Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Adam Voigt]

Distro: Redhat 8.0
Kernel: Redhat 2.4.18-27.8.0 (Latest)
Mobo: Aptron P4VXAD
CPU: 2.4 GHZ Intel Pentium 4
RAM: 512 MB
Kernel Params: ide=nodma, pci=biosirq (the dmesg suggested it =))

Leaving the computer running all night, with only a CPU
intensive task (i.e., seti@home), produced the attached
message log, of what appear to be almost on-the-clock
hourly OOPS's. This ofcourse led me to believe it was something
in my cron schedule, only problem is, it's empty, so it's
either syslogd or a built in system hourly task (no problems
on 8 other machines with this same kernel, distro, and config).

Behavior with the OOPS's, is sporatic, I can turn the machine
on, wait ten minutes, and log in, and do a "ls" and it will
OOPS, other times it will be hours before I see them.

One other problem, probably unrelated, the BIOS and the Kernel
both report seeing a "Realtek 8139" NIC on the computer, though
no such card exists and it is not built onto the mobo, only a
3COM 3c59x (PCI Card).

Included in the messages the kernel seemed to want me to
report to the list two bug reports =), seeing as how it even
has "cut here" in the messages file. Attached, is the full
/var/log/message, at the end of the file, the machine just quit,
as when I came in, no display would come up. I've also attached
the output from dmesg, lsmod, and lspci, if you need anything else,
or have any ideas, PLEASE don't hesitate to tell me.

-- 
Adam Voigt (adam@cryptocomm.com)
The Cryptocomm Group
My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Adam Voigt]

Sorry, the Kernel mailing list server won't let me attach the
files, so here are URL's:

http://64.238.252.49:8080/dmesg.txt
http://64.238.252.49:8080/lsmod.txt
http://64.238.252.49:8080/lspci.txt
http://64.238.252.49:8080/messages.txt

On Thu, 2003-03-27 at 09:42, Adam Voigt wrote:
    Distro: Redhat 8.0
    Kernel: Redhat 2.4.18-27.8.0 (Latest)
    Mobo: Aptron P4VXAD
    CPU: 2.4 GHZ Intel Pentium 4
    RAM: 512 MB
    Kernel Params: ide=nodma, pci=biosirq (the dmesg suggested it =))
    
    Leaving the computer running all night, with only a CPU
    intensive task (i.e., seti@home), produced the attached
    message log, of what appear to be almost on-the-clock
    hourly OOPS's. This ofcourse led me to believe it was something
    in my cron schedule, only problem is, it's empty, so it's
    either syslogd or a built in system hourly task (no problems
    on 8 other machines with this same kernel, distro, and config).
    
    Behavior with the OOPS's, is sporatic, I can turn the machine
    on, wait ten minutes, and log in, and do a "ls" and it will
    OOPS, other times it will be hours before I see them.
    
    One other problem, probably unrelated, the BIOS and the Kernel
    both report seeing a "Realtek 8139" NIC on the computer, though
    no such card exists and it is not built onto the mobo, only a
    3COM 3c59x (PCI Card).
    
    Included in the messages the kernel seemed to want me to
    report to the list two bug reports =), seeing as how it even
    has "cut here" in the messages file. Attached, is the full
    /var/log/message, at the end of the file, the machine just quit,
    as when I came in, no display would come up. I've also attached
    the output from dmesg, lsmod, and lspci, if you need anything else,
    or have any ideas, PLEASE don't hesitate to tell me.
    
    -- 
    Adam Voigt (adam@cryptocomm.com)
    The Cryptocomm Group
    My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc
-- 
Adam Voigt (adam@cryptocomm.com)
The Cryptocomm Group
My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Alan Cox]

On Thu, 2003-03-27 at 14:42, Adam Voigt wrote:
> Behavior with the OOPS's, is sporatic, I can turn the machine
> on, wait ten minutes, and log in, and do a "ls" and it will
> OOPS, other times it will be hours before I see them.

Does it pass things like memtest86

> One other problem, probably unrelated, the BIOS and the Kernel
> both report seeing a "Realtek 8139" NIC on the computer, though
> no such card exists and it is not built onto the mobo, only a
> 3COM 3c59x (PCI Card).

If its seen it will be there somewhere. It may just be integrated
into something and not actually used by the vendor.

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Adam Voigt]

Thanks for responding,

Yes, memtest passes with flying color's.

On Thu, 2003-03-27 at 12:04, Alan Cox wrote:
    On Thu, 2003-03-27 at 14:42, Adam Voigt wrote:
    > Behavior with the OOPS's, is sporatic, I can turn the machine
    > on, wait ten minutes, and log in, and do a "ls" and it will
    > OOPS, other times it will be hours before I see them.
    
    Does it pass things like memtest86
    
    > One other problem, probably unrelated, the BIOS and the Kernel
    > both report seeing a "Realtek 8139" NIC on the computer, though
    > no such card exists and it is not built onto the mobo, only a
    > 3COM 3c59x (PCI Card).
    
    If its seen it will be there somewhere. It may just be integrated
    into something and not actually used by the vendor.
    
-- 
Adam Voigt (adam@cryptocomm.com)
The Cryptocomm Group
My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Alan Cox]

On Thu, 2003-03-27 at 17:07, Adam Voigt wrote:
> Thanks for responding,
> 
> Yes, memtest passes with flying color's.

Next test is probably "noapm" as a boot option (so you dont
turn on any bios stuff that may be buggy) - also disable
bios power management and bios support for usb keyboard/mouse

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Adam Voigt]

Same thing, I ran a "find /" and when that didn't crash
it, I ran lynx, and tried to download the kernel from
kernel.org to try and make it die with all the compiling
activity, but just quiting lynx crashed it. Also,
I'm pretty sure this doesn't help, but the caps lock
and scroll lock keys are blinking in a regular interval.

On Thu, 2003-03-27 at 12:28, Alan Cox wrote:
> Next test is probably "noapm" as a boot option (so you dont
> turn on any bios stuff that may be buggy) - also disable
> bios power management and bios support for usb keyboard/mouse

-- 
Adam Voigt (adam@cryptocomm.com)
The Cryptocomm Group
My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Alan Cox]

On Thu, 2003-03-27 at 17:55, Adam Voigt wrote:
> Same thing, I ran a "find /" and when that didn't crash
> it, I ran lynx, and tried to download the kernel from
> kernel.org to try and make it die with all the compiling
> activity, but just quiting lynx crashed it. Also,
> I'm pretty sure this doesn't help, but the caps lock
> and scroll lock keys are blinking in a regular interval.

At this point I really suspect the hardware, assuming you
aren't loading junk weird modules and its not a misbuilt
kernel of somekind.

Re: Kernel Itself Reports Bug, Continuous OOPS's, and Phantom NIC Card

[Adam Voigt]

Alright, thanks for all your help.

On Thu, 2003-03-27 at 14:09, Alan Cox wrote:
> At this point I really suspect the hardware, assuming you
> aren't loading junk weird modules and its not a misbuilt
> kernel of somekind.
> 
-- 
Adam Voigt (adam@cryptocomm.com)
The Cryptocomm Group
My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc