LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Server (https://www.linuxquestions.org/questions/linux-server-73/)
-   -   help, resource usage problem :( (https://www.linuxquestions.org/questions/linux-server-73/help-resource-usage-problem-4175429873/)

mhrlive 10-01-2012 07:58 AM
help, resource usage problem :(
 
Hello,
I own two VPS, first one with 384MB RAM and the second one with 256MB RAM, the two servers has the same OS Centos 5.8 and the same packages ( LAMP) using just the same website, same DB, same httpd.conf and same my.cnf, the problem that annoying me and blows my mind is on the first server with 384MB its using only 80 to 95 MB of RAM but the second one with 256 its using all the RAM causing server to stop, tried to optimize tables, do anything found on google with no result.

top on the 256MB server currently show and it will goes up by the time:
Code:
top - 08:56:07 up 16:49,  1 user,  load average: 19.67, 6.95, 4.43
Tasks:  38 total,  1 running,  36 sleeping,  0 stopped,  1 zombie
Cpu(s): 12.7%us, 37.9%sy,  0.0%ni, 45.8%id,  1.7%wa,  0.0%hi,  2.0%si,  0.0%st
Mem:    262144k total,  227536k used,    34608k free,    3788k buffers
Swap:        0k total,        0k used,        0k free,        0k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
17283 apache    20  0 39992 6616 2656 S  7.3  2.5  0:01.60 httpd
16799 apache    20  0 40344 7228 3044 S  3.7  2.8  0:01.07 httpd
16798 apache    20  0 40344 7164 3012 S  2.3  2.7  0:02.39 httpd
12052 apache    20  0 40132 7272 3012 S  0.7  2.8  0:00.16 httpd
12056 apache    20  0 40424 7248 3004 S  0.3  2.8  0:00.06 httpd
15544 apache    20  0 40432 7448 3072 S  0.3  2.8  0:00.15 httpd
15549 apache    20  0 40332 7424 3096 S  0.3  2.8  0:00.07 httpd
16354 apache    20  0 40140 7296 3044 S  0.3  2.8  0:00.04 httpd
16704 apache    20  0 40424 7268 3016 S  0.3  2.8  0:00.10 httpd
16705 apache    20  0 40124 7284 3016 S  0.3  2.8  0:00.08 httpd
16758 apache    20  0 40124 7268 3004 S  0.3  2.8  0:00.06 httpd
16760 apache    20  0 40124 7260 3004 S  0.3  2.8  0:00.05 httpd
16797 apache    20  0 40424 7264 3016 S  0.3  2.8  0:00.09 httpd
    1 root      20  0  8356  728  676 S  0.0  0.3  30:00.42 init
 1105 apache    20  0 39640 4368  708 S  0.0  1.7  0:00.00 httpd
12053 apache    20  0 40712 7608 3084 S  0.0  2.9  0:00.40 httpd
12711 apache    20  0 40432 7428 3068 S  0.0  2.8  0:00.11 httpd
13312 apache    20  0 40424 7372 3076 S  0.0  2.8  0:00.24 httpd
14408 apache    20  0 40432 7524 3144 S  0.0  2.9  0:00.27 httpd
14425 apache    20  0    0    0    0 Z  0.0  0.0  0:00.10 httpd <defunct>
14461 apache    20  0 40132 7272 3016 S  0.0  2.8  0:00.12 httpd
15035 apache    20  0 40432 7500 3116 S  0.0  2.9  0:00.15 httpd
15367 apache    20  0 40432 7472 3096 S  0.0  2.9  0:00.21 httpd
15499 apache    20  0 40124 7468 3200 S  0.0  2.8  0:00.08 httpd
16091 apache    20  0 40432 7568 3072 S  0.0  2.9  0:00.06 httpd
16278 apache    20  0 40332 7392 3072 S  0.0  2.8  0:00.05 httpd
16353 apache    20  0 40432 7484 3044 S  0.0  2.9  0:00.03 httpd
16677 apache    20  0 40124 7264 3008 S  0.0  2.8  0:00.06 httpd
16757 apache    20  0 40432 7516 3048 S  0.0  2.9  0:00.05 httpd
16759 apache    20  0 40124 7236 3004 S  0.0  2.8  0:00.01 httpd
16796 apache    20  0 40344 7160 3012 S  0.0  2.7  0:01.68 httpd
17109 root      20  0  2300 1040  828 R  0.0  0.4  0:00.02 top
httpd:
Code:
<IfModule prefork.c>
StartServers      1
MinSpareServers    1
MaxSpareServers  3
ServerLimit      2000
MaxClients      1200
MaxRequestsPerChild  9000
</IfModule>

<IfModule worker.c>
StartServers        2
MaxClients        1600
MinSpareThreads    25
MaxSpareThreads    150
ThreadsPerChild    450
MaxRequestsPerChild  0
</IfModule>

Please help me to figure the cause .

unSpawn 10-01-2012 10:21 AM
Quote:
Originally Posted by mhrlive (Post 4793953)
on the first server with 384MB its using only 80 to 95 MB of RAM but the second one with 256 its using all the RAM causing server to stop
First of all Linux has a pretty good Virtual Memory Manager and seeing the VMM utilize all available memory is a good thing. Secondly if you say "causing server to stop" then there should either be daemon or syslog (OOM) messages to back that up (as in post some). Third I'm not going to compare your httpd.conf worker settings. If by posting it you meant to say you increased values then I'd suggest changing them back to default settings. People often think just increasing those values is good but it can actually be detrimental for performance. Your top output (why not show a comparison with the other server BTW?) shows no swap usage which usually is a bad sign, just one zombie process. Its output is nothing more than a snapshot and running a small SAR (Atop, Dstat, Collectl, whatever else you fancy), collecting system stats first and then comparing them could provide a better, more factual starting point IMHO because we don't even know if they're exercised in the same way. Finally I don't know the basic specs for CentOS-5.8 but just 256 MB RAM looks a bit low.


All times are GMT -5. The time now is 05:00 AM.