Results 1 to 12 of 12
  1. #1
    Join Date
    Apr 2005
    Posts
    537

    Strange Load problems (again)

    hello again all,

    ok, ive posted about this before, but now i have reason to ask again.

    my server is experiancing stupid (relitivly) loads even tho at this precise moment their is 15 members online, but my top output looks like the following

    Code:
    top - 20:59:51 up 2 days,  5:43,  1 user,  load average: 0.83, 0.75, 0.62
    Tasks:  66 total,   1 running,  65 sleeping,   0 stopped,   0 zombie
    Cpu(s): 45.2% us,  4.0% sy,  0.0% ni, 50.8% id,  0.0% wa,  0.0% hi,  0.0% si
    Mem:    507820k total,   477528k used,    30292k free,   129348k buffers
    Swap:  2104504k total,      160k used,  2104344k free,   248656k cached
    
      PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
    19914 nobody    15   0 15356 7088 2216 S 20.6  1.4   5:02.65 /usr/local/apache/bin/httpd -DSSL
    19921 nobody    17   0 15364 7232 2352 S 19.6  1.4   5:22.44 /usr/local/apache/bin/httpd -DSSL
    25952 root      16   0 25788 4080  728 S  5.6  0.8  24:39.38 ./sc_trans_linux
     3110 mysql     15   0  100m  15m 3400 S  2.7  3.2  94:10.06 /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/ser
     1756 root      17   0  3148  944  760 R  0.7  0.2   0:00.24 top -c
     1722 root      16   0  8300 2320 1848 S  0.3  0.5   0:00.16 sshd: root@pts/2
        1 root      16   0  3660  552  472 S  0.0  0.1   0:00.81 init [3]
        2 root      34  19     0    0    0 S  0.0  0.0   0:00.17 [ksoftirqd/0]
        3 root       5 -10     0    0    0 S  0.0  0.0   0:00.01 [events/0]
        4 root       5 -10     0    0    0 S  0.0  0.0   0:00.02 [khelper]
        5 root      15 -10     0    0    0 S  0.0  0.0   0:00.00 [kacpid]
       20 root       5 -10     0    0    0 S  0.0  0.0   0:00.00 [kblockd/0]
       38 root      15   0     0    0    0 S  0.0  0.0   0:00.23 [pdflush]
       39 root      15   0     0    0    0 S  0.0  0.0   0:00.67 [pdflush]
       41 root       8 -10     0    0    0 S  0.0  0.0   0:00.00 [aio/0]
       21 root      15   0     0    0    0 S  0.0  0.0   0:00.00 [khubd]
       40 root      15   0     0    0    0 S  0.0  0.0   0:02.56 [kswapd0]
      187 root      25   0     0    0    0 S  0.0  0.0   0:00.00 [kseriod]
      298 root      15   0     0    0    0 S  0.0  0.0   0:00.57 [kjournald]
     1333 root       6 -10  3116  468  380 S  0.0  0.1   0:00.03 udevd
     1725 root       6 -10     0    0    0 S  0.0  0.0   0:00.00 [kauditd]
     1765 root       6 -10     0    0    0 S  0.0  0.0   0:00.00 [kmirrord]
     1792 root      15   0     0    0    0 S  0.0  0.0   0:00.00 [kjournald]
     1793 root      15   0     0    0    0 S  0.0  0.0   0:02.45 [kjournald]
     1794 root      15   0     0    0    0 S  0.0  0.0   0:00.03 [kjournald]
     1795 root      15   0     0    0    0 S  0.0  0.0   0:00.78 [kjournald]
     1796 root      15   0     0    0    0 S  0.0  0.0   0:05.17 [kjournald]
     2649 root      16   0  2904  556  460 S  0.0  0.1   0:00.66 syslogd -m 0
     2653 root      16   0  2284  380  316 S  0.0  0.1   0:00.00 klogd -x
     2708 root      16   0  4392  532  272 S  0.0  0.1   0:00.02 rpc.idmapd
     2776 root      15   0  2664  508  300 S  0.0  0.1   0:00.02 /usr/sbin/smartd
     2785 root      18   0  1692  440  368 S  0.0  0.1   0:00.00 /usr/sbin/acpid
     2851 root      18   0  4884 1228 1040 S  0.0  0.2   0:00.00 /bin/bash
     2940 root      18   0  2924  756  624 S  0.0  0.1   0:00.01 xinetd -stayalive -pidfile /var/run/xinetd.pid
     2963 root      16   0  8036 2024  984 S  0.0  0.4   0:00.11 sendmail: accepting connections
     2971 smmsp     16   0  7544 1624  812 S  0.0  0.3   0:00.00 sendmail: Queue runner@01:00:00 for /var/spool/clientmqueue
     2987 root      16   0  6520  932  540 S  0.0  0.2   0:00.10 crond
     3014 dbus      16   0  3248  820  676 S  0.0  0.2   0:00.09 dbus-daemon-1 --system
     3023 root      16   0  6936 4636 1316 S  0.0  0.9   0:00.60 hald
     3044 root      21   0  1480  392  320 S  0.0  0.1   0:00.00 /usr/sbin/portsentry -tcp
     3077 root      21   0  5364 1244 1084 S  0.0  0.2   0:00.01 /bin/sh /usr/bin/mysqld_safe --datadir=/var/lib/mysql --pid-file=/var/lib/mysql/server.radiodmp.
     3181 root      16   0  7216 4912 1360 S  0.0  1.0   0:01.15 /usr/bin/perl /usr/libexec/webmin/miniserv.pl /etc/webmin/miniserv.conf
     3185 root      18   0  2636  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty1
     3186 root      18   0  2844  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty2
     3187 root      18   0  1564  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty3
     3188 root      18   0  1684  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty4
     3189 root      18   0  2036  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty5
     3190 root      18   0  1756  412  356 S  0.0  0.1   0:00.00 /sbin/mingetty tty6
     3834 named     16   0 37080 2828 1920 S  0.0  0.6   0:05.73 /usr/sbin/named -u named -t /var/named/chroot
    i havent got a clue whats causing this load,

