[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: To ext3 developers



Hi,

On Tue, May 21, 2002 at 04:17:06AM -0700, Halbert Thiodore wrote:
> On Tue, 21 May 2002, Stephen C. Tweedie wrote:
> > Yes, it is either a bug, or something [else].  It is hardware failure
> > in some cases, a core VFS bug in others, an ext3 bug in others.
> 
> Hardware failure is not possible in this case because everything is smooth
> under windows 98.

Don't you believe it --- we have had plenty of instances where Linux
was able to drive the hardware a bit harder than Windows, exposing
hardware problems that Windows did not see.  There are also plenty of
cases where there are hardware problems and the manufacturer "solved"
them by adding a Windows driver to disable the buggy feature.  Just
because something works in Windows and not in Linux does not mean
there's no hardware fault.

> > Please tell us the exact kernel version, hardware config, and the text
> > of the error.  Without precise information, we can't diagnose
> > problems, we can only guess.
 
> cpu: 0
> eip: 0010 [<c0134c1>] not tainted
              ^^^^^^^  
This should be an 8-digit number.

> eflags: 00010202
> eip is at (2.4.18-3)
> eax: cfda0302
> esi: 0000e
These two lines are incomplete: there should be other stuff on it.

> ds: 0018
> process swapper (pid 1, stackpage = c13153000)


> stack: 09000000 00000000 0000000 000000 00000000 00000002 00000000
>        cfda32d0 ce6cd230 0000000 ........ etc.
The stack bit doesn't matter quite so much;

> call trace: [<c0138765>]
> [<c013924>]
> [<c0137ae6>]
> [<c01382bf>]
> [<c013843a>]
but this bit matters a lot, and again it is truncated.

An accurate oops message would be useful here.  

When things die, are you getting random different error messages, or
is it always the same message?

Cheers,
 Stephen





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]