m 18m 5492 R 3 0.3 0:00.09 php5
20729 azmn 16 0 124m 16m 4500 R 3 0.3 0:00.08 php5
20742 azmn 16 0 123m 16m 4472 R 2 0.3 0:00.05 php5
....
.....