[PLUG] Speaking Of VIA And IDE

[Mike Russell]

you could try putting the drive in another system, bring it up as a slave, 
then see if you can browse it to leach your files out of it.
if you don't have a spare system, i live in aloha, OR. i have some free 
time lately.

mike

At 02:47 PM 4/11/02, you wrote:
>Last year, I posted here about a kernel panic which seemed a result of the
>VIA "686 bug".  Terry Griffin helped me into a newer kernel, and the problem
>seemed to go away.  But yesterday, after some months trying to get IWill or
>Fry's to take back my misdesigned motherboard (for all the time I spent
>working with these morons, I could have bought a dozen m/b's, but I had a
>morbid desire to know if they would ever capitulate), I finally got the
>store manager at Fry's to credit me for it.  As a replacement, I ended up
>buying an MSI K7T Turbo2.  Ridiculously enough, it has a very similar
>chipset, but a woman at Pacific Solutions on Powell (who seemed
>knowledgeable enough:  she had been haunted in her own way by this VIA
>stupidity) assured me this board did not possess the same fault as my
>previous board.  Besides that, it's the only decent board I could find.
>
>While I was configuring the K7T for my dual boot system, I started Linux a
>few times and then accidentally set the CAS latency incorrectly in the BIOS,
>booted once, and suddenly started having the selfsame problem as before.
>The panic message is "VFS: unable to mount root FS on 03:44, and above that
>something about "group descriptors corrupted".
>
>What kind of an idiot are you?, you may ask.  Well, this is not the same
>chipset, at least not quite.  At any rate, I can now install Mandrake 8.1,
>which is something I couldn't do with my old m/b.  This time I delved a
>little deeper, because now I have things on my root partition that I'd
>really hate to lose.  I started the rescue system from my install CD and ran
>e2fsck on the partition, but that wouldn't start, saying it had found an
>invalid superblock.  I also tried specifying subsequent blocks (in multiples
>of 8192), but no dice.  It seems that partition is fscked once again, though
>it has been working perfectly for months now (read:  no, I don't believe the
>hard drive is gone bad).
>
>So I'm at the same loss I was at before, to know whether it's my new,
>"corrected" hardware that still sucks, or SuSE 7.1, or Linux itself (gasp!
>Well, much as I despise M$, Windows has never trashed a partition on me for
>no good reason.  I run 98SE on this same box, and although I had to reboot
>it a couple dozen times after installing this board, nothing was damaged).
>I'm running kernel 2.4.16, which was a constant all through this
>installation.  Now, feel free to speculate on what's going wrong here, but
>what I really want to know is whether there is any hope for my data.  A lot
>of it is backed up (I'm learning), but all my emails were/are in there, not
>to mention countless hours of configuration.  Is there aught I can do?
>
>--Jason Van Cleve
>
>
>_______________________________________________
>PLUG mailing list
>PLUG at lists.pdxlinux.org
>http://lists.pdxlinux.org/mailman/listinfo/plug