Alan Secker wrote:

> Having just polished off the a lean installation of Mandriva Spring
> 2007 with the installation of a network printe, I closed down and went
> home. On reboting this morning my /usr partition on hde8 looked
> decidedly iffy. I rebooted on knoppix and ran:
>
> e2fsck /dev/hde8 and got something like:
>
> Attempts to read block on hde8 resulted in a short read while reading
> block 148. Then similar re journal superblock and then ext3 journal for
> /dev/hde8.
>
> fsck -y -C -V /dev/hde8
>
> gave the same messages.
>
> Is it recoverable or should I reinstall over it?

A superblock should have enough duplicates to be recoverable, you may
however find some files or clutter of them, in lost+found. The main issue
seems to be your harddrive. Test it running smartctl, maybe from a knoppix
cd. Run badblocks as well but _after_ you have backed up your most
important data because a dying harddrive may as well quit completely when
stressed by badblocks. Remember, when bad blocks become visible to the OS,
the harddrive already may have used up all the spare sectors.
--
vista policy violation: Microsoft optical mouse found penguin patterns
on mousepad. Partition scan in progress to remove offending
incompatible products. Reactivate MS software.
Linux 2.6.17mm,Xorg7.2/nvidia [LinuxCounter#295241,ICQ#4918962]

On Tue, 19 Jun 2007 22:23:04 +0200, Walter Mautner wrote:

> Alan Secker wrote:
>
>> Having just polished off the a lean installation of Mandriva Spring
>> 2007 with the installation of a network printe, I closed down and went
>> home. On reboting this morning my /usr partition on hde8 looked
>> decidedly iffy. I rebooted on knoppix and ran:
>>
>> e2fsck /dev/hde8 and got something like:
>>
>> Attempts to read block on hde8 resulted in a short read while reading
>> block 148. Then similar re journal superblock and then ext3 journal
>> for /dev/hde8.
>>
>> fsck -y -C -V /dev/hde8
>>
>> gave the same messages.
>>
>> Is it recoverable or should I reinstall over it?
>
> A superblock should have enough duplicates to be recoverable, you may
> however find some files or clutter of them, in lost+found. The main
> issue seems to be your harddrive. Test it running smartctl, maybe from a
> knoppix cd. Run badblocks as well but _after_ you have backed up your
> most important data because a dying harddrive may as well quit
> completely when stressed by badblocks. Remember, when bad blocks become
> visible to the OS, the harddrive already may have used up all the spare
> sectors.

Thanks. smartctlk confirmed that the drive was OK but nothing else
worked. I reinstalled, formatting only a couple of partitions and
restored everything. It is fine now. Thanks for your advice.

Ragards

Alan

On Tue, 19 Jun 2007 10:23:31 +0100, Alan Secker wrote:

> Having just polished off the a lean installation of Mandriva Spring
> 2007 with the installation of a network printe, I closed down and went
> home. On reboting this morning my /usr partition on hde8 looked
> decidedly iffy. I rebooted on knoppix and ran:
>
> e2fsck /dev/hde8 and got something like:
>
> Attempts to read block on hde8 resulted in a short read while reading
> block 148. Then similar re journal superblock and then ext3 journal for
> /dev/hde8.
>
> fsck -y -C -V /dev/hde8
>
> gave the same messages.
>
> Is it recoverable or should I reinstall over it?

Try running 'badblocks' on the partition before you do anything.