Another way is to use a livecd to run fsck on the drive.
fsck -y -c /dev/locationofdrivePartition
-c will also run a simple read-test on sectors of the drive, in case there are
'bad-sectors' - This will take a wee while. -y will just answer yes to any and
all questions (which may be undesirable to some, but in case are lots of read
errors, will attempt to fix them (marking them as bad and attempt to move data
elsewhere))
-cc will run a non-destructive read-write test, a more thorough way to test
for sector errors on the drive. This WILL take a long time!
Cloning the disk then fixing it and imaging back to another drive maybe a
quicker option tbh.
I've seen these errors before and they can cause massive IO holdups, see what
'smartinfo --all /dev/hdlocation' says regarding errors and health state.
Ian
On Monday 30 January 2012 18:16:48 Vic wrote:
> > 1) Is that it for the drive
>
> Yes.
>
> > 2) Is that it for the fs but the drive can be reused (i.e. a reinstall)?
>
> Probably not.
>
> The error you posted is a failure to read the journal. That means you
> might have lost any recent writes, but the data may be largely preserved -
> so far.
>
> What I do in these situations is to copy the partition image somewhere
> else (using dd_rescue or similar) and then e2fsck it there.
>
> You should get all or nearly all of your data back. Linux filesystems are
> really quite resilient.
>
> Vic.
>
>
> --
> Please post to: Hampshire@???
> Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
> LUG URL: http://www.hantslug.org.uk
> --------------------------------------------------------------
--
Please post to: Hampshire@???
Web Interface:
https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL:
http://www.hantslug.org.uk
--------------------------------------------------------------
