Thanks for the quick reply...
On Mon, 28 Sep 1998 15:22:14 -0400 "Theodore Y. Ts'o" wrote:
> Umm... you have a very sick machine.....
:-(
>
> Here are the types of errors I see in messages:
> free_one_pmd: bad directory entry 00000029
> free_one_pmd: bad directory entry 08cea798
>
> This tends to indicate memory corruption problems, and is unrelated to
> the disk problems below....
Hmmm maybe some problematic RAM? Is that possible?
Anyone out there have a utility to thoroughly test out system ram?
>
> hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hdc: dma_intr: error=0x84 { BadSector DriveStatusError },
> LBAsect=11258246,
> sector=11257940
> hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hdc: dma_intr: error=0x84 { BadSector DriveStatusError },
> LBAsect=11258246,
> sector=11257940
>
> This almost certainly means that your IDE disk has bad blocks on your
> disk. These can be mapped out so that ext2 won't use those blocks, but
> if you have a large number of bad blocks on your disk, you're likely to
> see a progressive failure of your disk, and you should replace the disk
> before it eats all of your data....
Ouch... :-(
>
> When it was with the scsi, I was getting inode errors...
>
> I can't really speak to the SCSI errors, without seeing the exact error
> messages. But given that the above free_one_pmd error messages as well,
> I'm very suspicious that there isn't something else going on.
I'll forward the scsi errors when I get back and look for them...
>
> Is this a new system, or one which has been working in the past? If
> it's been working in the past, has anything changed about the system
> recently? If it's a new system, you might want to consider getting it
> exchanged or repaired while it's under warranty, since this smells like
> some manufacturer dumped you with all of their marginal stock....
>
Hmmm it is a new system, brand new supermicro P6DBE dual 400... I tend to think
the mobo itself is good, these errors don't happen continuously. Could it be
related to bad ram?
It'd be nice at least to do some sort of ram verification, to narrow down the
issues.
Thanks
ricardo
-
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/