Re: end_request: I/O error
Wow. Fun. Of course, it turned out that I was wrong, and it was a dead
drive. The errors started showing up in MacOS a couple of days later.
-Josiah
On Wed, 20 Nov 2002, Albert D. Cahalan wrote:
> end_request: I/O error, dev 03:03 (hda), sector 1869542
> hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
> hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=1871206,
> sector=1869542
>
> When I run badblocks, I get all kinds of bad blocks, but I can't figure
> out how to get e2fsck to actually add them to the badblocks inode and work
> around them. I would normally just get a new hard drive (it's a 20G - came
I used to have lots of these on my Mac Cube. The fix was tedious.
Get out of X, and get used to the "dd" command. Make sure you have
the console log level set so that you get error messages on your
console. (from my memory: adjust /proc/sys/kernel/printk) Kill the
syslogd and klogd daemons; it's better to use pencil and paper for this.
Read the whole drive, noting the errors. Something like this:
dd if=/dev/hda of=/dev/null bs=1m
Then you swat each error by copying /dev/zero over it. This is of
course hazardous to your data, but what choice do you have? You'll
need dd's seek and skip options. Watch out for the units; they may
be sectors, kilobytes, blocks, or whatever.
Eventually you should be able to read the drive without errors.
Then you use apt-get to reinstall everything that got destroyed.
Reply to: