Slashdot Mirror

I second that. Or dd_rescue, which sounds like the same thing and does function similarly, but is a different implementation.

Another very useful tool is testdisk, which can operate on the drive itself or more safely on a disk image made with ddrescue or dd_rescue. I used testdisk to retrieve several hundred GB of files from an almost brand new disk that had a load of bad blocks. A friend had just backed up all their files to an external drive, deleted the original, and then planned to copy things back after a reformat of their main system drive when they discovered the external had gone bad. Talk about bad timing. Lesson learned. They now make two backups. Anyway, it took weeks to read and re-read the files from the disk, and we didn't get 100% of it back, but we got over 90% of the files.

Assuming the hardware isn't completely dead, then dd_rescue and testdisk are the way to go. I've recovered stuff from very badly messed up disks with that combination. One drive had serious hardware problems, and would only mount intermittently if it was propped in a certain orientation. The risk is that the drive will get worse as you attempt recovery, and lessen your chances if you ever go the professional route; but if you aren't going to pay 4-figures for professional data recovery, then it's probably the closest you'll get to doing it yourself (with free tools).

In another instance,

Nicholas Harbour, who at the time was working for the Department of Defense Computer Forensics Lab (DCFL)

wrote a loving modified dd that writes to multiple files and streams to multiple programs at the same time. The program, dcfldd, also introduces the sorely missed VERIFY operation, and even block-by-block hashes, ( dcfldd Man page)

Maybe someone will combine this with dd_rescue, ddrescue and dd_rhelp to make the ultimate "Convert and Copy" utility :-)

Ah and I can dream of SCTP support too :-)

It's a legacy, maybe - but just try to find a command in Linux to rescan your SCSI-bus.
Well, there isn't. Instead, you are supposed to echo some values into certain parts of the procfs, or run some vendor-specific script.

It's a legacy, maybe - but just try to find a command in Linux to rescan your SCSI-bus.
Well, there isn't. Instead, you are supposed to echo some values into certain parts of the procfs, or run some vendor-specific script.

i'm too paranoid to use stuff like this for backups.. sure 25 gigs is nice but whats the use if i just burn everything in 35 copies on the disc incase one part becomes unreadable?

as media starts to hold more, i just start creating more copies of the same backup on the disc. bluray/hdvd scares me because if it gets scratched you lose so much more than if a cd gets scratched

For those that don't know, PAR2 files are parity files that can efficiently reconstruct missing or damaged blocks in your archive. If you have more PAR2 recovery blocks than damaged blocks, then you can completely reconstruct all of the damaged files in your archive. The best newbie explanation I've seen is the "PAR & PAR2 files" section from Slyck's Guide To The Newsgroups.

If I'm backing up to a data DVD-R (capacity 4,706,074,624 bytes), I'll leave around 4GB of space for the actual data and fill the rest (to the brim) with the PAR2 files that I created for that data. I name the PAR2 files starting with the letter 'z' so that they get burned on the outer edge of the DVD. When creating the PAR2 files, I choose a block size that is a multiple of 2048 bytes because that is the block size of a DVD sector.

Some easy-to-use tools to create PAR2 files:

• Windows: QuickPar (freeware)
• OS X: MacPar deLuxe (shareware, $15)
• Linux/OS Independent (GTK): GPar2

Some DVD data recovery software (to get every readable block off a damaged disc):

• Windows: IsoBuster (recovery functions are free)
• Linux/OS X (command line): dd_rescue

Thanks, WuphonsReach.

On plain dd, bs=4096 runs the risk of not overwriting up to 3 sectors at the end of the disc. If this worries you, but you still want the performance gain, use dd_rescue, with a command of the form dd_rescue /dev/random /dev/hdx which will use 64KB blocks until they're too big to function, then fall back down all the way to single sectors.

dd_rescue skips read errors, so maybe it skips write errors too (though maybe not, given its intended use). If it doesn't, it should at least provide a good start for a tool that does.

Disk drives generate heat (and fail first from it). To keep your drives excercised, check out Bonnie.

It is similar to dd, but it does not exit on I/O errors. So it is perfect for pulling as much data as possible from a bad drive. It also has a nifty optimization wherein it uses 16kB blocks to copy until it gets an error, then it goes down to 512 byte blocks so it can get as close to the corrupted sectors as possible.

I just used this to recover all but 500kB of data from a 120GB drive that went bad. The method was simple, albeit long:

• Buy new 250GB drive(WD 7200rpm 8MB cache for $160 from Best Buy(stupid $90 rebate)
• fdisk, mkfs, mount new drive on /mnt/new
• dd_rescue -A /dev/hdc1 /mnt/new/home.ext3 (-A so it fills in any errors with zeroes)
• wait many hours(did my taxes, :)
• fsck -p /mnt/new/home.ext3
• mount home.ext3 /mnt/loop -o loop
• cp -a /mnt/loop/* /mnt/new

Here's something to try:

This month, I had two disk-failure on a 1.0 TB software raid5 with ReiserFS.
I was able to copy most of the data with dd_rescue and myrescue.
By the time I was finished mucking around, I had dome mkraid -f several times, so there were spots of missing data on the disk. The filesystem would not mount. So I used resierfsck --rebuild-tree, and once it completed five days later, I was able to mount the filesystem, with most of the files intact.

• scsicontrol: scu, sg-utils
  
• scsiha: scsiadd or rescan-scsi-bus.sh
  
• stacker: mtx or scsi-changer

SCSI HOW-TO: 9.3 Tape driver (st)