Continuous harddisk activity

  • Dear All,


    I'm totally new in this forum and also in working with OMV.


    My setup is:
    Server: HP N54L
    Systemdisk: Intel SSD 320 Series 160GB (SSDSA2CW160)
    RAID1: 2x Seagate Barracuda LP 5900.12 2TB (ST32000542AS)


    Now the problem is the there is a continuous activity on the RAID disks all the time even though nobody is working with the system. I starts after I mounted the RAID and in my opinion it couldn't be right.


    I made a short video. It's very low but you can here the continuous sound of the HDDs at the end of the video.


    https://www.dropbox.com/s/tg0hkcfj1xxenod/VIDEO0018.mp4?dl=0 - The sound is much better when you donwload the file first ;)


    Maybe somebody had the same phenomenon before and fixed it or is able to give me some tips.


    Lars

  • Maybe because there are continuos writes.. When you mount your raid collectd creates a database and begins to collect data points.


    Most of the activity is on the system drive.

  • Shell in to OMV and create a script with the following:


    echo " TIME TID PRIO USER DISK READ DISK WRITE SWAPIN IO COMMAND"
    iotop -o -b -t -qqq


    Then you can run that and watch to see what is accessing the disk.



    Here are my notes. Some of the tuning is already in the lastest OMV but some still needs to be applied:
    ----------
    Have to keep adding interval back to /etc/collectd/collectd.conf:
    (after Hostname)
    Interval 60


    then service collectd restart to pickup the change.


    From - http://info4admins.com/tips-to…disk-in-debian-or-ubuntu/
    ----------------------------------------------------------------------------------
    cat /proc/sys/vm/dirty_writeback_centisecs - Default was 500 = 5 seconds
    echo 90000 > /proc/sys/vm/dirty_writeback_centisecs - Set to 15 minutes


    ---
    Added these raid tuning parms in to /etc/rc.local - as well as the one above!


    # RAID cache tuning - omv now setting these!
    #echo 8192 > /sys/block/md0/md/stripe_cache_size
    #echo 8192 > /sys/block/md1/md/stripe_cache_size


    ----------------------


    Just depends on the iotop output to see what is thrashing your disk.


    Hope this helps...

  • another way would be to run

    Code
    find /bin /dev /etc /home /lib /media /mnt /opt /root /sbin /srv /tmp /usr /var -cmin -5

    to find the files that have been written in the last 5 minutes, and see if there is anything being written in the data drives

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!