[solved with recent e2fsck] external usb disk won't mount, error

[doctordruidphd]

Re: external usb disk won't mount, error

e2fsck -b 32768

maybe add the -n

based on what you are seeing from the various commands, that's my guess.

Hey, if someone out there has done this, please don't hesitate to jump in.
Experience, even if limited, beats guessing from man pages any day.


Edit: not meaning to put you off, Telengard, but to be perfectly honest, I haven't had this problem, or done this fix before. So I'm slogging through the man pages as well, and learning a lot along the way. I guess I would want to at least try to find someone who has done this before I hit the final key on it. So maybe a post on the ubuntu forum is warranted -- they have a somewhat larger readership, and maybe someone there has actually done this, and knows where the landmines are. It's something I want to know how to do, too, so I'm following closely.

Edit further: I did some googling again, and found this:
http://www.cyberciti.biz/faq/recover...ted-partition/
(substitute e2fsck for fsck -- fsck is a frontend that calls e2fsck for ext file systems)
Problem here is I tried this on a usb drive I know is good, and the command
sudo e2fsck -b 32768 -n /dev/sdc2
produced a boatload of errors, while
sudo e2fsck /dev/sdc2
came up with no errors. So I don't know what's going on -- the file system I am checking is actually empty.

Also, there is a more involved procedure here:
http://kezhong.wordpress.com/2009/06...lock-recovery/
interestingly, this uses dd to copy one of the backup superblocks to the primary. Again, I'd sure hesitate on this one until I heard from someone who had successfully done it.

Sorry I can't offer definitive help, as I said, I'm here mostly to learn on this one.

[Telengard]

Re: external usb disk won't mount, error

e2fsck -b 32768

maybe add the -n

tmp$ sudo e2fsck -b 32768 -n /dev/sdb1
e2fsck 1.40.8 (13-Mar-2008)
e2fsck: Device or resource busy while trying to open /dev/sdb1
Filesystem mounted or opened exclusively by another program?
tmp$ fuser /dev/sdb1
tmp$

Hey, if someone out there has done this, please don't hesitate to jump in. Experience, even if limited, beats guessing from man pages any day.

Edit: not meaning to put you off, Telengard, but to be perfectly honest, I haven't had this problem, or done this fix before. So I'm slogging through the man pages as well, and learning a lot along the way. I guess I would want to at least try to find someone who has done this before I hit the final key on it. So maybe a post on the ubuntu forum is warranted -- they have a somewhat larger readership, and maybe someone there has actually done this, and knows where the landmines are. It's something I want to know how to do, too, so I'm following closely.

No kidding. This kind of thing is never simple, and more eyes on the problem can only help. I had a bad experience once with Windows chkdsk which cost me 30GB of priceless data, thus my reluctance to blindly trust an automated program. The data is on the platters, I just have to either:

• fix the filesystem, or
• recover it forensic-ally

Problem here is I tried this on a usb drive I know is good, and the command ... came up with no errors. So I don't know what's going on -- the file system I am checking is actually empty.

Want to reproduce my problems? You could copy a few gigs of files to the disk and suddenly unlpug the data cable somewhere in the middle of the process. Wash, rinse, repeat. Should put your disk in a similar condition to mine.

Sorry I can't offer definitive help, as I said, I'm here mostly to learn on this one.

Your moral support is woth its weight in gold! I'm learning, too.

http://kezhong.wordpress.com/2009/06...lock-recovery/
interestingly, this uses dd to copy one of the backup superblocks to the primary.

Why use dd to restore the primary superblock from the backup superblock? I thought e2fsck could do the job.

More interesting reading:

• Linux: Find Alternative Superblocks:
• Understanding UNIX / Linux filesystem Superblock:
• Linux: Recover Corrupted Partition From A Bad Superblock:

[Telengard]

