Data recovery

RegEdit

New Member
Reaction score
3
Location
Pacific Palisades, CA
What is the process that pro data recovery places do at just the analysis stage? I'm guessing that it's to listen to the drive, to gather SMART data about the drive and to see if R-Studio will quickly map out the contents of the drive and then recover the data.

I do simple data recovery with Windows (if possible) and then with R-Studio. Usually R-Studio will either recognize the drive or not. If not then I'm done and tell the customer to take it to a specialty recovery place. When R-Studio does work, it maps out the contents of the drive pretty quickly OR it will come across bad sectors (cyclical redundancy errors) and may start to drag on for hours. At that point should I image or clone the drive? Or should I image or clone the drive before even thinking about using R-Studio?
 
What is the process that pro data recovery places do at just the analysis stage? I'm guessing that it's to listen to the drive, to gather SMART data about the drive and to see if R-Studio will quickly map out the contents of the drive and then recover the data.

The "pro" process is completely different than what you can do with software alone (which makes recovery more effective and keeps the drive's data much safer).

We definitely use our ears, but that doesn't always tell you much as you'd think. A "clicking" sound, for example, could be caused by several different problems (firmware corruption, ROM corruption, bad heads, bad PCB, bad sectors, etc).

*A screeching + scratching sound however, always means head(s) have crashed onto platter and are scratching it. If this is the case, we stop here.

Checking SMART data is something we never do (besides a quick check to see if SMART has passed or failed), as it's not reliable and a better diagnosis is possible without this data (explained below). Connecting to R-studio would be out of the question, as involving an OS and any data recovery software at this stage would be dangerous (explained below).

Assuming it's not screeching or scratching, we visually inspect the PCB. Assuming the drive gets "ready," we connect every drive to a hardware data recovery tool. If possible, all modules + the PCB's rom information are backed up. Then the drive's firmware is checked to make sure it's healthy. Then individual heads are tested (if possible).

Often times however, the drive isn't "ready" - it's either not spinning at all, stuck in a "busy" state, clicking, or various other issues. The real problem is determined after a thorough diagnosis using the knowledge of each make/model drive gained over the years in conjunction with hardware data recovery tools that communicate with the drive's firmware (think: operating system) directly and/or using multimeters, donor parts, hot air guns, etc.

I assume clean room "pros" may open the drive's cover and assess damage if the drive is making scratching or screeching sounds or if the drive is diagnosed as having multiple bad heads (as if only 1 bad head, the best solution, if possible, would be disabling that head, cloning the drive, then attempt to swap the head stack or platters to recover the sectors from the "bad" head).

After this analysis, we disable certain aspects of the drive's firmware to help facilitate a smoother clone (i.e. clearing smart modules, disabling certain background tasks the drive is running, etc). We also fix any corrupt firmware modules and if necessary disable any bad/dead heads. We turn off smart, auto-relocation of bad sectors, and other features depending on the drive and model (and problem with drive).

Then we clone using hardware cloner. Depending on the drive's condition, we may clone "good" heads first. We may also target specific important folders, files, or file types first, then attempt to clone the remainder (this would be after imaging the MBR if PC formatted or catalog file if Mac formatted). We almost always go after a 100% image of the drive.

At that point should I image or clone the drive? Or should I image or clone the drive before even thinking about using R-Studio?
The latter. Cloning afterwards would defect it's purpose, which is preserving as much of the drive's data as possible. Once you clone as much as possible (ideally 100% or 99.99%), then you use software to scan the drive's partition, find/locate all the data, and transfer the data to a transfer hard drive. We also backup the data (in addition to the separate clone drive) to an encrypted disk image for safe keeping (and hold for ~30 days).

A streamlined outline of our process can be found here:
http://www.300dollardatarecovery.com/ourprocess.html

*2AM post, forgive any spelling & grammatical errors :)
 
The "pro" process is completely different than what you can do with software alone (which makes recovery more effective and keeps the drive's data much safer).

Connecting to R-studio would be out of the question, as involving an OS and any data recovery software at this stage would be dangerous (explained below).

Ahhh $300 DDR. I've heard about you folks as the go-to place for DR.

Unless there's some scraping heard in there, how much risk is there in simply hooking the drive up to a bench test computer to see if it's 1) recognized and healthy enough to transfer away with the command prompt or 2) healthy enough to image or clone the drive, and see if R-Studio will quickly map out the drive and then you proceed to recover the "Documents and Settings" or "Users" folder? From what I gather this may be what they do at places like Geek Squad before they give up and send it off to their lab back East.
I guess I'm also worried about legalities. When I get a computer in for repair, the first thing I test is the drive with Drive Fitness Test. If it fails I connect the drive to my bench test computer just to see if it's even recognized. If they have absolutely critical data like old baby photos or critical documents I don't bother. But often they say that their data is not THAT important. Then I might image the drive and see if R-Studio does its thing in short order. I'd say that more than 50% of the time R-Studio works, especially when I get these errors... "drive inaccessible - error performing inpage operation" or "0 bytes used / 0 bytes available space". If R-Studio doesn't work, has data really been damaged? What can happen?
 
Last edited:
Best thing to do if data is not that critical is to image that drive fast either ddrescue or ftk imager then run R-studio on the image.
If you want to lean how to fix drives you can take a long distance course $3000+shipping includes certification.

