On 10/31/06, Bruno Wolff III <bruno@xxxxxxxx> wrote:
On Tue, Oct 31, 2006 at 12:13:19 -0500, Michael Wiktowy <michael.wiktowy@xxxxxxxxx> wrote: > What could be causing this massive slowdown on the fsck? There didn't Bad sectors on the disk.
I finally made it through a pass using e2fsck -f -c -c -v -y /dev/hda2 and there was about a 1000 bad blocks. I am making another run to see if that number is stable but I don't know if e2fsck reports the total number of or the number of new bad blocks at the end of the run.
You should also look at how many sectors have been remapped. If it is more than a handful that indicates serious problems.
Is there an easy and quick way to query this info from an installed disk rather than running a badblock scan?
If you have safely copied the data off, you can also run badblocks from a live cd and write test the whole drive. That may help get some bad sectors remapped.
Yes ... that was the first thing that I did and I was running from a LiveCD of an unmentionable distro to do these checks. I was under the impression that e2fsck called the badblocks program when you used the -c flag. I'll look into that man page though. Thanks for your help. I still would like to know if reinstalling FC6 from scratch and choosing to format the partition will undo all this work and clear this badblocks table though. I suppose to get rid of the cruft I could just do the dreaded rm -Rf / before installing and not modify or format the existing partitions just to be sure. I recall there being a "Check partition for bad blocks" flag available on installation (that would make installations take *forever*) but I don't recall seeing it recently. /Mike