Re: external usb disk won't mount, error

tmp$ sudo dumpe2fs /dev/sdb1 | grep -i superblock
dumpe2fs 1.40.8 (13-Mar-2008)
 Primary superblock at 0, Group descriptors at 1-10
 Backup superblock at 32768, Group descriptors at 32769-32778
 Backup superblock at 98304, Group descriptors at 98305-98314
 Backup superblock at 163840, Group descriptors at 163841-163850
 Backup superblock at 229376, Group descriptors at 229377-229386
 Backup superblock at 294912, Group descriptors at 294913-294922
 Backup superblock at 819200, Group descriptors at 819201-819210
 Backup superblock at 884736, Group descriptors at 884737-884746
 Backup superblock at 1605632, Group descriptors at 1605633-1605642
 Backup superblock at 2654208, Group descriptors at 2654209-2654218
 Backup superblock at 4096000, Group descriptors at 4096001-4096010
 Backup superblock at 7962624, Group descriptors at 7962625-7962634
 Backup superblock at 11239424, Group descriptors at 11239425-11239434
 Backup superblock at 20480000, Group descriptors at 20480001-20480010
 Backup superblock at 23887872, Group descriptors at 23887873-23887882
tmp$

These superblocks all match those in reply #14. Here's something more interesting:

tmp$ sudo dumpe2fs /dev/sdb1 | tail
dumpe2fs 1.40.8 (13-Mar-2008)
Group 1191: (Blocks 39026688-39059455)
 Block bitmap at 39026688 (+0), Inode bitmap at 39026689 (+1)
 Inode table at 39026701-39027212 (+13)
 0 free blocks, 16384 free inodes, 0 directories
Group 1192: (Blocks 39059456-39062038)
 Block bitmap at 39059456 (+0), Inode bitmap at 39059457 (+1)
 Inode table at 39059469-39059980 (+13)
 0 free blocks, 16384 free inodes, 0 directories

dumpe2fs: /dev/sdb1: error reading bitmaps: Can't read an block bitmap
tmp$

So there's some filesystem corruption for you. Bitmaps are used for things like block allocation tables, the way the filesystem tracks which blocks are in use.

Feels like I'm chasing my tail. Time to try something a bit more risky.

[Telengard]

Re: external usb disk won't mount, error

e2fsck is bugged
e2fsck reports device as busy with -b parameter - Theodore Ts'o
Which explains why I've had no luck getting it to do anything :P

Sadly, I still have no darn luck with it. The following is after booting from a 9.10 Karmic live CD.

ubuntu@ubuntu:~$ sudo e2fsck -n -b 32768 -B 4096 /dev/sdb1
e2fsck 1.41.9 (22-Aug-2009)
e2fsck: Bad magic number in super-block while trying to open /dev/sdb1

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>

ubuntu@ubuntu:~$ sudo e2fsck -n -B 4096 /dev/sdb1
e2fsck 1.41.9 (22-Aug-2009)
e2fsck: Group descriptors look bad... trying backup blocks...
e2fsck: Bad magic number in super-block when using the backup blocks
e2fsck: going back to original superblock
e2fsck: Device or resource busy while trying to open /dev/sdb1
Filesystem mounted or opened exclusively by another program?
ubuntu@ubuntu:~$ sudo e2fsck -b 32768 -B 4096 /dev/sdb1
e2fsck 1.41.9 (22-Aug-2009)
e2fsck: Bad magic number in super-block while trying to open /dev/sdb1

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>

After reading Linux EXT2/EXT3 Superblock Recovery « Kezhong's Weblog, I've begun to think that maybe the superblock thing is a red herring. I am not getting the message "Couldn’t find valid filesystem superblock." from dumpe2fs. As shown in reply #17:

dumpe2fs: /dev/sdb1: error reading bitmaps: Can't read an block bitmap

[doctordruidphd]

Re: external usb disk won't mount, error

Why use dd to restore the primary superblock from the backup superblock? I thought e2fsck could do the job.

My understanding of this is they are using dd to copy one of the backup superblocks to the primary. That is what e2fsck is supposed to do, but it doesn't look like it wants to.

I tried compiling the scandrive program, and it's no go. Too many errors to work through.

Did you try this one -- mount with the sb= option? Should be harmless if it fails:

You can also use superblock stored at 32768 to mount partition, enter:
# mount sb={alternative-superblock} /dev/device /mnt
# mount sb=32768 /dev/sda2 /mnt

The correct form seems to be:
sudo mount -t ext2 -o sb=32768 /dev/sdb1 /media/testing

But I am starting to suspect, as you are, that the superblock thing isn't the real problem.
May have to start working through the google hits on *linux ext2 recovery magic number", there are a lot of them.

