Attempt to read block from filesystem resulted in short read while trying to

the_ultimate_samurai · 05-31-2010, 04:13 PM

so i pulled out a sata chord from my motherboard while it was still on (i know stupid right? i am normally religious about turning off my computer before removing a chord, often going so far as to completely unplug it and wait fro the light to go off, but i had always assumed it was just me being irrational...) well after doing so my computer locked up...i had to do a forced restart...after doing that my secondary HDD (/dev/hdb) not longer reads. fdisk /dev/hdb gives this:

Unable to read /dev/hdb
(there is more information in the actual console...):

Code:

May 31 17:01:15 debian kernel: [ 3009.176919]  hdb:<3>end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] __ratelimit: 22 messages suppressed
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] ldm_validate_partition_table(): Disk read failed.
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.176919] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.176919] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.196957] Dev hdb: unable to read RDB block 0
May 31 17:01:15 debian kernel: [ 3009.196969] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.197015] Buffer I/O error on device hdb, logical block 0
May 31 17:01:15 debian kernel: [ 3009.197068] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.197130] end_request: I/O error, dev hdb, sector 24
May 31 17:01:15 debian kernel: [ 3009.200945] end_request: I/O error, dev hdb, sector 24
May 31 17:01:15 debian kernel: [ 3009.200945] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.200945]  unable to read partition table
May 31 17:01:15 debian kernel: [ 3009.200945] end_request: I/O error, dev hdb, sector 0
May 31 17:01:15 debian kernel: [ 3009.200945] end_request: I/O error, dev hdb, sector 2615992
May 31 17:01:15 debian kernel: [ 3009.200945] end_request: I/O error, dev hdb, sector 2615992

fsck gives:

fsck.ext3: Attempt to read block from filesystem resulted in short read while trying to open /dev/hdb
Could this be a zero-length partition?
(again the same stuff as above shows up in the actual console...)

mke2fs -n /dev/hdb gives:

Code:

mke2fs 1.41.3 (12-Oct-2008)
/dev/hdb is entire device, not just one partition!
Proceed anyway? (y,n) y
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
81760 inodes, 327000 blocks
16350 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=335544320
10 block groups
32768 blocks per group, 32768 fragments per group
8176 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912

and e2fsck -b on all those backup superblocks gives:

Code:

e2fsck 1.41.3 (12-Oct-2008)
e2fsck: Invalid argument while trying to open /dev/hdb

The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

please help me, thats some of my most important data i REALLY want to recover it (some of that data has survived the crash of 4+ hdds and is over 10 years old)

unSpawn · 05-31-2010, 04:31 PM

Was the "victim drive" part of a RAID setup or LVM or did it have encrypted partitions on it? How many partitions were on it aprox? Why do you run 'mke2fs -n /dev/hdb' on the whole disk? Do you have (file, any, whatever) backups? Can you boot a Live CD and run 'testdisk /log /debug /dev/hdb', quit it, then read (and post?) the log file contents testdisk created?

the_ultimate_samurai · 05-31-2010, 04:54 PM

Quote:

Originally Posted by unSpawn

Was the "victim drive" part of a RAID setup or LVM or did it have encrypted partitions on it? How many partitions were on it aprox? Why do you run 'mke2fs -n /dev/hdb' on the whole disk? Do you have (file, any, whatever) backups? Can you boot a Live CD and run 'testdisk /log /debug /dev/hdb', quit it, then read (and post?) the log file contents testdisk created?

was not raid, probably not lvm(least i never explicitly told it to be...maybe implicit...)no ecrypted partitions aprox 2 partitions i believe (main partition and a swap partition which was roughly a gig or so i think) i ran mke2fs -n on the whole partition because i have no idea what im doing, running on /dev/hdb1 gives:

Could not stat /dev/hdb1 --- No such file or directory

The device apparently does not exist; did you specify it correctly?

hdb is the only one present in /dev neither hdb1 nor hdb5(the swap) are present in dev..
do i need a live cd for testdisk? the hdd is secondary so i can boot into linux just fine, or will it cause problems if i use it here?

as for backups...very little...just my music...the rest is all on that drive. it was a 500 gig so it was used for storing MOST my data. the only thing that could back it up is my 1tb which is also broken (due to WD being idiots)

unSpawn · 05-31-2010, 05:34 PM

Quote:

Originally Posted by the_ultimate_samurai

was not raid, probably not lvm (..) no encrypted partitions

Good.

Quote:

Originally Posted by the_ultimate_samurai

hdb is the only one present in /dev neither hdb1 nor hdb5(the swap) are present in dev..

Once a drives partition table goes belly up don't expect it to show up in /dev...

Quote:

Originally Posted by the_ultimate_samurai

do i need a live cd for testdisk? the hdd is secondary so i can boot into linux just fine, or will it cause problems if i use it here?

A Live CD might be easier as it frees the other disk as write target if necessary but if you can run testdisk fine that's OK.

Quote:

Originally Posted by the_ultimate_samurai

my 1tb which is also broken

Cool. At least that'll stop you from breaking it further ;-p

the_ultimate_samurai · 05-31-2010, 05:58 PM

well i can run testdisk but the size given for the drive is wrong, first it said ~1 GB now it has :

Disk /dev/hdb - 12785069 TB / 11627953 TiB - X6ضضضضض ضض ضض ~Vضضض

in both cases there was gibberish after, this time it seems to be hebrew...i didnt continue...

PTrenholme · 05-31-2010, 08:23 PM

You might want to try the the parted command (parted /dec/sdb) and see if you can use the "rescue" command (use help rescue for syntax) to recreate you disk partition(s). (You need to be "root" to run parted.)

the_ultimate_samurai · 06-02-2010, 08:09 PM

hmm it seems to have resolved itself. after i plugged the sata cable back in (as i had some work to do on another HDD, dont worry im fairly sure i didnt break it...) and turned it on, then turned it off again properly and removed the cable...it worked...bios musta gotten all confused by the sudden removal of a sata cable...