> I had occasion to reboot and got GRUB Error 17. I ran my Mepis LiveCD and
dmesg output shows a few filesystems the system was expecting to find.
Assuming grub was correctly installed and functional, and that nothing
and nobody tinkered with the bootloader, /etc/fstab, or the disk
partitioning, you have a problem.
If the dmesg output comes from the live CD system, you clearly have a
dead hard disk:
> [17179587.620000] VFS: Can't find ext3 filesystem on dev hda.
> [17179587.624000] VFS: Can't find an ext2 filesystem on dev hda.
> [17179587.640000] ReiserFS: hda: warning: sh-2021: reiserfs_fill_super: can
> not find reiserfs on hda
> [17179587.640000] FAT: bogus number of reserved sectors
> [17179587.640000] VFS: Can't find a valid FAT filesystem on dev hda.
The live system detected these filesystems, tried to mount them, and
got nowhere.
> [17179604.844000] hda: dma_intr: status=0x51 { DriveReady SeekComplete
> Error }
> [17179604.844000] hda: dma_intr: error=0x40 { UncorrectableError },
> LBAsect=2303280, sector=2303280
> [17179604.844000] ide: failed opcode was: unknown
> [17179604.844000] end_request: I/O error, dev hda, sector 2303280
> [17179604.844000] Buffer I/O error on device hda3, logical block 0
Hardware error. Could also be in the IDE controller or the cable.
> [17179610.216000] ide: failed opcode was: unknown
> [17179610.240000] end_request: I/O error, dev hda, sector 2303280
> [17179610.240000] Buffer I/O error on device hda3, logical block 0
> [17179614.204000] hda: dma_intr: status=0x51 { DriveReady SeekComplete
> Error }
> [17179614.228000] hda: dma_intr: error=0x40 { UncorrectableError },
> LBAsect=2303280, sector=2303280
>
> ***Dozens (hundreds?) of repeats of of the above five lines***
Unless you can find a broken cable etc, R.I.P.
Output of smartctl -son -A /dev/hda please?
Run a disk selftest with: smartctl -son -tlong /dev/hda
Check its progress and result with smartctl -a /dev/hda
But before doing anything much at all, try and retrieve your important
data now. To a different disk, of course. Stop writing to the
problematic one right now. Avoid power cycling the computer.
Don't run filesystem checks until you have verified the disk is ok. If
the disk is not ok, run filesystem tests only as last resort if you need
to copy data off, otherwise don't bother. Treat all files on disk as
damaged, until(!) you can verify otherwise by checksumming them.
If the disk unexpectedly does turn out to be ok (fixed the cable, mobo,
..?), don't bother with fsck, the correct command to use is mkfs, unless
you can confirm that the damage was definitely only minor.
Btw badblocks is often not a useful tool these days. It is useful only
in specific cases, and then only if you understand why. Advice to the
contrary is still common, but outdated by at least 10 years...
> [EMAIL PROTECTED] fdisk /dev/hda3
The correct command is: fdisk -l
Volker
--
Volker Kuhlmann is list0570 with the domain in header
http://volker.dnsalias.net/ Please do not CC list postings to me.
