Loading ...
Sorry, an error occurred while loading the content.

nVidia FX1500 gone bad

Expand Messages
  • John D Groenveld
    For those of you still running Sun s final volume Solaris x64 stationary workstations, here s a reminder to do some preventative maintenance for dead or dust
    Message 1 of 4 , Dec 2, 2012
    View Source
    • 0 Attachment
      For those of you still running Sun's final volume Solaris x64
      stationary workstations, here's a reminder to do some
      preventative maintenance for dead or dust bunnied fans:
      <URL:http://www.cse.psu.edu/~groenvel/ultra40-fx1500.jpg>

      I missed the magic smoke escaping but the random Solaris
      crashes were the give-away.

      John
      groenveld@...
    • palowoda
      ... Over the years the three most random crashes I ve noticed with hardware is: 1.) CPU fan going out. 2.) Cheap capacitors on the nvidia cards. And sometimes
      Message 2 of 4 , Dec 2, 2012
      View Source
      • 0 Attachment
        --- In solarisx86@yahoogroups.com, John D Groenveld <jdg117@...> wrote:
        >
        > For those of you still running Sun's final volume Solaris x64
        > stationary workstations, here's a reminder to do some
        > preventative maintenance for dead or dust bunnied fans:
        > <URL:http://www.cse.psu.edu/~groenvel/ultra40-fx1500.jpg>
        >
        > I missed the magic smoke escaping but the random Solaris
        > crashes were the give-away.
        >

        Over the years the three most random crashes I've noticed with
        hardware is:

        1.) CPU fan going out.
        2.) Cheap capacitors on the nvidia cards. And sometimes motherboards.
        3.) Cheap plastic fans that just seize up on the nvidia cards.

        You can never get away without inspection on some schedule.

        ---Bob
      • John D Groenveld
        ... Have the gamers and other customers who demand HPGPUs convinced graphics and mainboard manufacturers to monitor these via IPMI? All the other fans in the
        Message 3 of 4 , Dec 2, 2012
        View Source
        • 0 Attachment
          In message <k9gv1d+tbp4@...>, "palowoda" writes:
          >Over the years the three most random crashes I've noticed with
          >hardware is:
          >
          >1.) CPU fan going out.
          >2.) Cheap capacitors on the nvidia cards. And sometimes motherboards.
          >3.) Cheap plastic fans that just seize up on the nvidia cards.
          >
          >You can never get away without inspection on some schedule.

          Have the gamers and other customers who demand HPGPUs convinced
          graphics and mainboard manufacturers to monitor these via IPMI?
          All the other fans in the ancient Ultra 40 had a BIOS monitor.

          John
          groenveld@...
        • John Martin
          ... The Xorg extension NV-CONTROL allows you to monitor the GPU core temperature (GPUCoreTemp) and fan speed (GPUCurrentFanSpeed). From a GeForce GT 640: $
          Message 4 of 4 , Dec 3, 2012
          View Source
          • 0 Attachment
            On 12/02/12 20:41, John D Groenveld wrote:
            > In message <k9gv1d+tbp4@...
            > <mailto:k9gv1d%2Btbp4%40eGroups.com>>, "palowoda" writes:
            > >Over the years the three most random crashes I've noticed with
            > >hardware is:
            > >
            > >1.) CPU fan going out.
            > >2.) Cheap capacitors on the nvidia cards. And sometimes motherboards.
            > >3.) Cheap plastic fans that just seize up on the nvidia cards.
            > >
            > >You can never get away without inspection on some schedule.
            >
            > Have the gamers and other customers who demand HPGPUs convinced
            > graphics and mainboard manufacturers to monitor these via IPMI?
            > All the other fans in the ancient Ultra 40 had a BIOS monitor.

            The Xorg extension NV-CONTROL allows you to monitor the
            GPU core temperature (GPUCoreTemp) and fan speed (GPUCurrentFanSpeed).

            From a GeForce GT 640:

            $ nvidia-settings -q GPUCoreTemp

            Attribute 'GPUCoreTemp' (:0.0): 27. // degC
            'GPUCoreTemp' is an integer attribute.
            'GPUCoreTemp' is a read-only attribute.
            'GPUCoreTemp' can use the following target types: X Screen, GPU.

            $

            $ nvidia-settings -q fans

            1 Fan on :0

            [0] :0[fan:0] (Fan 0)

            $ nvidia-settings -q :0[fan:0]/GPUCurrentFanSpeed

            Attribute 'GPUCurrentFanSpeed' (:0[fan:0]): 30. // percent
            The valid values for 'GPUCurrentFanSpeed' are in the range 0 - 100
            (inclusive).
            'GPUCurrentFanSpeed' can use the following target types: Fan.

            $
          Your message has been successfully submitted and would be delivered to recipients shortly.