The Community Forums

Interact with an entire community of cPanel & WHM users!
  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Load Average, Exim, Trouble Ahead?

Discussion in 'General Discussion' started by crwilliams, Aug 21, 2005.

  1. crwilliams

    crwilliams Well-Known Member

    Joined:
    Sep 10, 2002
    Messages:
    49
    Likes Received:
    0
    Trophy Points:
    6
    Once upon a time, this server of mine cooked along beautifully with a load average consistently under 1.0, until I started having trouble with spam and dictionary attacks. I was getting upwards of 2000 messages a day stuck in the mail queue, almost all of them bounces from the dictionary attacks. The load average started noticeably creeping up, and only stopping Exim for a few minutes brought it under control. Jonathon's MailScanner package with the Dictionary Attack ACL was a lifesaver with the dictionary attack and spam problem, and I'm no longer getting messages clogged in the mail queue (and I've since learned my lesson about fail versus blackhole).

    But now, something still isn't quite right, and my load average on this server is still consistently over 2.0, with it going hours at a time over 3 or 4, sometimes shooting up to over 6 or 7. I must have done something in between the time I had the spam problem and had MailScanner installed that is still causing problems.

    My biggest concern right now is whether or not the load average is going to cause a serious hardware issue or something equally dire if I don't get it taken care of soon. From a user's perspective, the server seems fine. It doesn't feel sluggish, and all services are available. But at what point do you really begin to worry about load average? Is it one of those things that will bite you in the behind if you don't do something about it ASAP, even if services themselves seem fine? Am I looking at a potentially big problem, or am I over-reacting?

    This is a P4 3.06 GHz processor with 1GB of RAM. It normally processes about 15,000 emails per day (12,000 incoming, and about 4,000 outgoing). On a slow day, it only processes about a third of that. It has about 100 domains on it, but they're very low traffic. Most of the activity is mail, and with a few Mailman lists (low memberships, the most at about 120 people) and we're running Sympa on the server, so there are a few Sympa mailing lists (also low memberships). There haven't been any domains or new services/programs added in over a year, either. Until the recent spam issue and load average spike, this server was highly consistent, and hadn't had a single problem.

    These are a couple of today's readings from top, one sorted by CPU usage, and the second sorted by memory just 2 minutes later. These are pretty light load averages, comparatively speaking.

    16:19:47 up 17:28, 2 users, load average: 3.15, 3.05, 3.21
    106 processes: 97 sleeping, 4 running, 3 zombie, 2 stopped
    CPU states: cpu user nice system irq softirq iowait idle
    total 60.6% 0.0% 11.6% 17.8% 0.8% 0.2% 9.0%
    Mem: 1030560k av, 932744k used, 97816k free, 0k shrd, 76160k buff
    694616k actv, 81468k in_d, 13756k in_c
    Swap: 2048276k av, 112016k used, 1936260k free 511144k cached

    PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
    1618 root 15 0 4880 152 88 S 1.4 0.0 0:11 0 /usr/local/apache/bin/httpd -DSSL
    12 root 15 0 0 0 0 SW 0.4 0.0 11:16 0 kjournald
    1649 named 24 0 17264 16M 1136 S 0.4 1.6 7:48 0 /usr/sbin/named -u named
    2347 root 15 0 19404 12M 488 S 0.2 1.2 0:05 0 initlog -q -c /usr/local/cpanel/bin/cppop
    1 root 17 0 116 80 56 S 0.0 0.0 0:05 0 init [3]
    2 root 15 0 0 0 0 SW 0.0 0.0 0:00 0 keventd
    3 root 34 19 0 0 0 SWN 0.0 0.0 0:00 0 ksoftirqd/0
    6 root 25 0 0 0 0 SW 0.0 0.0 0:00 0 bdflush
    4 root 15 0 0 0 0 SW 0.0 0.0 0:10 0 kswapd
    5 root 15 0 0 0 0 SW 0.0 0.0 0:37 0 kscand
    7 root 15 0 0 0 0 SW 0.0 0.0 0:03 0 kupdated
    8 root 25 0 0 0 0 SW 0.0 0.0 0:00 0 mdrecoveryd
    67 root 25 0 0 0 0 SW 0.0 0.0 0:00 0 khubd
    605 root 15 0 0 0 0 SW 0.0 0.0 0:00 0 kjournald
    792 root 15 0 0 0 0 SW 0.0 0.0 0:00 0 loop0
    872 root 16 0 212 4 0 S 0.0 0.0 0:00 0 /bin/bash /etc/rc.d/rc 3
    1376 root 15 0 384 356 300 R 0.0 0.0 0:09 0 syslogd -m 0
    1380 root 15 0 156 100 96 S 0.0 0.0 0:00 0 klogd -x
    1451 root 23 0 720 628 468 S 0.0 0.0 0:00 0 /usr/sbin/sshd
    1465 root 15 0 468 420 340 S 0.0 0.0 0:00 0 xinetd -stayalive -pidfile /var/run/xinetd.pid
    1490 root 20 0 2044 1172 732 S 0.0 0.1 0:00 0 chkservd
    1550 mailnull 15 0 1160 916 844 S 0.0 0.0 0:00 0 /usr/sbin/exim -bd -oX 26
    1556 mailnull 15 0 1220 976 900 S 0.0 0.0 0:10 0 /usr/sbin/exim -bd
    1563 mailnull 16 0 1144 880 824 S 0.0 0.0 0:00 0 /usr/sbin/exim -C /etc/exim_outgoing.conf -q60m
    1577 mailnull 16 0 1136 824 820 S 0.0 0.0 0:00 0 /usr/sbin/exim -tls-on-connect -bd -oX 465
    1582 root 15 0 716 664 428 S 0.0 0.0 0:04 0 antirelayd
    1735 root 15 0 4020 2484 564 S 0.0 0.2 0:16 0 cppop - accepting on port 110
    1739 mailnull 15 0 15048 3232 696 S 0.0 0.3 0:12 0 /usr/bin/perl -I/usr/mailscanner/lib /usr/mailscann
    1772 root 15 0 448 440 368 S 0.0 0.0 0:00 0 crond