[doctordruidphd]

Re: external usb disk won't mount, error

My experiment, for what it's worth:

1. Create msdos partition on usb device /dev/sdc (gparted)
2. Create ext2 file system on /dev/sdc1 (gparted)
3. Mount /dev/sdc1 on /media/junk:
sudo mount -t ext2 -o users,rw,exec /dev/sdc1 /media/junk
4. Fix permissions and ownership (done)
5. Copy files over (done)
6. sync & unmount (done)
7. run e2fsck

$ sudo e2fsck -f /dev/sdc1
e2fsck 1.41.14 (22-Dec-2010)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
junk: 288/245760 files (4.5% non-contiguous), 282512/981760 blocks

8. run dumpe2fs

$ sudo dumpe2fs /dev/sdc1 | grep super
dumpe2fs 1.41.14 (22-Dec-2010)
Filesystem features: ext_attr resize_inode dir_index filetype sparse_super large_file
Primary superblock at 0, Group descriptors at 1-1
Backup superblock at 32768, Group descriptors at 32769-32769
Backup superblock at 98304, Group descriptors at 98305-98305
Backup superblock at 163840, Group descriptors at 163841-163841
Backup superblock at 229376, Group descriptors at 229377-229377
Backup superblock at 294912, Group descriptors at 294913-294913
Backup superblock at 819200, Group descriptors at 819201-819201
Backup superblock at 884736, Group descriptors at 884737-884737

9. Scramble the primary superblock

$ sudo dd if=/dev/zero count=1 bs=4096 seek=0 of=/dev/sdc1

10. verify the dirty deed

$ sudo dumpe2fs /dev/sdc1 | grep super
[sudo] password for greenman:
dumpe2fs 1.41.14 (22-Dec-2010)
dumpe2fs: Bad magic number in super-block while trying to open /dev/sdc1
Couldn't find valid filesystem superblock.

11. Try mouonting with the sb option -- won't do it (why not??)

$ sudo mount -t ext2 -o sb=32768 /dev/sdc1 /media/junk
mount: wrong fs type, bad option, bad superblock on /dev/sdc1,
missing codepage or helper program, or other error
In some cases useful info is found in syslog - try
dmesg | tail or so

12. Try e2fsck

$ sudo e2fsck /dev/sdc1
e2fsck 1.41.14 (22-Dec-2010)
e2fsck: Superblock invalid, trying backup blocks...
junk was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong for group #4 (32254, counted=17918).
Fix<y>? yes

Free blocks count wrong for group #5 (32013, counted=0).
Fix<y>? yes

Free blocks count wrong for group #6 (32254, counted=0).
Fix<y>? yes

Free blocks count wrong for group #7 (32013, counted=25252).
Fix<y>? yes

Free blocks count wrong for group #8 (32254, counted=17918).
Fix<y>? yes

Free blocks count wrong for group #9 (32013, counted=0).
Fix<y>? yes

Free blocks count wrong for group #10 (32254, counted=0).
Fix<y>? yes

Free blocks count wrong for group #11 (32254, counted=0).
Fix<y>? yes

Free blocks count wrong for group #12 (32254, counted=0).
Fix<y>? yes

Free blocks count wrong for group #13 (32254, counted=0).
Fix<y>? yes

Free blocks count wrong for group #14 (32254, counted=30147).
Fix<y>? yes

Free blocks count wrong for group #28 (32254, counted=29932).
Fix<y>? yes

Free blocks count wrong (964406, counted=699248).
Fix<y>? yes

Free inodes count wrong for group #0 (8181, counted=8180).
Fix<y>? yes

Free inodes count wrong for group #4 (8192, counted=8145).
Fix<y>? yes

Directories count wrong for group #4 (0, counted=2).
Fix<y>? yes

Free inodes count wrong for group #7 (8192, counted=8189).
Fix<y>? yes

Directories count wrong for group #7 (0, counted=2).
Fix<y>? yes

Free inodes count wrong for group #8 (8192, counted=8016).
Fix<y>? yes

Directories count wrong for group #8 (0, counted=2).
Fix<y>? yes

Free inodes count wrong for group #28 (8192, counted=8142).
Fix<y>? yes

Directories count wrong for group #28 (0, counted=1).
Fix<y>? yes

Free inodes count wrong (245749, counted=245472).
Fix<y>? yes


junk: ***** FILE SYSTEM WAS MODIFIED *****
junk: 288/245760 files (4.5% non-contiguous), 282512/981760 blocks

13. Try mounting:

sudo mount -t ext2 -o users,rw,exec /dev/sdc1 /media/junk

works -- it's all there.

