Re: bad blocks in under 2.5 years! So much for HP ...

From: milkyway (d0mufasa_at_hotmail.com)
Date: 11/03/05 

Date: 3 Nov 2005 12:24:39 -0800

Hello all and thanks sooooooooooooo much for responding ;-P

To the questions:

>>You did apply all patches, including kernel?
I did not apply any patches to the kernel at all. When doing the uname
-a on the system, what I have written above is what I got. I have been
using straight what was out of the SuSe Box.

>> The problem might be well related to reiserfs, have seen it

>>Same here, although I admit there seems to be a lot less complaining about
>>reiserfs generally in message traffic I read than just a couple years ago.

>>Still, I ask for ext2 or ext3 when I do installs, myself. It's good enough
>>for my general purposes, although some folks really swear by reiser. There
>>is much to be said for choice, and especially in Linux/FOSS. :)

When I had made the partitions, I thought they were of the fstype of
ext2 - or something for Linux altogether, I had no idea that this
reiserfs was involved at all until I had looked at the error messages

>>In the long run, your best bet is to replace the defective hard disk.
>>Frequently, once a drive begins developing bad blocks, the damage is
>>progressive -- new bad blocks will continue to crop up until the disk is
>>so damaged that it can't be used at all.

That is what I am in the process of doing. I took it to a shop near
where I live. The person
said that they would work on saving the data but I had saved what I
wanted onto these small
disks you plug into the USB ports. Anywho, I have 3 of them and I saved
what I wanted on
2 of them (duplication). The only I did not save were my database files
but the DB was
so small, I think it can be reproduced.

My concern now is contacting the guy tomorrow and telling him not to
worry about saving
the data and to just focus on replacing (successfully) the disk drive.

>>Actually, I prefer SCSI anyway because I do some IO intensive
>>database stuff where the ability to run drives simultaneously to avoid seek
>>contention and to queue commands is important.

I am on the hunt for a new computer just in case the things with this
laptop don't work.
I am just looking for something to finish my project with - so - I am
thinking I will save the SCSI approach for later.

>>What brand of drive is it, anyway? I'm still trying to figure out which make of
>>notebook drive has respectable reliability, which seems to be sliding downhill
>>everyday

I don't know the brand. I just took it in today for repair. Again, I
need to contact the
guy and just ask him to replace the drive altogether.

Question: Is there a stable brand of hard disk that I should shoot for?
Any to
avoid? ANY type of advice on this point would be great!

>>Here is one possible solution:
>>[snip]
>>4) you can now mount that backupfile via the loop device and recover the
>>data on another computer or new hard disk.

The thing is, I can mount the partitions. I used the Suse Emergency
Recovery approach and was able to get into a mode where I could mount,
etc. I was able to mount the partition that was causing me issues. When
running the "badblocks" command, several blocks were listed.

Again, I was able to mount the partition though. Could that mean
something? I had read that if one was able to write to the blocks, then
somehow the problem would correct itself. There were several blocks.
Right now, I don't have the system with me - it is at the shop -

TIA

Relevant Pages

  • Re: Moving Along with Fedora!
    ... but I would like to know the windows equivalent of the Linux ... you "mount" them to the tree. ... all the letter designations for drives have disappeared. ... a drive, the partitions have device named like /dev/hda1 or /dev/hda2, ...
    (Fedora)
  • Re: bad blocks in under 2.5 years! So much for HP ...
    ... >>database stuff where the ability to run drives simultaneously to avoid seek ... Is there a stable brand of hard disk that I should shoot for? ... I can mount the partitions. ...
    (alt.os.linux.suse)
  • Re: bad blocks in under 2.5 years! So much for HP ...
    ... >>database stuff where the ability to run drives simultaneously to avoid seek ... Is there a stable brand of hard disk that I should shoot for? ... I can mount the partitions. ...
    (comp.os.linux.hardware)
  • Re: Fedora (Core 1) trying to mount ntfs HD as root, and other problems
    ... > drives. ... and I get a help screen of syntax of mount commands. ... > winmodem isn't working to download anything into the linux partitions now. ... Try to solve problems yourself first: man <command> first, ...
    (alt.os.linux)
  • Re: Cannot write to drive "Media is write protected"
    ... those hard disk partitions wich i have partitioned with vistas own disk ... Vista is the drive Vista is installed on; ... one or two hard drives because they are ...
    (microsoft.public.windows.vista.hardware_devices)