Distance Learning Data Recovery Training

Also you can get what is called a portable clean room they are not too expensive to open drives.
 
Last edited:
Also you can get what is called a portable clean room they are not too expensive to open drives.

Note that there are other "issues" which arise when opening a cover besides dust/contamination, such as cover alignment, which a portable clean room can't help with.

Ahhh $300 DDR. I've heard about you folks as the go-to place for DR.

:D

Unless there's some scraping heard in there, how much risk is there in simply hooking the drive up to a bench test computer to see if it's 1) recognized and healthy enough to transfer away with the command prompt or 2) healthy enough to image or clone the drive, and see if R-Studio will quickly map out the drive and then you proceed to recover the "Documents and Settings" or "Users" folder?

I don't know exactly how risky it is, since I never work that way, but there IS a risk. The first step, rather than connecting to test computer, should be cloning. If it's not healthy enough to clone, then it's not something you should work with (if it can't be cloned, then software recovery tools likely will not be able to recover it, and trying to will likely make the drive's condition worse).

Note that in "cloning" a drive, you are ideally skipping over bad sector on the first pass of the clone, and quickly imaging all the good sectors. Then you go back for another pass (or several passes) to work on the skipped/bad/slow sectors. If you let the software recovery programs run on the drive, they don't care which areas are good/bad and will "stress out" the drive when working on the bad sections/sectors (which could destroy the drive). Make sense?

From what I gather this may be what they do at places like Geek Squad before they give up and send it off to their lab back East.
You won't learn anything good by following what the Geek Squad does.
 
Note that in "cloning" a drive, you are ideally skipping over bad sector on the first pass of the clone, and quickly imaging all the good sectors. Then you go back for another pass (or several passes) to work on the skipped/bad/slow sectors. If you let the software recovery programs run on the drive, they don't care which areas are good/bad and will "stress out" the drive when working on the bad sections/sectors (which could destroy the drive). Make sense?
Cool. Which cloning program do you recommend for doing these strategic passes?
 
correct me if I'm wrong, but does cloning move the heads one track at a time while copying has the heads moving somewhat randomly over the platters? Is that why cloning is less stressful and somewhat safer than copying?
 
correct me if I'm wrong, but does cloning move the heads one track at a time while copying has the heads moving somewhat randomly over the platters? Is that why cloning is less stressful and somewhat safer than copying?

Exactly! And keep in mind, there is cloning and there is cloning. Cloning is good because it does exactly as you say. Cloning using the DDI is "streets ahead" since you can target MFT, target files, map and skip bad/weak heads, disable certain drive function, and much more. It's cloning on bath salts.

Check out this DDI Whitepaper for more than you probably want to know: http://deepspar.com/pdf/DeepSparDiskImagingWhitepaper3.pdf
 
I wish I could give customers a better idea as to what the risks are when I get these errors... "drive inaccessible - error performing inpage operation" or when it says "0 bytes used / 0 bytes available space". For me, these are errors that R-Studio has done well in working past.

images

raw-drive.jpg
 
Last edited:
Sorry, since I don't hook drives directly up to Windows (it's been said many times before, not to do that), I don't know what those errors mean, besides something with the drive is "bad."
 
I wish I could give customers a better idea as to what the risks are when I get these errors... "drive inaccessible - error performing inpage operation" or when it says "0 bytes used / 0 bytes available space". For me, these are errors that R-Studio has done well in working past.

images

raw-drive.jpg

Posisble error is corrupted MFT$. The Master File Table (MFT) contains metadata about every file and folders. It includes filenames, locations, size, and etc.
 
Note that there are other "issues" which arise when opening a cover besides dust/contamination, such as cover alignment, which a portable clean room can't help with.

That is true with some of the newer drives seems drive mfg's are going to lengths to prevent hard drive repairs.
 
Send it to a data recovery pro for a proper assessment before it is too late. Remember, if the price is too high, you can always turn it down and then kill the drive yourself. However, if you kill it first, there is no going back.
 
Posisble error is corrupted MFT$. The Master File Table (MFT) contains metadata about every file and folders. It includes filenames, locations, size, and etc.
Not a error in MFT. My guess is that it reads sector 0 (MBR) partially or fully, but when it gets to the boot sector, it is either missing or corrupt. This could be because of a bad sector or weak/failing heads. Either way, it should be properly diagnosed and cloned before it is completely killed.
 
Not a error in MFT. My guess is that it reads sector 0 (MBR) partially or fully, but when it gets to the boot sector, it is either missing or corrupt. This could be because of a bad sector or weak/failing heads. Either way, it should be properly diagnosed and cloned before it is completely killed.

I always appreciate info's from the master! Thanks

I'm just curious, once the image is done, With DDI I normally deactivate MBR before connecting the drive to Windows. With disable MBR, I was able to get drive and file info. Does that mean the MBR is not 100% disable?
 
I always appreciate info's from the master! Thanks

I'm just curious, once the image is done, With DDI I normally deactivate MBR before connecting the drive to Windows. With disable MBR, I was able to get drive and file info. Does that mean the MBR is not 100% disable?
At the end of sector 0 (MBR) there are four characters "55AA". When you deactivate the MBR, it chances to "55BB".
 
Back
Top