LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Server (https://www.linuxquestions.org/questions/linux-server-73/)
-   -   how to reduce load average linux (https://www.linuxquestions.org/questions/linux-server-73/how-to-reduce-load-average-linux-930841/)

pavez01 02-23-2012 04:22 AM

how to reduce load average linux
 
I am using CentOS 5.3 As a Mail Server. But System Load Average is too much High. For that reason System is responding very slow.
TOP output:
top - 16:21:30 up 4:39, 1 user, load average: 77.91, 87.64, 84.56
Tasks: 641 total, 2 running, 628 sleeping, 0 stopped, 11 zombie
Cpu(s): 4.6%us, 0.9%sy, 0.0%ni, 33.6%id, 60.9%wa, 0.0%hi, 0.1%si, 0.0%st
Mem: 8309316k total, 2556448k used, 5752868k free, 13320k buffers
Swap: 8385888k total, 0k used, 8385888k free, 395804k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
18570 apache 23 0 36804 12m 2560 S 2.3 0.2 0:00.07 httpd
16005 apache 16 0 49964 25m 3088 S 2.0 0.3 0:01.02 httpd
17127 apache 15 0 46368 22m 3100 S 1.7 0.3 0:00.69 httpd
17164 apache 15 0 38640 14m 2948 S 1.7 0.2 0:00.26 httpd
18234 apache 16 0 34160 9m 2756 S 1.7 0.1 0:00.09 httpd
13033 apache 15 0 41348 17m 3040 S 1.3 0.2 0:00.48 httpd
15854 apache 15 0 34652 10m 3000 S 1.3 0.1 0:00.30 httpd


What should i do now to reduce Load Average???

Please help.

deep27ak 02-23-2012 05:02 AM

check out for no. of sleeping and zombie process in your server for which I would advice to kill all not required running processes

Code:

Tasks: 641 total, 2 running, 628 sleeping, 0 stopped, 11 zombie

devUnix 02-23-2012 05:19 AM

You can reboot your system once to see if all looks okay afterwards.

Killing Zombie and Sleeping Processes in that order can also help.

Use 'top' and 'ps aux' commands to figure what processes are causing the Load Average to go high. (You already know those commands.)

pavez01 02-23-2012 06:22 AM

Thanks All.

I have restarted Server. But within 10 minutes Load Average became high. :-(

TenTenths 02-23-2012 06:39 AM

Ok, you've got 60.9%wa which would indicate that the disk is being heavily used. Either you're getting a LOT of mail in or out, or you have an application / website that is making your disk very busy. Try investigating to see if that is the case.

pavez01 02-23-2012 06:58 AM

Dear TenTenths, You are right. A huge mail transaction has been done by the Server. I have administrated the server since 2009. But last couple of weeks i am facing the problem..

TenTenths 02-23-2012 07:00 AM

It could also mean a disk fault causing it to have to read/write/error-correct.

pavez01 02-23-2012 07:04 AM

Could be. Server is using local HDD and there is no Amber in HDD. How could i know if there any error?

TenTenths 02-23-2012 07:05 AM

check /var/log/messages and see if there's anything in there that could indicate a problem.

suprstar 02-23-2012 07:05 AM

Possibly a degraded RAID, those sometimes have to work extra extra hard to get the same job done.

apt-get install iotop

and you can see the heavy hitters on the disk i/o.


All times are GMT -5. The time now is 08:51 AM.