Hi, I have a Dell Dimension 2400 and (I just called Dell) the hard drive has indeed (taking deep breaths, one....two....three....) crashed. Much of my data---the absolutely vital stuff---is backed up. However, I am a writer and I am under a tight deadline and the physical process of reinstalling (repartitioning for the dual boot, setting up dual monitors and Xinerama, reinstalling the variety of programs I use (Xine, Skype, Audacity...getting English/Japanese/Korean input in email and word processing...) terrifies me. I may have also lost a few files and the "sent" emails for about 2 months.

My questions are these:

1. Dell says that the error is a #7, and the disk is spinning too fast for the data to be readable. Has anyone had this error before and is there a possible way to fix it? If I bought an identical drive and swapped the motor or something, would that work?

2. Assuming I _can_ get the disk to work, what is the best way to migrate the data from the various Linux partitions to the new hard drive? Should I set it up as a slave and just cp? Or back up to DVD in .tar files which I then restore? Can I copy the entire /usr/bin/ directory without problems?

3. If I can't get the drive back, how in the future might I avoid some of this hassle should a drive ever crash again? I have a small home network...anyone have slick ideas? The key thing is that I just don't have time to be a _real_ sys admin on my own system, especially when a deadline is looming...I need quick and relatively user-friendly ways to automate backup and recovery....



Thanks so much!

If the drive motor is indeed going haywire there may be no fix other than attempting a replacement which I would never recommend unless you already know what you are doing. My understanding is that even if you think you are purchasing the same model hard drive it may be a different lot and not have the exact same parts. To me reinstalling is a lot less scarry that ripping open my hard drive and possibly turning it into a heap of useless parts.
I would think the smart thing to do would be to focus on recovering any personal files you do not have backed up.
I had a drive fail on me and it would fail to boot. I was able to install it as a slave, set the BIOS to not detect it and boot Linux from the master drive. I was able to boot up and copy my files off the bad drive.
While researching ways to save data off a bad drive it seems it is common as a last resort to put the drive in the freezer (protected in some way I am sure) for some time. Google a bit and see if it is something you want to try.
You did not mention how Dell was able to determine "error is a #7, and the disk is spinning too fast for the data to be readable". You can get a utility from the drive manufacturer to test the drive.

I have tried the trick of putting the drive in the freezer before and was able to get data off the drive so give that a try.

As for preventing this from happening again I found a tool called partimage at http://www.partimage.org/ that will allow you to make a backup image of your system to another system. I found it fairly straight forward to use. I used the rescue cd at http://www.sysresccd.org/ to boot my pc I was backing up and ran partimaged on another pc. was able to back up the pc accross the network. It also restored fine.

Recovering from a hard drive crash unfortunately is no sure bet. If the cause of the problem is rooted in a mechnical problem, such the spin rate being off, realistically that is not a user-fixable issue. If I were you I'd suggest trying two things.

1. Use a Live CD such as Knoppix to see if the data is accessible, and if so, back it up immediately. You'll still need to deal with the fact that your drive can't be trusted anymore and needs to be replaced, but at least your data will be safe. In terms of the actual backup process, whatever process you're comfortable with should be fine. As a writer I'd assume virtually all of your critical data would be text documents, and a plain copy would work just fine.

2. If that is unsuccesful, you can try the fsck command (file system check). man fsck for more details.

Long term, you probably already know the best way to avoid similar situations in the future: more frequent backups. In truth the failure rate of quality-brand drives is pretty low, but they do occur and it's almost always at the worst possible time. Good luck with things

First, backup your data by your favorite method. Get your system bootable, whather you use something like Knoppix or Damn Small Linux (good but different) or you go buy another disk and put your old one on the same cable as your CD-ROM.

Second, figure out who made the old disk and then get their diagnostic tools (you may need wine to use this tool). Often it will have a low-level format utility in it. I have had lots of luck using this utility with a couple different drives, so give it a shot overnight. It will take some time (several hours).

Complete your reinstall.

Using the low-level format utility often has fixed my bum disks for good. If it isn't thrashing (making a loud clanking noise), you may can get a couple years out of it after this. If it is thrashing, it's gone. Also, I have had to do a low-level format on one of my disks two or three times before that disk stabilized, but it did, so give it a couple more shots if it fails initially. In the mean time, find you another hard disk.