Results 1 to 9 of 9

Thread: HELP!!!! HUGE Load spikes

  1. #1
    Join Date
    Apr 2008
    Posts
    34
    Rep Power
    7

    Unhappy HELP!!!! HUGE Load spikes

    HELP!!!
    I initially had huge load spikes on my Zimbra server (close to 800, yes, 800)
    Anyway, it seems to have subsided now but I'm still looking for the source of the spikes.
    Meanhwhile, this is what I have right now



    top - 11:46:21 up 16 days, 18:01, 2 users, load average: 13.15, 11.00, 13.63
    Tasks: 107 total, 1 running, 105 sleeping, 1 stopped, 0 zombie
    Cpu(s): 34.4%us, 8.9%sy, 0.0%ni, 50.4%id, 5.2%wa, 0.0%hi, 1.1%si, 0.0%st
    Mem: 16633240k total, 9412788k used, 7220452k free, 316656k buffers
    Swap: 2031608k total, 0k used, 2031608k free, 5454096k cached

    PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
    953 zimbra 18 0 2116m 1.6g 5424 S 142 9.9 98:31.93 java
    424 zimbra 15 0 1738m 953m 4176 S 16 5.9 4:51.61 mysqld
    21884 zimbra 16 0 3260 2072 856 S 9 0.0 1:18.52 htop

    467 root 10 -5 0 0 0 S 3 0.0 36:06.96 kjournald
    32075 zimbra 25 0 237m 98m 38m S 2 0.6 1:47.17 slapd
    12291 named 25 0 69968 4088 2032 S 1 0.0 2:43.98 named
    15641 root 15 0 1792 828 628 S 1 0.0 0:57.36 syslogd
    398 zimbra 16 0 9616 4424 2168 S 0 0.0 0:16.62 zmlogger
    32651 zimbra 18 0 115m 19m 3548 S 0 0.1 1:24.62 mysqld
    1 root 15 0 2036 644 552 S 0 0.0 0:01.33 init
    2 root RT 0 0 0 0 S 0 0.0 0:01.57 migration/0
    3 root 34 19 0 0 0 S 0 0.0 0:01.64 ksoftirqd/0
    4 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/0
    5 root RT 0 0 0 0 S 0 0.0 0:02.35 migration/1
    6 root 34 19 0 0 0 S 0 0.0 0:00.73 ksoftirqd/1
    7 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/1


    Forgive if I'm wrong but I don't think those bold lines are normal.
    What could be the cause?
    HELP!

    ice, ice baby

  2. #2
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    I'd suggest you upgrade to the most recent release and see if the problem still occurs.
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  3. #3
    Join Date
    Apr 2008
    Posts
    34
    Rep Power
    7

    Default

    Getting approval for upgrade is a real pain in my organization.
    I remember the last time, I was forced to do a stress test and the results weren't exceptional.
    Management eventually disregarded my request.
    Is there any other option or tweak other than upgrading to zimbra 5?

    ice, ice baby

  4. #4
    Join Date
    Apr 2008
    Posts
    34
    Rep Power
    7

    Default

    Also, how I have logger running, could this be a contributory factor to MYSQL activity?
    Is there a way of limiting the size of the MYSQL database for logging?
    Alternatively, I wouldn't mind specifying the number of days the logger holds data.
    Thank you

    ice, ice baby

  5. #5
    Join Date
    Jan 2008
    Posts
    658
    Rep Power
    8

    Default

    'zmloggerctl stop' and find out

    What exactly did you mean by "how I have logger running, could this be a contributory factor to MYSQL activity?"? Did you tweak logger in some way?

  6. #6
    Join Date
    Apr 2008
    Posts
    34
    Rep Power
    7

    Default

    I stopped it before and I have restarted it. No, I didn't tweak it.
    I just wonder where the HIGH load averages come from

    ice, ice baby

  7. #7
    Join Date
    Jan 2008
    Posts
    658
    Rep Power
    8

    Default

    Ok. When you stopped logger, did the loads go down?

    According to the code posted above, CPU usage doesn't look high nor does I/O loads. Is that normal for your system? If you haven't already, I'd suggest installing iostat and see what I/O usage looks like.

  8. #8
    Join Date
    Apr 2008
    Posts
    34
    Rep Power
    7

    Default

    Dude, I assure, the load has been even higher than below...
    Attached Images Attached Images

    ice, ice baby

  9. #9
    Join Date
    Jan 2008
    Posts
    658
    Rep Power
    8

    Default

    Quote Originally Posted by yusufu View Post
    Dude, I assure, the load has been even higher than below...

    I understand that and don't need proof.. No one here is doubting that. When you run top, are wait times usually fairly high or do they remain low? A lot of times when a system has high load averages it's because it's waiting on I/O to do something. You can use iostat to tell you what kind of I/O usage your system is using.

    You said at the top of the thread that loads have subsided. Did you do anything to make them go down?

Similar Threads

  1. Replies: 10
    Last Post: 10-17-2012, 06:45 PM
  2. Long Folder List causing harsh load times?
    By SpEnTBoY in forum Administrators
    Replies: 2
    Last Post: 03-08-2011, 06:40 AM
  3. Replies: 6
    Last Post: 06-10-2008, 08:15 PM
  4. New Load Testing Tools?
    By tross in forum Administrators
    Replies: 3
    Last Post: 05-15-2008, 04:10 PM
  5. Zmlogger and CPU Spikes
    By andreychek in forum Administrators
    Replies: 4
    Last Post: 11-01-2006, 07:01 PM

Posting Permissions

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