Filed under: Software
Totus Copy, for your dying hard drive
Performing regular backups of important data has been greatly simplified with the advent of Time Machine. For most people, in most cases, that's enough to ensure a good night's sleep. But for those situations where a hard drive in its death throes means a last minute panic to save whatever data can be salvaged, Totus Copy can make a bad day a little better.
The documentation for this new software from Dragon Forged clearly states that running it can further damage your failing drive, but that's not such a bad price to pay if it can save your data. It requires a hard disk that can still mount, and features several means of rescuing your flailing files, including transferring single files (less likely to fail than batches), minimal scanning to reduce failures and error correction with no time limit on transfers. It can also target specific directories or even file types.
Thankfully, I'm unable to test the application right now. But if you're in a situation where this might (now, or eventually) be a lifesaver, Dragon Forged is offering TUAW readers 25% off the $20 pricetag until June 1st. Just enter coupon code "BHBB0008F" when you purchase. More information, documentation and a trial download are all available on the product page.
_thumbnail.jpg "Take a stroll down memory lane")
Reader Comments (Page 1 of 1)
mark said 1:15PM on 5-27-2008
Paying for a program that hasn't even been tested once? No thanks. You should be grateful for beta tester results, and not be charging them.
Reply
Ed said 1:21PM on 5-27-2008
mark, you're missing the point. Brett, the author of this post is not the person who wrote the program. I'm sure the program's developers have smashed many a hard disk then used their program on it.
Reply
Josh Wardell said 1:31PM on 5-27-2008
What about software solutions for drives that don't mount in the OS?
Reply
Piotr Malecki said 2:00PM on 5-27-2008
For those drives the only solution is either dd [http://en.wikipedia.org/wiki/Dd_(Unix)] or a professional data recovery service [anywhere from a couple hundred to a couple thousand dollars, usually no guarantee, but sometimes money back if they can't recover anything]. If you have critical data, especially business data that doesn't have a backup or a recent one, then a couple thousand dollars is nothing.
Rod said 2:13PM on 5-27-2008
Try Spinrite for those cases. As long as the drive is recognized by the bios, it can usually help. I've used it on my office machines and it's saved my butt so many times I've lost count.
RobK said 6:28PM on 5-27-2008
I do not recommend Spinrite as the first approach to backup data on a failing drive. Spinrite does NOT backup data to a separate drive. Spinrite tries to repair the drive which can be VERY risky if the drive hardware is near failure. (Sorry Steve Gibson.)
I would also NOT recommend dd as dd is not designed to backup data from flaky drives.
I would first recommend a program that tries to backup the data. Once the data is backed up, you can try programs the Spinrite to repair the drive.
There are a couple of choices to backup a flaky drive:
1. There is an excellent FREE Open Source program called GNU ddrescue. You can install in via Darwin Ports. Or just compile the source code yourself.
(see http://www.gnu.org/software/ddrescue/ddrescue.html)
Or just boot off a Linux LiveCD (like SystemRescue - http://www.sysresccd.org/Main_Page) and use dd_rescue. dd_rescue works on all filesystems.
2. For those who want to pay $$, try CopyCatX which will retry bad sectors a number of times until the data is copied.
RobK said 6:31PM on 5-27-2008
Oops. I forgot to mention that there are two free programs with similar names -- GNU ddrescue and dd_rescue. You want to use GNU ddrescue. NOT the older obsolete dd_rescue. I know it is confusing.
Brett said 2:35PM on 5-27-2008
If you really have a dying drive and have access to a pc and can remove the hard drive, spinrite (from grc.com) is the answer for you.
Reply
Sarbo said 3:35PM on 5-27-2008
This app may help you copy some files off but if your disk is failing you are better off using Drive Genius and Data Rescue II to make a disc image and recover from the image of the drive. Totus might be helpful if you have one critical file you need to get off but if you need to recover more than a few files you should make a single pass to image the disc and then work with recovery software on that.
Reply
Kyle said 3:50PM on 5-27-2008
Sarbo,
Totus Copy does this more effectively then Drive Genius, or Data Rescue. Both those software titles will attempt to bulk copy the data onto a disk image. Totus Copy will copy the files (granted not to an image) and have the ability to target files of a specific kind, skip unneeded directories, ignore invisible system files, and skip application bundles. Plus without resumable copying the other utilities will hang and fail on groups of bad blocks.
Sarbo said 5:30PM on 5-27-2008
@Kyle,
I'm talking about a device level image not an OSX image. I'm not saying Totus isn't a good tool, but when a device is failing you have to triage. As I said earlier, if a there are a few key files then Totus could be a good solution, but if you need maximum recovery then you should make a device level image. For really damaged disks, hardware level reverse cycle clone may be efficacious, but that is not a consumer end-user feature--well, at least not until it is bundled with the tools needed to extract data from the clone.
Reply
Tim Rutter said 10:42PM on 5-27-2008
System requirements: "A failing hard disk that still mounts".
Ah yeah not that good ddrescue and plain old dd dont have that limitation, If the device attaches bus and can be seen at /dev/diskX , You will be able to image the whole drive or just one partition.
Then mount the image and then fix the filesystem.
See: http://www.macosxhints.com/article.php?story=20050302225659382
Reply
RobK said 11:45AM on 5-28-2008
The macosxhints article was written before GNU ddrescue was developed.
I would not recommend using dd. There are many drawbacks with using dd. It's slow even on the error-free stuff since it's doing tiny reads and writes. It spends a lot of time chewing through the erroneous parts of the drive, rather than reading as much error-free stuff as it can, THEN going back to do the hard stuff.
The GNU ddrescue first recovers data from all the good sectors on the drive and then goes back and tries to recover data from the bad sectors on the drive. You will likely recover more data this way. If you try to recover data from bad sectors before all the data on the good sectors has been recovered, you risk losing the data on the good sectors if the drive hardware is failing.
Another advantage of GNU ddrescue is that it enables you to do multiple passes to recover your data. For example:
# first, grab most of the error-free areas in a hurry:
ddrescue -B -n /dev/rdisk0 /dev/rdisk1 rescued.log
# then try to recover data from the bad sectors without pushing the drive too hard:
ddrescue -B -r 1 /dev/rdisk0 /dev/rdisk1 rescued.log
(the "-r 1" is necessary to make it retry the failed areas up to 1 time)
# then try to recover as much of the dicy areas as possible:
ddrescue -B -r 10 /dev/rdisk0 /dev/rdisk1 rescued.log
(the "-r 10" is necessary to make it retry the failed areas up to 10 times)
There is agood interview with the developed of GNU ddrescue here:
http://blue-gnu.biz/content/interview_gnu_ddrescue_039_s_antonio_diaz_diaz
Reply