Results 1 to 6 of 6
  1. #1

    server optimization issue

    Hello

    I have server (Dual Xeon 3.06, 2G RAM, 2 x 120 GB Hard Drive) and am
    curious, if it's configured propertly. Almost all the time, "top" gives me
    350% - 390% idle, while "average load" ranges from 1 (night, low loadings)
    to 10 (business time, high loadings).

    I tried to tune apache and increased number of threads but it didn't help.
    What could cause such a big "average load" while CPUs are almost idle and
    what system parameters could be optimized?

    And, second question, is it ok, that "top" process takes itself 15% of CPU
    time.

    Below I have listed output of "top". Any thoughs would be very appreciated.
    Thanks in advance.

    "top" output:

    Code:
    # top
     22:04:09  up 13 days, 15:14,  9 users,  load average: 0.75, 1.08, 1.36
    290 processes: 283 sleeping, 1 running, 3 zombie, 3 stopped
    CPU states:  cpu    user    nice  system    irq  softirq  iowait    idle
               total    8.8%    0.0%   15.6%   0.0%     0.0%    0.0%  374.8%
               cpu00    1.6%    0.0%    0.8%   0.0%     0.0%    0.0%   97.5%
               cpu01    3.3%    0.0%    0.8%   0.0%     0.0%    0.0%   95.8%
               cpu02    0.0%    0.0%    0.0%   0.0%     0.0%    0.0%  100.0%
               cpu03    4.1%    0.0%   14.1%   0.0%     0.0%    0.0%   81.6%
    Mem:  2068176k av, 1973448k used,   94728k free,       0k shrd,  161820k
    buff
           684396k active,            1125064k inactive
    Swap: 2096472k av,   86868k used, 2009604k free                 1168384k
    cached
    
      PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME CPU COMMAND
     7548 root      17   0  1420 1420   856 R    16.6  0.0   0:00   3
     4109 root      16   0 34288  33M 34204 S     6.6  1.6   0:31   0 httpd
    30322 root      15   0  1400 1400   496 S     0.8  0.0   0:05   1 exclog
        1 root      16   0   396  372   340 S     0.0  0.0   0:27   2 init
        2 root      RT   0     0    0     0 SW    0.0  0.0   0:00   0 swapper
        3 root      RT   0     0    0     0 SW    0.0  0.0   0:00   1 swapper
        4 root      RT   0     0    0     0 SW    0.0  0.0   0:00   2 swapper
    From Adult Web Hosting Service with love

  2. #2
    Can anyone offer me some ideas on this issue? Thanks.
    From Adult Web Hosting Service with love

  3. #3
    Join Date
    Dec 2001
    Location
    NYC, NY
    Posts
    799
    it could be disk I/O your load average isn't just for your CPU.. it is for everything from disk I/O/CPU/RAM/SWAP and such

    run a vmstat 1 when you see the load jump.. also what kernel are you running?
    Blog your life away
    http://photoblog.com

  4. #4
    Join Date
    Jun 2002
    Posts
    66
    can you post your httpd.conf (the configuration part) and your my.cnf .
    Carlos Carluccio
    COLO@ Colocation. Perfected.
    Atlanta - Dallas - Phoenix - Clifton - Los Angeles
    Total Server Solutions - Servers - US Based Support - Server Management

  5. #5
    Thanks for help, need to check "vmstat 1" - very interesting
    From Adult Web Hosting Service with love

  6. #6
    Join Date
    Apr 2005
    Location
    Sweden
    Posts
    241
    Also look at the output from "atop", if you have that installed. If not, install it . It gives you a nice view of cpu/memory consumption and also network and harddrive i/o. Paste the output of a full page here, a page that had at least 8 seconds interval between that and the previous page (so that everything is averaged out a little bit). I think default is 10 seconds though, but not sure.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •