Strange discrepancy with gmirror and iostat - BSD

This is a discussion on Strange discrepancy with gmirror and iostat - BSD ; I just noticed that one of the drives in a mirror apparently has a busier SATA interface: $ iostat -w 1 -x device r/s w/s kr/s kw/s wait svc_t b ad4 52.6 93.4 3355.1 7104.2 0 46.7 16 ad6 52.6 ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: Strange discrepancy with gmirror and iostat

  1. Strange discrepancy with gmirror and iostat

    I just noticed that one of the drives in a mirror apparently has a
    busier SATA interface:

    $ iostat -w 1 -x
    device r/s w/s kr/s kw/s wait svc_t b
    ad4 52.6 93.4 3355.1 7104.2 0 46.7 16
    ad6 52.6 93.7 3355.1 7138.1 1 95.1 87
    .....
    ad4 52.1 73.9 3299.8 6951.5 0 35.2 14
    ad6 52.1 73.9 3297.9 6951.5 2 54.6 87
    .....
    ad4 50.3 92.2 3198.2 7175.0 0 14.0 20
    ad6 50.4 92.2 3200.0 7175.0 1 43.8 73

    Note that the first few values are near identical, which is to be
    expected since it's a mirror. However, the svc_t and b values are
    signifcantly higher for ad6.

    Any ideas what may be causing this, and whether it's a measuring quirk
    or an indication of an actual problem? There are no errors reported in
    the logs. FreeBSD 7.0 amd64. TIA for any suggestions...

  2. Re: Strange discrepancy with gmirror and iostat

    On Thu, 1 May 2008 11:13:55 -0700 (PDT), googlegroups@sensation.net.au
    wrote:

    >I just noticed that one of the drives in a mirror apparently has a
    >busier SATA interface:
    >
    >$ iostat -w 1 -x
    >device r/s w/s kr/s kw/s wait svc_t b
    >ad4 52.6 93.4 3355.1 7104.2 0 46.7 16
    >ad6 52.6 93.7 3355.1 7138.1 1 95.1 87
    >....
    >ad4 52.1 73.9 3299.8 6951.5 0 35.2 14
    >ad6 52.1 73.9 3297.9 6951.5 2 54.6 87
    >....
    >ad4 50.3 92.2 3198.2 7175.0 0 14.0 20
    >ad6 50.4 92.2 3200.0 7175.0 1 43.8 73
    >
    >Note that the first few values are near identical, which is to be
    >expected since it's a mirror. However, the svc_t and b values are
    >signifcantly higher for ad6.
    >
    >Any ideas what may be causing this, and whether it's a measuring quirk
    >or an indication of an actual problem?


    Maybe the two drives don't have the same cache/buffer size?

    >There are no errors reported in
    >the logs. FreeBSD 7.0 amd64. TIA for any suggestions...



  3. Re: Strange discrepancy with gmirror and iostat

    On May 3, 4:57 am, no...@nowhere.com (Speechless) wrote:
    > Maybe the two drives don't have the same cache/buffer size?


    Purchased at the same time, identical model and firmware according to
    dmesg

    ad4: 152627MB at ata2-master SATA300
    ad6: 152627MB at ata3-master SATA300

+ Reply to Thread