    16:21:39 up 17:30, 2 users, load average: 2.52, 2.82, 3.10
    104 processes: 99 sleeping, 2 running, 1 zombie, 2 stopped
    CPU states: cpu user nice system irq softirq iowait idle
    total 55.2% 0.0% 19.0% 16.8% 0.4% 5.0% 3.6%
    Mem: 1030560k av, 966860k used, 63700k free, 0k shrd, 76868k buff
    729916k actv, 82660k in_d, 12960k in_c
    Swap: 2048276k av, 111168k used, 1937108k free 514580k cached

    PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
    1178 mailnull 15 0 53856 17M 2492 S 0.2 1.7 0:00 0 MailWatch SQL
    1649 named 24 0 17264 16M 1136 S 0.6 1.6 7:49 0 /usr/sbin/named -u named
    1810 mysql 15 0 21856 13M 1284 S 0.2 1.3 0:07 0 /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mys
    2347 root 15 0 19424 12M 488 S 0.0 1.2 0:05 0 initlog -q -c /usr/local/cpanel/bin/cppop
    30552 sympa 15 0 13344 11M 2092 S 0.0 1.1 0:00 0 /usr/local/bin/perl /home/sympa/bin/task_manager.pl
    30516 sympa 15 0 13332 8956 2056 S 0.0 0.8 0:00 0 /usr/local/bin/perl /home/sympa/bin/archived.pl
    30504 sympa 15 0 13772 8344 2064 S 0.0 0.8 0:00 0 /usr/local/bin/perl /home/sympa/bin/sympa.pl
    12913 unitb 34 19 7944 7924 1132 T N 0.0 0.7 0:00 0 /usr/bin/perl /usr/local/cpanel/3rdparty/bin/awstat
    12911 unitb 34 19 7328 7292 840 S N 0.0 0.7 0:00 0 cpanellogd - http logs for unitb
    2309 root 39 19 7148 7104 660 S N 0.0 0.6 0:01 0 cpanellogd - setting up logs for unitb
    18284 root 38 19 7220 6936 1780 S N 0.0 0.6 0:00 0 cpanellogd - setting up logs for ancalime
    19220 ancalime 39 19 7320 6212 1784 S N 0.0 0.6 0:00 0 cpanellogd - http logs for ancalime
    19223 ancalime 34 19 8312 5460 1556 T N 0.0 0.5 0:00 0 /usr/bin/perl /usr/local/cpanel/3rdparty/bin/awstat
    8877 root 15 0 5108 5080 848 S 0.0 0.4 0:00 0 cpsrvd - waiting for connections
    31374 mailman 15 0 5196 5004 1976 S 0.0 0.4 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31377 mailman 15 0 4936 4592 1960 S 0.0 0.4 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31379 mailman 15 0 4976 4584 2016 S 0.0 0.4 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    2349 root 15 0 4324 4324 900 S 0.0 0.4 0:04 0 /usr/local/cpanel/bin/cppop
    31371 mailman 23 0 4780 4276 1968 S 0.0 0.4 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31376 mailman 15 0 4700 4120 1888 S 0.0 0.3 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    27444 mailnull 15 0 4128 4104 1720 S 0.0 0.3 0:00 0 /usr/bin/perl /usr/local/cpanel/bin/eximstats
    31381 mailman 15 0 4700 3936 1888 S 0.0 0.3 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31380 mailman 15 0 4820 3888 1940 S 0.0 0.3 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31378 mailman 15 0 4740 3852 1888 S 0.0 0.3 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    31375 mailman 15 0 4704 3308 1888 S 0.0 0.3 0:00 0 /usr/bin/python2 /usr/local/cpanel/3rdparty/mailman
    1739 mailnull 17 0 15048 3240 696 S 0.0 0.3 0:12 0 /usr/bin/perl -I/usr/mailscanner/lib /usr/mailscann
    1735 root 15 0 4020 2488 564 S 0.0 0.2 0:16 0 cppop - accepting on port 110
    5337 zr 15 0 1676 1588 1256 S 0.0 0.1 0:00 0 sshd: zr@pts/2
     
  2. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    It's not going to cause hardware problems - at least, it shouldn't. The key part in this is:
    Which is always your best barometer of server and service performance.

    Ultimately, unless you have a particular problem which is always possible, then the only way to improve your load average would be to throw hardware at it (dual processors, more memory, faster disks, etc).
     
Loading...

Share This Page