NOTE: The method of using dd to copy a backup superblock to the primary, i.e.E

sudo dd if=/dev/sdc1 count=1 bs=4096 skip=32768 seek=0 of=/dev/sdc1

did not work, no matter how I play with the parameters. Probably the same reson the sb option doesn't work with mount, unknown what that reason is.

[Snowhog]

Re: external usb disk won't mount, error

NOTE: The method of using dd to copy a backup superblock to the primary, i.e.E

sudo dd if=/dev/sdc1 count=1 bs=4096 skip=32768 seek=0 of=/dev/sdc1

did not work, no matter how I play with the parameters. Probably the same reson the sb option doesn't work with mount, unknown what that reason is.

Have you seen/read Advanced Find ext2 ext3 Backup SuperBlock

[doctordruidphd]

Re: external usb disk won't mount, error

Yeah, basically the same thing I tried. The dd superblock restore still doesn't work, but the -b option for e2fsck does.

Maybe try a current (oneiric) livecd, newer version of e2fsck might work better?

[Telengard]

Re: external usb disk won't mount, error

Did you try this one -- mount with the sb= option? Should be harmless if it fails:

You can also use superblock stored at 32768 to mount partition, enter:
# mount sb={alternative-superblock} /dev/device /mnt
# mount sb=32768 /dev/sda2 /mnt

The correct form seems to be:
sudo mount -t ext2 -o sb=32768 /dev/sdb1 /media/testing

tmp$ sudo mount /dev/sdb1 /media/Data/ -t ext2 -o sb=32768
mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
    missing codepage or helper program, or other error
    In some cases useful info is found in syslog - try
    dmesg | tail or so

tmp$ dmesg | tail
[  97.942411] input: Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball as /devices/pci0000:00/0000:00:1d.3/usb3/3-1/3-1:1.1/input/input7
[  97.986287] input,hidraw1: USB HID v1.11 Mouse [Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball] on usb-0000:00:1d.3-1
[44770.792857] usb 3-1: USB disconnect, address 3
[44771.509055] usb 3-1: new full speed USB device using uhci_hcd and address 4
[44771.710198] usb 3-1: configuration #1 chosen from 1 choice
[44771.720166] input: Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball as /devices/pci0000:00/0000:00:1d.3/usb3/3-1/3-1:1.0/input/input8
[44771.744464] input,hidraw0: USB HID v1.11 Keyboard [Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball] on usb-0000:00:1d.3-1
[44771.751156] input: Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball as /devices/pci0000:00/0000:00:1d.3/usb3/3-1/3-1:1.1/input/input9
[44771.788951] input,hidraw1: USB HID v1.11 Mouse [Unicomp Inc. Ruffian v2.5 Keyboard w/ Trackball] on usb-0000:00:1d.3-1
[51181.279306] VFS: Can't find an ext2 filesystem on dev sdb1.
tmp$

No dice

May have to start working through the google hits on *linux ext2 recovery magic number", there are a lot of them.

Ugh ...
I tried searching on dumpe2fs "error reading bitmaps: Can't read an block bitmap", but found nothing of use in the first page or so.

11. Try mouonting with the sb option -- won't do it (why not??)

'tis a mystery to myself as well

12. Try e2fsck

$ sudo e2fsck /dev/sdc1
e2fsck 1.41.14 (22-Dec-2010)
e2fsck: Superblock invalid, trying backup blocks...
junk was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
. . .

Wait, you actually got e2fsck to do something useful! [me=Telengard]wonders about the different version numbers of e2fsck.[/me]

tmp$ lsb_release -d
Description:  Ubuntu 8.04.4 LTS
tmp$ e2fsck -V
e2fsck 1.40.8 (13-Mar-2008)
    Using EXT2FS Library version 1.40.8, 13-Mar-2008

ubuntu@ubuntu:/tmp$ lsb_release -dc
Description:  Ubuntu 9.10
Codename:    karmic
ubuntu@ubuntu:/tmp$ e2fsck -V
e2fsck 1.41.9 (22-Aug-2009)
    Using EXT2FS Library version 1.41.9, 22-Aug-2009

Maybe try a current (oneiric) livecd, newer version of e2fsck might work better?

It is probably worth me giving it a try, especially considering the findings documented in post #18.

