Can't mount external drive---interpret this error message?

[cdupree]

Okay, looks like my external hard drive is pretty screwed. Can someone who understands what's going on tell me how bad it is and what I should do about it?

[115867.131621] sd 21:0:0:0: [sdi] No Caching mode page present
[115867.131631] sd 21:0:0:0: [sdi] Assuming drive cache: write through
[115867.131639] sd 21:0:0:0: [sdi] Attached SCSI removable disk
[115867.543463] usb 1-2: USB disconnect, device number 14
[117094.280185] EXT3-fs error (device sdd4): ext3_check_descriptors: Block bitmap for group 640 not in group (block 7)!
[117094.284404] EXT3-fs (sdd4): error: group descriptors corrupted

[Detonate]

Open a terminal and try this.

sudo e2fsck -p /dev/sdd4

and see if that repairs the problem.

[cdupree]

When I do that I get the following:

redExt3: Note: if several inode or block bitmap blocks or part
of the inode table require relocation, you may wish to try
running e2fsck with the '-b 32768' option first. The problem
may lie only with the primary block group descriptors, and
the backup block group descriptors may be OK.

redExt3: Block bitmap for group 640 is not in group. (block 7)


redExt3: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY.
(i.e., without -a or -p options)

[Detonate]

Did you try running with the -b 32768 option?
Also try running e3fsck without any options.

[Snowhog]

Is the drive unmounted?

[cdupree]

The drive will not mount. I should have said more clearly in my initial statement that what I showed as results came from trying to mount the drive and getting an error message telling me to look at "dmesg | tail". The results I showed were that output.

When I run e2fsck with the -b 32768 option I get

e2fsck 1.42 (29-Nov-2011) [followed by a stream of entries like this:]
Block bitmap for group 1920 is not in group. (block 0)
Relocate<y>?

And I get something similar when I run with no options.

[Detonate]

e2fsck 1.42 (29-Nov-2011) [followed by a stream of entries like this:]
Block bitmap for group 1920 is not in group. (block 0)
Relocate<y>?

Did you answer yes? I think that will fix the problem.

[Detonate]

I think the most common causes of errors such as the one you are experiencing is that the drive was removed or improperly unmounted while file operations were being processed. If you had to do a hard shutdown with the drive mounted or if you had a power failure that is probably what happened. If we can't fix it with e2fsck we may be looking at reformatting that partition.

[cdupree]

When e2fsck asked me questions I answered yes, but there was a seemingly unending stream of those questions. So I'll probably have to reformat and lose the data, huh?

[Detonate]

Run e2fsck again with the -y option. This will cause e2fsck to automatically assume assume the yes answer to every question. If that does not fix it, yes you may have to reformat.

[cdupree]

Well, that didn't fix it, so I guess I have to reformat.

Is there a recommended or favorite tool for that? I've got "Partition Manager" and "KDE Partition Manager" in my K menu. I've generally used GPartEd in the past. Any suggestions? And thanks for the attempts to help with the disk!

[Detonate]

They all work equally well.