Bad block on drive using RAID-1

Bad block on drive using RAID-1

Post by Sacha J. Bernstei » Thu, 02 Nov 2000 07:13:53



I have a RAID-1 setup going using kernel version 2.2.17, and it appears that
there is a bad block on one of the two drives.  I don't consider this a
whole-drive failure, and would like to continue using that drive.  Here are
the messages I'm receiving:

ckraid returns:

read error on device file /dev/hdg4, block #23461956 = 0x1660044

and the kernel reports:

Oct 31 03:14:34 bardeen kernel: hdg: read_intr: status=0x59 { DriveReady
SeekComplete DataRequest Error }
Oct 31 03:14:34 bardeen kernel: hdg: read_intr: error=0x40 {
UncorrectableError }, LBAsect=57189447, sector=46923912
Oct 31 03:14:34 bardeen kernel: end_request: I/O error, dev 22:04 (hdg),
sector 46923912

So, is there anything I can do to tell the kernel to not use this block (on
either drive) and to rebuild the RAID set skipping the bad section?  What
are my options other than replacing an entire drive that only has a small
section bad?

Thanks,

Sacha


 
 
 

Bad block on drive using RAID-1

Post by David Efflan » Sun, 31 Dec 1899 09:00:00



Quote:

>I have a RAID-1 setup going using kernel version 2.2.17, and it appears that
>there is a bad block on one of the two drives.  I don't consider this a
>whole-drive failure, and would like to continue using that drive.  Here are
>the messages I'm receiving:

>ckraid returns:

>read error on device file /dev/hdg4, block #23461956 = 0x1660044

>and the kernel reports:

>Oct 31 03:14:34 bardeen kernel: hdg: read_intr: status=0x59 { DriveReady
>SeekComplete DataRequest Error }
>Oct 31 03:14:34 bardeen kernel: hdg: read_intr: error=0x40 {
>UncorrectableError }, LBAsect=57189447, sector=46923912
>Oct 31 03:14:34 bardeen kernel: end_request: I/O error, dev 22:04 (hdg),
>sector 46923912

>So, is there anything I can do to tell the kernel to not use this block (on
>either drive) and to rebuild the RAID set skipping the bad section?  What
>are my options other than replacing an entire drive that only has a small
>section bad?

IDE drives normally reserve some unseen disk space to automatically map
into any bad sectors.  The fact that you see a bad sector that was not
seen before is a bad sign, it could mean that the drive is self
destructing.

If the drive manufacturer has a diagnostic program, see what it says.  I
had a WD 2.1GB drive once, that appeared to pass even the distructive test
of writing all 1's or 0's to the drive.  But when I tried to write mixed
data to the latter 3/4 of the drive, scandisk in Win95 kept finding more
bad sectors, and the non-destructive WD diagnostics at that point said
that the drive was bad with "no error sites remaining".  The warranty
replacement worked flawlessly except for a temporary glitch below.

One other thought with your number of drives is overheating.  When I was
getting lost interrupt errors I discovered that my case was packed with
lint in front of the inlet fan.  Clearing that resolved my drive interrupt
problems with no loss of data (other than gaps in the logs), even though
it had gotten to the point that I could not 'shutdown'.

--

http://www.autox.chicago.il.us/  http://www.berniesfloral.net/
http://cgi-help.virtualave.net/  http://hammer.prohosting.com/~cgi-wiz/

 
 
 

1. Finding out if there are bad blocks in the bad blocks list

Hi

I'm trying to find out whether there are any bad blocks in the ext2fs
bad block list. I think there might be, because I had a whole lot of
ext2fs error messages popping up during a compile, and when I removed
the directory I was working with and ran e2fsck -cf /dev/hdc6, it told
me the filesystem was modified, even though none of the regular checks
had failed (I'd previously run e2fsck -f /dev/hdc6, so all the duplicate
blocks and unattached inodes and stuff had been sorted out). I think the
drive might still be under warranty, so if there are bad blocks I'll
take it back.

TIA
Bruce
--
Please remove the ".nospam" from my address before replying
/--------------------------------------------------------------------\
| Bruce Merry (Entropy)            | bmerry at iafrica dot com       |
| Proud user of Linux!             | http://www.cs.uct.ac.za/~bmerry |
|               When you make your mark on the world,                |
|                 watch out for people with erasers.                 |
\--------------------------------------------------------------------/

2. Help why do I get these errors?

3. How do you add a bad block to e2fs bad block list?

4. mail for root

5. RAID - can it work if bad blocks exist

6. IP Masquerade question

7. Bad Blocks on IDE drive, fsck doesn't fix it

8. RadioShack to Feature Microsoft Wares

9. How do I check for bad blocks on my hard drive?

10. Bad block on disk drive

11. Bad Blocks on SCSI Drive

12. frequency of dup/bad blocks on hard drive

13. Bad blocks not getting remapped on Maxtor drive