I suppose this question isn't necessarily *nix-related, but since the drive in question was formatted with ReiserFS, I would think there would be 'nix tools to do the job... Anyway, I had a hard drive that crashed recently (a Maxtor, the second time that a computer that I have built has had a Maxtor that crashed... It seems that I have much better luck with Western Digital Drives). The trouble is, I neglected to backup when the warning signs first appeared, thinking that it was a problem with my kernel only. I installed a replacement drive, and when I try to mount the old one, it says something about it "not being a directory" (I unfortunately don't recall the exact error message, but I can find out and post it if it is useful). So my question is, are there any tools or methods that exist that I can use to retrieve my data? Needless to say, this has finally taught me the necessity of backing up frequently...

You could try to capture an image of the affected partitions (before the drive goes even more south), and then try working with the image(s) instead. The dd command should handle that.

dd if=/dev/broken_partition of=my_partition_image

You can then mount the image using the loopback interface. Maybe you could even try running reiserfs repair tools on them etc. Try keeping an unaltered image of the partition around, if you can. You might need to start over from scratch, and the drive might not be possible to image at that point.

Me going off topic (as normal, for me):

Quote: Original post by Strife
Anyway, I had a hard drive that crashed recently (a Maxtor, the second time that a computer that I have built has had a Maxtor that crashed... It seems that I have much better luck with Western Digital Drives).

I was always "raised" (for lack of a better word) to choose Western Digital over Maxtor. However, I've had about the same experience with either brand's IDE drives (I've seen two of each fail in recent memory). So, I've started using Seagate just to give something else a try. So far, so good.

Quote: Original post by Anonymous Poster
You could try to capture an image of the affected partitions (before the drive goes even more south), and then try working with the image(s) instead. The dd command should handle that.

dd if=/dev/broken_partition of=my_partition_image

You can then mount the image using the loopback interface. Maybe you could even try running reiserfs repair tools on them etc. Try keeping an unaltered image of the partition around, if you can. You might need to start over from scratch, and the drive might not be possible to image at that point.

I think this is what I'll try next. I had tried to create an image using a hard disk analyzing/recovery tool I found on the Internet, but when it got to the first bad sector, it crapped out. Hopefully good ol' dd will work.

If dd craps out too, try specifying the parameter conv=noerror as that will ignore read errors.