Re: linux-kernel-digest V1 #190

From: Robert Dinse (nanook@eskimo.com)
Date: Fri Feb 11 2000 - 13:54:02 EST


On Sat, 12 Feb 2000 00:20:31 wzdd@tig.com.au wrote:
>
> From: wzdd@tig.com.au
> Date: Sat, 12 Feb 2000 00:20:31 +1100 (EST)
> Subject: Massive filesystem corruption with 2.2.14?
>
> Hi Linux users,
>
> I just finished salvaging data from a friend's RedHat 6.1 system,
> running kernel 2.2.14, after its / partition (the only Linux native
> partition on the system) became corrupt. The computer is a fairly new
> (several months) Pentium 3 system, and also has a windows 98 partition
> that wasn't affected, which makes me less inclined to consider hardware
> failure. I was wondering if anyone on this list knew anything about
> possible poor interactions with the hardware on this class of PC.
>
> The first few MB of the partition were completely obliterated - however
> none of the backup superblocks I passed to e2fsck worked either. It
> occurred sometime between shutdown and startup - amusingly enough it
> first appeared after rebooting from Windows 98, but we're not suspecting
> a Microsoft conspiracy yet.
>
> mke2fs -S followed by e2fsck recovered enough to copy almost all of
> /home onto a backup drive, but this was one of the few inodes that
> wasn't obviously corrupt (many entries were owned by seemingly random
> UIDs, had insanely huge file sizes, and / or were special files).
>
> Anyway, there are several possibilities I'm considering, including
> flakey hardware and the heat, but in the circumstances it seems more
> likely that some part of the kernel decided it didn't like some part of
> the hardware.
>
> Extra random information: BX chipset, hard drive supports UDMA-66 but
> I'm almost certain it was not enabled. Hard drive is a Seagate and is
> the primary master. The only other IDE device is a CD-ROM, which is
> secondary master.
>
> Any help or me-toos appreciated.
>
> - - Nicholas FitzRoy-Dale

     After having Windows-98 trash Linux partitions and scribble the partition
table numerous times, I finally segregated them onto their own drives, a drive
for Windows and a drive for Linux. They seem to happily co-exist that way, but
I was not able to stop Windows from corrupting Linux when it was on the same
drive. This was with Asus motherboards on a variety of different systems
ranging from a Pentium 133 to a Pentium 450, and with a variety of different
drives include WD, Seagate, and IBM, though I did at one time have a system
with a SCSI drive sharing Windows and Linux and didn't experience that problem.

     I've come to the conclusion that Windows is just plain evil.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:20 EST