    cureent details on the server are its using apache to serve dynamic pages. and ive offloaded all images to lighttpd which isnt even registering a load so thats good. the site running on server is very mysql heavy but ive tweaked mysql to the best i can. not sure about any tweaks i can do to apache tho

    thanks

  2. #2
    Join Date
    Jan 2006
    Location
    Jersey
    Posts
    2,971
    Well, the load stats doesnt look too bad. What are you basing you "heavy load" issues on? pages loading slow? server responding to commands slowly? or something else?
    Email: info ///at/// honelive.com

  3. #3
    Join Date
    Apr 2005
    Posts
    537
    the fact the sites taking seconds to pull pages up and normally doesnt, and the fact that the servers only a 2ghz celeron so any load over about 1.5 tends to pretty much kill it

  4. #4
    Join Date
    Apr 2005
    Location
    Jacksonville, FL
    Posts
    981
    Your assessment about killing a 2ghz celeron with a 1.5 load is a bit off, to say the least. Load average is not what you should necessarily be looking at, but rather the idle percentage of your CPU. If your machine is hacked, the 1.5 load you mention will not be indicative of the actual load average in the case of a process-hiding kernel module. 'top' will never see hacked processes, and thus, the load they generate won't be reflected. But hackers can't seem to modify the idle CPU percentage; use that as your benchmark. If you have a CPU that's consistently 10% idle and a load average of 0.3, you'll know something is not right.

    Now selling BigVPS's!
    Jacksonville Colocation and dedicated servers by colo4jax
    We are *not* a reseller. We own our servers, switches, routers and racks.

  5. #5
    Join Date
    Apr 2002
    Location
    USA
    Posts
    5,783
    run netstat -n and look at the amount of network connections and to what ports.

  6. #6
    Join Date
    Jan 2007
    Location
    Netherlands
    Posts
    20
    try to reload /restart mysql

    then see if the load is still high of that it may be an other process

    but that is =/var/lib/mysql/server.radiodmp.
    streaming radio ?

  7. #7
    Join Date
    Apr 2005
    Posts
    537
    the servers hostname is server.radiodmp.net - i do run a shoutcast stream(and sc_trans) off it but the load that places is usually nil. now ive got a weird problem. last night lighttpd was happily serving images on port 81 now it isnt and wont even load any images, and yes ive tried restarting it. HELP. as its crashing my site

  8. #8
    Join Date
    Apr 2005
    Posts
    537
    scratch the lighttpd problem, seems apf had magicly restarted itself

  9. #9
    Quote Originally Posted by tical
    Your assessment about killing a 2ghz celeron with a 1.5 load is a bit off, to say the least.
    Totally agree with tical.
    The server will run ok with 3+ load (but less than 5 )

    The top processes are from your website:

    19914 nobody 15 0 15356 7088 2216 S 20.6 1.4 5:02.65 /usr/local/apache/bin/httpd -DSSL
    19921 nobody 17 0 15364 7232 2352 S 19.6 1.4 5:22.44 /usr/local/apache/bin/httpd -DSSL

    There is no IO waiting, so it couldn't be logs processing.
    PremiumReseller.com Hyper-V SSD VPS USA London Singapore
    Reseller Hosting Cpanel PURE SSD CloudLinux Softaculous
    Windows Reseller Asp.NET 4.5 MSSQL SmarterMail Enterprise

  10. #10
    Join Date
    Apr 2005
    Posts
    537
    ill keep an eye on it then, out of curiosity though would disabling SSL in apache help at all (because i dont need ssl )

  11. #11
    Join Date
    Jan 2007
    Location
    Netherlands
    Posts
    20
    try to clear your /tmp
    and restart httpd

    why is /usr/local/apache/bin/httpd -DSSL started by nobody and not as user apache ?

  12. #12
    Join Date
    Apr 2005
    Posts
    537
    because i havent configured apache to run as a user (yet). how do i disable ssl on apache though ?

Posting Permissions

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