Re: Nforce2 sporadic hangs/freezes

From: Laurel Amberdine (circle_at_mtco.com)
Date: 09/30/03

  • Next message: Bill Marcum: "Re: USB Keyboard problem" 
    Date: 30 Sep 2003 05:06:14 GMT

    On 29 Sep 2003 07:02:22 -0700, scmail@gmx.de <scmail@gmx.de> wrote:
    >
    > I have an Asus A7N8X-X Nforce2 Board and have the same problem. If I
    > have a lot of I/O Traffice, the system will be freezed. So I turned
    > off DMA und it works, ok I have not tested like you about 6hours, but
    > I think it is a bug in kernel with nforce. Today I will try to use
    > NOAPIC and ACPI=OFF und DMA ON. I hope that this will work, because I
    > can't live with PIO, it is to slow!

    Interesting.

    I have an A7N8X motherboard. I was having lockup problems, but I could
    see what was happening. (SuSE has a text console that shows kernel
    messages. Or something. I guess.) The IDE channel was not responding.
    There would be lots of retries, and eventually a reset of the bus. After
    the reset things would operate okay for a short while, then lock up again.
    While I was "locked up" I could use the keyboard and mouse, but programs
    didn't do anything.

    I'd assumed it was my drive going bad and got a SATA HD to replace it.
    It now works without lockups, nice and fast with no problems, other than
    the initial difficulty of getting Linux on to it!

    However, having seen what you guys are saying, I wonder if there is a
    problem with the bus itself. I was never able to get any SMART errors out
    of my supposedly-failing drive.

    Before I got the new drive, I noticed that both increased activity and
    higher case temperatures caused more lockups. When I opened the little
    door in the front of my case and stuck a fan to blow in there, the lockups
    decreased. Maybe a problem with the IDE system and heat (and the kernel
    not handling the issue)? 

    -- 
- Laurel   *  *  *   http://amberdine.com
  • Next message: Bill Marcum: "Re: USB Keyboard problem"

    Relevant Pages

    • Re: RT bug with 2.6.13-rt4 and 3c905c tornado
      ... > This card works perfectly on the same PCI-X slot with a non RT kernel. ... use serial logging and the NMI watchdog to debug hard lockups (see the ... Use CONFIG_DETECT_SOFTLOCKUP=y to detect soft lockups. ...
      (Linux-Kernel)
    • Re: HPET regression in 2.6.26 versus 2.6.25 -- connection between HPET and lockups found
      ... As part of my experiments to determine the root cause of my lockups, ... return value to indicate that the HPET resource was not assigned. ... Trying to discover if something was going haywire in this part of the code, ... I tried to capture some data which I could save until just before the kernel ...
      (Linux-Kernel)
    • Re: Race condition in 2.4 tasklet handling (cli() broken?)
      ... TeJun Huh wrote: ... > It would be helpful if you can tell us more about your lockups. ... All installations are SuSE 8.2 but with standard kernel. ... Trace; c01397f8 ...
      (Linux-Kernel)
    • KDE Kernel Configurator
      ... I have put together an RH9 machine for a client to use as a small ... That went very well apart from lockups due to glx ... a go at enabling support for somathing on my own RH9 box. ... The KDE Control Centre has no help files for Kernel ...
      (linux.redhat)
    • Re: NForce2 pseudoscience stability testing (2.6.0-test11)
      ... It seems that compiling Local-APIC out of kernel has ... It's been about a month without lockups for 2.4 ... hasn't locked up either but it gets a new kernel and a reboot ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)