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.

How to trace high loads?

Discussion in 'Workarounds and Optimization' started by SuperBaby, Aug 7, 2010.

  1. SuperBaby

    SuperBaby Well-Known Member

    Joined:
    Nov 27, 2003
    Messages:
    331
    Likes Received:
    0
    Trophy Points:
    16
    Location:
    Thailand
    cPanel Access Level:
    Website Owner
    Twitter:
    I get warning emails from my server almost every alternate day. It said the server loading was too high. This week it happened on Sun, Mon, Wed and Fri as observed under Munin.

    I checked the backup schedule and also cronjobs. No clue.

    How do I trace the cause of the high loads. Thanks.
     
  2. asent

    asent Active Member

    Joined:
    Apr 5, 2005
    Messages:
    33
    Likes Received:
    0
    Trophy Points:
    6
    Watch it in real time with top.
     
  3. SuperBaby

    SuperBaby Well-Known Member

    Joined:
    Nov 27, 2003
    Messages:
    331
    Likes Received:
    0
    Trophy Points:
    16
    Location:
    Thailand
    cPanel Access Level:
    Website Owner
    Twitter:
    The problem is, when I received the email notification, the high load is over. I need to back trace.
     
  4. cPanelDon

    cPanelDon cPanel Quality Assurance Analyst
    Staff Member

    Joined:
    Nov 5, 2008
    Messages:
    2,557
    Likes Received:
    7
    Trophy Points:
    38
    Location:
    Houston, Texas, U.S.A.
    cPanel Access Level:
    DataCenter Provider
    Twitter:
    As a starting point, you may review averaged daily statistics using WHM via the following menu path: WHM: Main >> Server Status >> Daily Process Log

    What exact information is included in the e-mails you are receiving? Can you provide an example of the full mail headers and or message content?

    Are you seeing only increased server load, or are you also seeing other symptoms or indications of a possible problem?

    Is the affected system using both suPHP and suEXEC? If yes, this will make it easier to track which users may be contributing to an increased server load. The status of suPHP and suEXEC can be determined using the following command via root SSH access:
    Code:
    # /usr/local/cpanel/bin/rebuild_phpconf --current
     
  5. SuperBaby

    SuperBaby Well-Known Member

    Joined:
    Nov 27, 2003
    Messages:
    331
    Likes Received:
    0
    Trophy Points:
    16
    Location:
    Thailand
    cPanel Access Level:
    Website Owner
    Twitter:
    1) I do not use suPHP or suEXEC.

    2) Here is the alert from the server company:
    Code:
    Service: check_unix_load
    Host: C1XXXX-user
    Address: 75.100.100.100
    State: WARNING
    
    Date/Time: Sat Aug 7 17:16:43 CDT 2010
    
    Additional Info: 5-min load currently 7.04, up 75 days, 23:15
    3) Here is the alert from ConfigServer:
    Code:
    lfd: High 5 minute load average alert - 6.12
    
    Time:                    Fri Aug  6 06:15:07 2010 +0800
    1 Min Load Avg:          7.35
    5 Min Load Avg:          6.12
    15 Min Load Avg:         3.59
    Running/Total Processes: 7/141
     
  6. cPanelDon

    cPanelDon cPanel Quality Assurance Analyst
    Staff Member

    Joined:
    Nov 5, 2008
    Messages:
    2,557
    Likes Received:
    7
    Trophy Points:
    38
    Location:
    Houston, Texas, U.S.A.
    cPanel Access Level:
    DataCenter Provider
    Twitter:
    What are the hardware specifications of the system? A server load of ~7 may be relatively insignificant depending on the overall hardware configuration and available resources. What is the output of the following command (to view CPU information)?
    Code:
    # cat /proc/cpuinfo
    I recommend monitoring the server more directly during the hours or time frames reported via e-mail, and while monitoring I would evaluate system responsiveness and also check if there may be increased I/O wait times or other symptoms discovered. Via root SSH access you may consider monitoring via tools such as ps, pstree, top, htop, mytop, vmstat, and or any other utilities you find available.
     
  7. B12Org

    B12Org Well-Known Member

    Joined:
    Jul 15, 2003
    Messages:
    692
    Likes Received:
    1
    Trophy Points:
    18
    Location:
    Seattle Washington
    cPanel Access Level:
    Root Administrator
    with out taking a running shapshot and having PIDs of processes it would be hard to back trace this - if you had PIDs you could then enable logging to log the pids in the httpd log format and then parse out the pid from the log files to get a specific domain and file that is causing issues.
     
Loading...

Share This Page