Have you seen/read Advanced Find ext2 ext3 Backup SuperBlock

Hey, Snowhog! Glad to see doctordruidphd and I aren't the only ones here

I'm getting a new hard disk to in case I must resort to forensics. Have a look and see if it seems good:
http://www.newegg.com/Product/Produc...82E16822324043

My gf is being admitted to the hospital tomorrow for a heart-related procedure, so I don't expect to be able to work more on this until Tuesday or so.

[Snowhog]

Re: external usb disk won't mount, error

Maybe [SOLVED] Cannot Mount Volume - External Hard Drive (see #2)

[doctordruidphd]

Re: external usb disk won't mount, error

It is always best to use the most recent version of utilities like e2fsck. I believe the version cited on the page in Snowhog's post is the buggy one, and it gave the expected message.

The disk you are looking at from newegg looks OK, but I would t5ry a google for its model number plus "linux fail" or something like that, just to make sure.

Sorry to hear about your gf, hope all turns out well. Might be a good time to put this on the shelf for a day or two. We'll be here.

[Snowhog]

Re: external usb disk won't mount, error

It is always best to use the most recent version of utilities like e2fsck. I believe the version cited on the page in Snowhog's post is the buggy one, and it gave the expected message.

The referenced post referred to using TestDisk, which is why I pointed to it. Maybe TestDisk can repair the damage (latest version: 6.13, November 15, 2011).

[doctordruidphd]

Re: external usb disk won't mount, error

Unfortunately, the poster didn't mention HOW he used testdisk to repair the system. I have so far not been able to find the way. What I did find is this:

http://www.cgsecurity.org/wiki/Advan...kup_SuperBlock

but e2fsck (fsck is just a frontend to e2fsck) needs to be working for this method.
Still searching google, but all I'm finding is direct repair of FAT/NTFS. Linux repair seems to need the above method.

[Telengard]

Re: external usb disk won't mount, error

The disk you are looking at from newegg looks OK, but I would t5ry a google for its model number plus "linux fail" or something like that, just to make sure.

If the drive arrives DOA obviously it will go back. Otherwise I plan to test it very rigorously before entrusting my data to it.

Sorry to hear about your gf, hope all turns out well. Might be a good time to put this on the shelf for a day or two. We'll be here.

Everything turned out fine.
Her doctor was just being overly cautious due to her family history.

Maybe TestDisk can repair the damage (latest version: 6.13, November 15, 2011).

Returning to this project, I find I'm just about ready to give up hope on e2fsck, although maybe I'll have a go with a more recent version just to make sure. After that, TestDisk seems like the next logical step. After that, maybe Foremost.

I can't thank you all enough for the support. Even though it seems like we're just feeling around in the dark, I've already learned some things.

[doctordruidphd]

Re: external usb disk won't mount, error

From everything I have been able to find with google, TestDisk by itself cannot repair an ext file system; what it can do is identify where the backup superblocks are. e2fsck does the actual repair. Since the "busy" bug seems to have been fixed in later versions, going with a current livecd seems the best option. There do exist commercial solutions to this problem that you can find with google, I can't vouch for any of them, but the cost vs. your time with something like foremost is something to consider.

Before going that route, I would try a post on the ubuntu forum, though not in the "hardy" section, but in the General Help section, for reasons of broader readership. Also if the problem persists you might try contacting the e2fsck developer through the sourceforge web page:
http://e2fsprogs.sourceforge.net/
You never know -- it might be an easy fix for someone who knows all the tricks.

Edit: one more (ha!) thing: I notice from the website that 1.42 is the latest e2fsprogs version; what's in oneiric is 1.41. Before undertaking the compiling yourself, you might want to go with the (admittedly experimental) precise cd, as it uses 1.42.

oneiric:
$ apt-cache policy e2fsprogs
e2fsprogs:
Installed: 1.41.14-1ubuntu3
Candidate: 1.41.14-1ubuntu3
Version table:
*** 1.41.14-1ubuntu3 0
500 http://us.archive.ubuntu.com/ubuntu/ oneiric/main amd64 Packages
100 /var/lib/dpkg/status

precise:
# apt-cache policy e2fsprogs
e2fsprogs:
Installed: 1.42~WIP-2011-10-16-1ubuntu1
Candidate: 1.42~WIP-2011-10-16-1ubuntu1
Version table:
*** 1.42~WIP-2011-10-16-1ubuntu1 0
500 http://us.archive.ubuntu.com/ubuntu/ precise/main amd64 Packages
100 /var/lib/dpkg/status