On 5/2/99 11:31, Gevaerts Frank at [EMAIL PROTECTED] wrote:
>I think your video card has nothing to do with it.
>Try giving "mem=8M" or something like that at the LILO prompt. Maybe it
>sometimes detects your memory size wrong.
>Is your memory fast enough for your BIOS settings ?
>
I will try "mem=32M" at the lilo prompt to see if that works (it was my
video card that has, I believe, 8MB), but I am skeptical that that will
have any effect. This machine has been running fine for over a year now
with no memory related problems that I have noticed. As for the memory
speed, each of the chips is 60ns, which is as it should be.
>Anyway, next time you get an Oops, write everything down and read
>/usr/src/linux/Documentation/oops-tracing.txt. Once you have followed its
>instructions you have something the kernel developpers can analyse, in
>case you have a real kernel bug.
>
Well, that's good to know for the future, and if I get another oops
message, I will write it down. But, my ability to compare the Oops
message with anything in oops-tracing.txt is directly dependent on my
finding the solution to my current problem.
Cheers,
Sean
>Frank
>
>
>On Sun, 2 May 1999, T. Sean (Theo) Schulze wrote:
>
>> About a week and a half ago, I came home to find my screen filled with
>> looked like memory stack dump. At the bottom of the screen were a couple
>> of lines of text that read (loosely): "Aaiee...tried to kill the idle
>> task!" and "Kernel panic...killing the interrupt handler." I have been
>> working this problem on and off ever since then, and still don't have it
>> solved. The machine is one I assembled from bits, and it is/was running
>> SuSE Linux 6.0 with kernel 2.2.1. It had 40MB of RAM, but while
>> troubleshooting, I found that one of my SIMMs was not getting reported as
>> EDO and so I swapped it and its complement out. The box now has 32MB
>> RAM. The video card is a generic S3 VirGE/DX with 8(?)MB and 170MHz
>> RAMDAC. The monitor is a Cornerstone 45/101nf 19".
>>
>> When I first saw what was on the screen, I tried switching consoles using
>> the Alt-F* combinations, but that didn't work. I tried telnetting in
>> from my NetBSD box, but that didn't work. Ctl-C, Ctl-D and Ctl-Alt-Del
>> were likewise ineffective. So, I resigned myself to being in one of
>> those states where the computer is truly frozen and switched it off.
>> When I first tried to reboot, I got the message that my kernel was not a
>> valid compressed image. Another attempt at reboot gave me a full blown
>> "Oops: 0000" error message with stack dump. Since then, I have tried
>> booting with my older kernel (2.2.0), my rescue kernel (2.0.36), Tom's
>> Root/Boot disk, an emergency disk built on my SuSE Linux box but using an
>> earlier kernel, a Slackware boot/root disk combination, and (ugh!) a
>> Windows95 startup disk.
>>
>> * The emergency disk and the Slackware boot disk fail to boot the
>> machine, and they give me the error message that there is not enough
>> memory (although I haven't tried these since troubleshooting my RAM).
>>
>> * Tom's root/boot disk will either cease booting with the not enough
>> memory error, or it will begin to boot, get to the point where the vga
>> mode is to be selected, and then reboot the machine without booting when
>> a vga mode is chosen. (This seems to be the case, no matter what vga
>> mode is selected, although I haven't tried all of them yet using
>> tomsrtbt.)
>>
>> * The older kernel and the rescue kernel throw the machine into reboot
>> when the vga mode is selected.
>>
>> * The Windows95 startup disk seems to have no problem at all and boots
>> the machine up to the A:\ prompt. That is to say, the machine remains in
>> an (arguably) useless condition. :-)
>>
>> * I have also tried several times to just boot "normally", that is,
>> using my current kernel. The day before yesterday, the machine booted up
>> to checking the disks with fsck. It got to the partition with my /usr
>> partition on it and reported that fsck could not complete the check. It
>> put me into single user mode and told me to do the fsck manually. Since
>> it was pretty late by then, and since I wanted to read up a bit on manual
>> fscking, I didn't do it right then. Not wanting to leave the machine
>> running overnight in its disable condition, I shut it down. (Moral:
>> Never, ever voluntarily give up a gain you've made towards a solution.)
>> When I tried to reboot again the next day, I was back at the invalid
>> compressed image error message or the Oops. The last time I tried to
>> boot from using my normal kernel, I was presented with a colorful "crazy
>> quilt" of colored bars about 1/4cm wide and 3/4cm long each filled with
>> at least one ASCII character. Some of them had more than one character
>> in them, and in others the characters were blinking.
>>
>> I am wondering how to proceed, and I am hoping for some advice from the
>> Linux veterans here on the list. I am thinking that the problem might be
>> with my video card, since I have recently installed this monitor, and I
>> suspect that it may be more monitor than this card was designed to
>> handle. Today I will swap out the video card with an older card I have,
>> an S3 Trio64V+, and hook up my old 15" Packard-Hell monitor to it to see
>> what that does.
>>
>> Other than that, does anyone else have any thoughts on what could be
>> causing this? Once I get this fixed, I plan on saving my course work
>> from my C++ classes and then doing a total reinstall as part of a move to
>> a dual boot Windows98/SuSE Linux 6.0 system. (I need Windows98 for a
>> programming class, otherwise I would still be Windows-free.)
>>
>> TIA,
>> Sean.
>>
Theo. Sean Schulze
[EMAIL PROTECTED] [EMAIL PROTECTED]
**************************************************
The cost of living hasn't affected its popularity.
