Did my drive just die?

mijohnst · 11-23-2005, 04:51 PM

My box was acting funny so I decided to reboot it. When it started coming up I get about 20 lines of this:

Buffer I/O error on device hda2, Logical block 0
Buffer I/O error on device hda2, Logical block 1
Buffer I/O error on device hda2, Logical block 2
Buffer I/O error on device hda2, Logical block 3
Buffer I/O error on device hda2, Logical block 4
Buffer I/O error on device hda2, Logical block 5
Buffer I/O error on device hda2, Logical block 6
Buffer I/O error on device hda2, Logical block 7
Buffer I/O error on device hda2, Logical block 0
Buffer I/O error on device hda2, Logical block 1
Buffer I/O error on device hda2, Logical block 2
Buffer I/O error on device hda2, Logical block 3
Buffer I/O error on device hda2, Logical block 4
Buffer I/O error on device hda2, Logical block 5
Buffer I/O error on device hda2, Logical block 6
Buffer I/O error on device hda2, Logical block 7

mkrootdev: label / not found
mount:error 2 mounting ext3

pivotroot: pivot_root (/sysroot,/sysroot,init) failed :2
umount /initrd/proc failed :2
kernel panic: no init found. Try passing Init= option to kernel

It looks like it can't read my fstab file or something. Is there any way to get the fstab working if that's the case? When I try and load the Fedora 3 rescue disk it says that it can't find any partitions. The BOIS isn't reporting it as a bad drive or anything and if I do a badblocks test it spins the drive with no errors.

Any suggetions?

tuxdev · 11-23-2005, 05:37 PM

Have you fscked it yet? Might not be possible, if the partitions are gone. If that is the case, you need more local help.

mijohnst · 11-23-2005, 05:41 PM

I did try and fsck and an e2fsck and they both fail saying something about possible zero lenth or something. I'm starting to doubt myself now because I made a change to the FSTAB file earlier and starting to wonder if that change broke it. Are there tools to recover or create a new FSTAB file?

mijohnst · 11-23-2005, 06:52 PM

I think I know what my problem is. I found somewhere that if you're getting a pivotroot: pivot_root (/sysroot,/sysroot,init) failed :2 error, then it's because the 'initrd' is not there.

If I boot up to the rescue disk, I can mount /dev/hda1 which is my boot directory. If I try and mount /dev/hda2 though it says"

mount: Mounting /dev/hda2 on /mnt/system failed: Input/output error

slantoflight · 11-24-2005, 12:02 AM

Any easy to tell if you harddrive is physically damaged is to pull it out of its bay and listen to it. Does it sound any different than before? Is it consistent or does rapidly spin up and then abruptly spin down?

It sounds like atleast one of your partitions is corrupted. This doesn't easily happen on a journaling file system.

mijohnst · 11-24-2005, 09:06 AM

Yikes... Corrupt journaling doesn't sound good. I've pulled the drive and it sounds normal. How would I fix that one partition? I've tried e2fsck and it fails. Seems like I've had this problem before a long time ago and I got it fixed...

mijohnst · 11-24-2005, 10:34 PM

I can't figure this out... When I try to do fsck or e2fsck it fails saying that it's possibly a zero-length partition or something. How to I force it to check the partition? I'm running out of ideas... lol

Matir · 11-24-2005, 11:16 PM

Can you post the output of fdisk -l?

mijohnst · 11-24-2005, 11:35 PM

When I run that command, I get this:

Code:

Device          Boot          Start          End          Blocks                ID          System
/dev/hda1       *              1               13           104391             83          Linux
/dev/hda2                      14           29496        236822197+     83          Linux 
/dev/hda3                      29497     30515        8185117+         82          Linux Swap

When I boot to the rescue disk and try and mount /dev/hda2 using 'mount /dev/hda2 /mnt/system' I get this:

Code:

Mount: mounting /dev/hda2 on /mnt/system failed: Input/Output error.

When I try and run a e2fsck I get this error:

Code:

e2fsck: Attempt to read block from file system resulted in short read while trying tro open /dev/hda2

slantoflight · 11-25-2005, 10:35 AM

I hate to say it, but you may just have to reformat your partition. Sounds like your fscked out of options.

mijohnst · 11-25-2005, 11:38 AM

Don't say that! lol

purelithium · 11-25-2005, 11:44 AM

Try running the proprietary disk tools that come with your drive. Seatools for Seagate, Data Lifeguard for WD, Power Max for Maxtor. They usually have a drive test tool to see if your drive is physically failing.

mijohnst · 11-25-2005, 02:10 PM

Something interesting to add... I just downloaded Tom's boot disk and booted another machine of mine that's working just fine and tried to do an fsck and e2fsck and I get the same exact "superblock might be corrupt" message.

Matir · 11-25-2005, 03:21 PM

You are fscking /dev/hda{1,2,3}, not just /dev/hda, right? That often gets "superblock might be corrupt" messages.

mijohnst · 11-25-2005, 04:21 PM

I am doing /dev/hda2. I can mount hda1 just fine. Anyway, as a last ditch effore I've run a "mke2fs -S /dev/hda2" followed with "e2fsck -f /dev/hda2" and I'm seeing lots being changed and lots being removed from the tree. I hope that very little of this is the data I need off that partition...

Seems like I've had to do this before and I was able to get my data by looking through lostandfound directories. Is this a common problem with e2fs? Should I be moving all my partitions to e3fs?

Thanks for the help!