Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Help with disk crash (long)




Hello to all -- 

========
Summary -- I lost a hard drive, and am looking for a kind soul
knowledgeable in the ways of the ext2fs file system,
to help me recover my users files.
========

During the storms here in Boston last weekend, I lost a hard drive -- 
probably due to power failure at just the right(wrong) time.

Of course, it contains  /home, and of course, there are files
there that aren't backed up.

All the info on structure of drives (eg, e2fsck -l) was, of course,
on the corrupted drive.  

e2fsck fails to repair it.

Here's what I know about the drive -- hdc

>From an earlier log, this is presumably right -- 

Jun 30 18:48:26 ppro kernel: ide_setup: hdc=638,255,63 
Jun 30 18:48:29 ppro kernel:     ide1: BM-DMA at 0xfcb8-0xfcbf, BIOS settings: hdc:pio, hdd:pio 
Jun 30 18:48:29 ppro kernel: hdc: Maxtor 85250D6, ATA DISK drive 
Jun 30 18:48:29 ppro kernel: hdc: Maxtor 85250D6, 5009MB w/256kB Cache, CHS=638/255/63 
Jun 30 18:48:29 ppro kernel:  hdc: hdc1 hdc2 hdc3 hdc4 < hdc5 hdc6 hdc7 hdc8 > 
Jun 30 20:48:14 ppro kernel: ide_setup: hdc=638,255,63 
Jun 30 20:47:32 ppro fsck: /dev/hdc5: clean, 48678/257040 files, 923070/1028128 blocks 
Jun 30 20:47:32 ppro fsck: /dev/hdc3: clean, 3606/14056 files, 42899/56227 blocks 
Jun 30 20:47:32 ppro fsck: /dev/hdc8: clean, 111641/602112 files, 2118925/2401686 blocks 
Jun 30 20:48:17 ppro kernel:     ide1: BM-DMA at 0xfcb8-0xfcbf, BIOS settings: hdc:pio, hdd:pio 
Jun 30 20:48:17 ppro kernel: hdc: Maxtor 85250D6, ATA DISK drive 
Jun 30 20:48:17 ppro kernel: hdc: Maxtor 85250D6, 5009MB w/256kB Cache, CHS=638/255/63 
Jun 30 20:48:17 ppro kernel:  hdc: hdc1 hdc2 hdc3 hdc4 < hdc5 hdc6 hdc7 hdc8 > 

Note hdc4 is an extended partition.  The data I want is in hdc5.  
I could loose everything else and still be moderately happy.

Here's what fstab was set to do - 
/dev/hdc3               /oldslash               ext2    defaults        1 2
/dev/hdc8               /oldslash/usr           ext2   defaults 1 2
/dev/hdc5               /home                   ext2    defaults        1 2
/dev/hdc6              swap                    swap    defaults        0 0

oldslash was / for an older installation.
After the crash, the drive looked like this -- 

Jul  7 22:20:47 ppro kernel: ide_setup: hdc=638,255,63 
Jul  7 22:17:50 ppro swapon: swapon: /dev/hdc6: No such device 
Jul  7 22:20:25 ppro mount: mount: wrong fs type, bad option, bad superblock on /dev/hdc5, 
Jul  7 22:20:41 ppro mount: mount: wrong fs type, bad option, bad superblock on /dev/hdc3, 
Jul  7 22:20:50 ppro kernel:     ide1: BM-DMA at 0xfcb8-0xfcbf, BIOS settings: hdc:pio, hdd:pio 
Jul  7 22:20:50 ppro kernel: hdc: Maxtor 87000 A8 COMET, ATA DISK drive 
Jul  7 22:20:50 ppro kernel: hdc: Maxtor 87000 A8 COMET, 1669MB w/256kB Cache, CHS=638/255/63 
Jul  7 22:20:50 ppro kernel:  hdc:hdc: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error } 
Jul  7 22:20:50 ppro kernel: hdc: read_intr: error=0x40 { UncorrectableError }, LBAsect=0, sector=0 
Jul  7 22:20:50 ppro kernel: end_request: I/O error, dev 16:00 (hdc), sector 0 

Possibly, a Bad Thing happened here -- maybe e2fsck tried to "repair" the drive with
incorrect information when power came back up?

After a second powerup, it looked like this

Jul  7 23:40:30 ppro kernel: ide_setup: hdc=638,255,63 
Jul  7 23:26:46 ppro swapon: swapon: /dev/hdc6: No such device 
Jul  7 23:40:08 ppro mount: mount: wrong fs type, bad option, bad superblock on /dev/hdc5, 
Jul  7 23:40:24 ppro mount: mount: wrong fs type, bad option, bad superblock on /dev/hdc3, 
Jul  7 23:40:33 ppro kernel:     ide1: BM-DMA at 0xfcb8-0xfcbf, BIOS settings: hdc:pio, hdd:pio 
Jul  7 23:40:33 ppro kernel: hdc: Maxtor 85250D6, ATA DISK drive 
Jul  7 23:40:33 ppro kernel: hdc: Maxtor 85250D6, 5009MB w/256kB Cache, CHS=638/255/63 
Jul  7 23:40:33 ppro kernel:  hdc: hdc1 hdc2 hdc3 hdc4 < > 

The model is right, again, but the extended partition is gone.  Also, can't find superblock  on any except
/dev/hdc1.  That partition actually looks ok.

Here's what fdisk says -- 

# fdisk -l /dev/hdc
Warning: ignoring extra data in partition table 5
Warning: ignoring extra data in partition table 5
Warning: ignoring extra data in partition table 5
Warning: invalid flag 0x6a49 of partition table 5 will be corrected by w(rite)

Disk /dev/hdc: 255 heads, 63 sectors, 638 cylinders
Units = cylinders of 16065 * 512 bytes

   Device Boot    Start       End    Blocks   Id  System
/dev/hdc1             1         7     56196   83  Linux
/dev/hdc2             8       135   1028160   83  Linux
/dev/hdc3           136       142     56227+  83  Linux
/dev/hdc4           143       638   3984120    5  Extended
/dev/hdc5   ?     81888    165783 673889067+  57  Unknown


1-4 actually look ok here, but the superblock info seems corrupted.

I tried the testdisk program 

TestDisk 3.5, Data Recovery Utility by Christophe GRENIER, July 12 2001
grenier at nef.esiea.fr
http://www.esiea.fr/public_html/Christophe.GRENIER/
Linux version

which gave some more detailed info that I can't interpret, and
the log file is quite long so it's not included here.
(I'll happily pass it along to anyone who's interested).

  My default plan is to delete the partitions 1-4
and recreate them with fdisk, using the start/end blocks as above.
Then create a single hdc5 partition in the extended partition and pray
the data will appear.

So -- if you've read this far you probably have a clue what I should do.
Thanks in advance for any help.

			Steve







 





-
Subcription/unsubscription/info requests: send e-mail with
"subscribe", "unsubscribe", or "info" on the first line of the
message body to discuss-request at blu.org (Subject line is ignored).




BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org