Portal Home > Knowledgebase > Articles Database > huge load on our server
huge load on our server
Posted by mixmox, 09-11-2010, 03:16 AM |
hello every one.
we have a dedicate server which host 1 music website.
this week we have huge load 21 15 29 and ... .
How to reduce load?
we use cent os + directadmin + apache
|
Posted by mugo, 09-11-2010, 04:35 AM |
That very much depends on where the bottleneck is. Why don't you dump the outputs of
top
free -m
to begin with. If you have substantial visitors, or even a few, hitting streamed or downloaded music files, that alone can drive up at least your bandwidth usage.
What is your port speed? If you are capped at, say, 10M, just a few cable-modem / DSL users could bring you to a stand-still while streaming or downloading large files.
The more info you give, the better quality answer you will receive.
|
Posted by mixmox, 09-11-2010, 09:06 AM |
root@da ~]# free -m
total used free shared buffers cached
Mem: 2017 1785 232 0 163 1089
-/+ buffers/cache: 532 1485
Swap: 2047 0 2047
|
Posted by mixmox, 09-11-2010, 09:08 AM |
top - 09:06:45 up 2 days, 3:49, 2 users, load average: 0.00, 0.29, 0.66
Tasks: 119 total, 2 running, 112 sleeping, 5 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni, 99.8%id, 0.0%wa, 0.2%hi, 0.0%si, 0.0%st
Mem: 2065948k total, 1828336k used, 237612k free, 167032k buffers
Swap: 2096472k total, 116k used, 2096356k free, 1116016k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1 root 15 0 2156 640 552 S 0.0 0.0 0:00.46 init
2 root RT -5 0 0 0 S 0.0 0.0 0:04.11 migration/0
3 root 34 19 0 0 0 S 0.0 0.0 0:00.03 ksoftirqd/0
4 root RT -5 0 0 0 S 0.0 0.0 0:00.00 watchdog/0
5 root RT -5 0 0 0 S 0.0 0.0 0:03.74 migration/1
6 root 34 19 0 0 0 S 0.0 0.0 0:00.03 ksoftirqd/1
7 root RT -5 0 0 0 S 0.0 0.0 0:00.00 watchdog/1
8 root 10 -5 0 0 0 S 0.0 0.0 0:00.01 events/0
9 root 10 -5 0 0 0 S 0.0 0.0 0:00.02 events/1
10 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 khelper
11 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 kthread
15 root 10 -5 0 0 0 S 0.0 0.0 0:00.64 kblockd/0
16 root 10 -5 0 0 0 S 0.0 0.0 0:06.47 kblockd/1
17 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 kacpid
122 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 cqueue/0
123 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 cqueue/1
126 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 khubd
128 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 kseriod
195 root 24 0 0 0 0 S 0.0 0.0 0:00.00 khungtaskd
197 root 15 0 0 0 0 S 0.0 0.0 0:25.60 pdflush
198 root 10 -5 0 0 0 S 0.0 0.0 0:48.18 kswapd0
199 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 aio/0
200 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 aio/1
351 root 11 -5 0 0 0 S 0.0 0.0 0:00.00 kpsmoused
382 root 13 -5 0 0 0 S 0.0 0.0 0:00.00 ata/0
383 root 13 -5 0 0 0 S 0.0 0.0 0:00.00 ata/1
384 root 14 -5 0 0 0 S 0.0 0.0 0:00.00 ata_aux
384 root 14 -5 0 0 0 S 0.0 0.0 0:00.00
|
Posted by Hillockhosting, 09-11-2010, 09:09 AM |
give the top results when load is high.
Right now I see the load of 0.00, 0.29, 0.66 which wont be of any help to check the issue.
|
Posted by M Bacon, 09-11-2010, 09:46 AM |
What is your server specifications and are you running any kind of service that may spike the load?
|
Posted by mugo, 09-11-2010, 11:26 AM |
you may also want to consider installing sysstat. That will measure and keep a running historicals on all resources.
|
Posted by SunShellHosting, 09-11-2010, 11:43 AM |
Switch to CloudLinux and you'll not see high load. They have 30 days trial as well
|
Posted by mixmox, 09-11-2010, 06:43 PM |
top - 18:40:56 up 9 min, 3 users, load average: 6.24, 4.85, 2.90
Tasks: 152 total, 1 running, 151 sleeping, 0 stopped, 0 zombie
Cpu(s): 5.7%us, 1.3%sy, 0.0%ni, 91.1%id, 1.9%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 2065948k total, 698276k used, 1367672k free, 40436k buffers
Swap: 2096472k total, 0k used, 2096472k free, 393516k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
6220 apache 15 0 50356 30m 1684 S 0.7 1.5 0:00.04 httpd
6249 apache 16 0 50356 30m 1700 S 0.7 1.5 0:00.02 httpd
6548 root 15 0 2424 1056 800 R 0.7 0.1 0:00.03 top
5636 apache 15 0 50492 30m 1824 S 0.3 1.5 0:00.37 httpd
6233 apache 16 0 50356 30m 1696 S 0.3 1.5 0:00.02 httpd
6280 apache 16 0 50356 30m 1696 S 0.3 1.5 0:00.04 httpd
6305 apache 15 0 50348 30m 1696 S 0.3 1.5 0:00.02 httpd
6347 apache 15 0 50356 30m 1748 S 0.3 1.5 0:00.03 httpd
6348 apache 17 0 50356 30m 1696 S 0.3 1.5 0:00.02 httpd
1 root 15 0 2156 644 552 S 0.0 0.0 0:00.38 init
2 root RT -5 0 0 0 S 0.0 0.0 0:00.00 migration/0
3 root 34 19 0 0 0 S 0.0 0.0 0:00.00 ksoftirqd/0
4 root RT -5 0 0 0 S 0.0 0.0 0:00.00 watchdog/0
5 root RT -5 0 0 0 S 0.0 0.0 0:00.00 migration/1
6 root 34 19 0 0 0 S 0.0 0.0 0:00.00 ksoftirqd/1
7 root RT -5 0 0 0 S 0.0 0.0 0:00.00 watchdog/1
8 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 events/0
|
Posted by mugo, 09-11-2010, 06:56 PM |
Yep, it looks like your httpd process is the culprit, memory is ok, I would bet this are some of your larger files that are being hit, probably by a few users or more at a time.
There are a few ways to go here...one basic method to see which files (while the load is high), if, say, your web path is /var/www/html, do a
lsof |grep /var/www/html
That will show you the files opened right then, and how many are being hit at once.
netstat -n |grep 80 will show you your connections and amounts there-of...
you can grep out just established (netstat -n |grep ESTAB)
You can dump any of these to files by just adding
> /path/to/files/dump.txt
at the end of the command (if you didn't already know that)
|
Posted by mixmox, 09-12-2010, 05:52 PM |
One Minute - 38.91
Five Minutes - 12.16
Fifteen Minutes - 4.29
top - 16:56:03 up 2 min, 1 user, load average: 38.91, 12.16, 4.29
Tasks: 424 total, 6 running, 409 sleeping, 0 stopped, 9 zombie
Cpu(s): 40.6%us, 8.2%sy, 0.0%ni, 25.4%id, 25.3%wa, 0.1%hi, 0.3%si, 0.0%st
Mem: 2065948k total, 1166968k used, 898980k free, 33872k buffers
Swap: 2096472k total, 0k used, 2096472k free, 392732k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
4356 alex 15 0 21632 9288 4520 S 5.1 0.4 0:00.12 php-cgi
4374 alex 16 0 21628 9284 4520 S 5.1 0.4 0:00.12 php-cgi
4243 alex 15 0 22596 9.9m 4588 S 4.2 0.5 0:00.21 php-cgi
4345 alex 16 0 21632 9284 4516 S 4.2 0.4 0:00.11 php-cgi
4351 alex 15 0 22448 9500 4540 S 4.2 0.5 0:00.11 php-cgi
4354 alex 16 0 21628 9280 4520 S 4.2 0.4 0:00.11 php-cgi
4355 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.12 php-cgi
4357 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4363 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4367 alex 16 0 21628 9280 4516 S 4.2 0.4 0:00.11 php-cgi
4368 alex 16 0 21632 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4372 alex 16 0 21632 9284 4516 S 4.2 0.4 0:00.11 php-cgi
4378 alex 16 0 21632 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4379 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4382 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4385 alex 16 0 21628 9284 4520 S 4.2 0.4 0:00.11 php-cgi
4390 alex 16 0 21632 9284 4516 S 4.2 0.4 0:00.11 php-cgi
4278 alex 15 0 22600 9.9m 4588 S 3.4 0.5 0:00.22 php-cgi
4444 alex 16 0 21160 7976 4128 S 3.4 0.4 0:00.04 php-cgi
4449 alex 15 0 21160 7852 4028 S 3.4 0.4 0:00.04 php-cgi
4451 alex 15 0 21160 7852 4028 S 3.4 0.4 0:00.04 php-cgi
4273 alex 15 0 22592 9.9m 4588 S 2.5 0.5 0:00.22 php-cgi
4307 mysql 15 0 99.2m 26m 3000 S 2.5 1.3 0:00.05 mysqld
|
Posted by akasharya, 09-13-2010, 01:23 AM |
As you can see in this line "Cpu(s): 40.6%us, 8.2%sy, 0.0%ni, 25.4%id, 25.3%wa, 0.1%hi, " cpu is used by 40% by users(not by any system user, as you can see in sy=8.2%) and later you can see user alex is running a process php-cgi, the culprit is user "alex" so you can stop this process for temporary period using skill and then monitor the server load...or if you are using cpanel/whm then you can suspend this account for a temporary period.
|
Posted by inspiron, 09-14-2010, 07:45 PM |
Check the ownerships and the permission of the script is set correct.
|
Add to Favourites Print this Article
Also Read