Quote:
Are you saying that you do want to recover the 6 files on the drive that are "stored" on the sectors that now are "bad" ? It might be possible at least partially. You will get the parts of the files that are stored on the good sectors and if you do patch sysfile 93 to stop re-location, etc you can use something like PC-3000 DE and do multiple reads on those bad sectors to see if hopefully one of the reads can get some data. If not at least you will still have the rest of the file. If it's a movie you might get away with just a little bit of damage (missing frames) ...
That's what I asked about months ago in that other thread I linked. Last time I tried to work on it, the “reallocated” count was indeed constantly rising, even when the drive was idle. What tools are necessary to patch sysfile 93 ? (Maybe you already replied that, I didn't re-read the whole thread, but there was a lot of new information to process, and the general feeling from the replies I got was quite discouraging.) I don't have PC-3000 DE and certainly couldn't afford it right now. Most I could do is try again with ddrescue or HDDSuperClone. All six files are TV recordings, which were available online for just one week. I could find one on eMule P2P network (strictly identical file, same size and checked with WinHex) ; among the others, some are severely damaged, some have only a few small bad spots (see the ddrescueview[*] screenshots in that SuperUser thread linked above), one of them has only one bad 4KB cluster : I was hoping to at least get that one 100% recovered, for OCDness' sake, but indeed, just one missing cluster is barely noticeable.
I still consider myself lucky, especially now that I know about those drives' reputation to deteriorate quickly once they start to have issues. And anyway I'm hoarding so many such TV recordings (documentaries, movies and whatnot) that I'm actually watching a very small proportion of them, so I might have never watched those ones anyway !
We live in crazy hectic times, it's become strictly impossible to keep up with everything, and it's getting worse on a daily basis...
[*] I had the idea of using ddrescue's own “generate” mode to generate logfiles purposefully aimed at locating the bad spots within each file, and showing them simply and clearly when opened in ddrescueview. I don't know if there's another commonly used method to do this, but that one works like a charm. Copied from the SuperUser thread :
“I had the idea to use ddrescue's “generate” mode, which creates logfiles (or mapfiles as they're called now) by parsing the output and considering that totally empty sectors are unread sectors, marked “?”, the rest being marked “+”. Since ddrescue expects an input file and an output file, but only the output file is actually parsed in that mode, I created dummy input files with this command, which copies only 1MB but extends the size to the size of the output files (just to save time and space) :
ddrescue -s 1048576 -x 883789824 201707222358.mp4 201707222358copy.mp4
Then I ran the “generate” command :
ddrescue -G 201707222358copy.mp4 201707222358.mp4 201707222358-generate.log
And then I opened those files with ddrescueview [...].”
Quote:
You don't even have to go to eBay fo find "what looks like high" prices for data recovery - viewtopic.php?f=3&t=36887
Very interesting thread. I've found these statements to be relevant to what I wrote earlier :
“The money I make on donors is what pays for my time spent wiping, testing and keeping inventory on thousands of drives.”
“One possibility is to apply a "no result - no fee" or a "no result - small administrative fee" policy and take more margins on you work in order to cover the risk of an unsuccessful recovery.”
“For a short while, I contemplated sharing that risk, i.e., if the recovery is successful, they pay the full cost of the donor and I keep the donor, but if not, I absorb half the donor's cost and also keep the donor. Given the high cost of donors, I've decided that's just not feasible for a small-time DR shop like mine. If they don't want to pay for the donor no matter the recovery outcome, I'll just wish them best of success with whatever they decide to do.”
And the whole last post is about the idea of balancing out the price asked for each client with the risk of unrecoverable failure which costs time and sometimes parts with little to no reward. So just looking at the “profit” generated for re-selling drives, or to perform an easy task, is not enough to get a broader picture of how profitable the whole venture is.
Quote:
Another example - viewtopic.php?f=3&t=36904 - Flash pen drives, way smaller than usb - Logical (up to 32Gb) - $175 ; Minor (logical/quick physical fix) - $350 CAD.
Indeed that seems quite high, even in Canadian Dollars. But the first post says that increasing the price results in having less clients, all else being equal. And there is competition, so after some time the prices tend to average out, at least in populated areas where such services are relatively easy to find. Also, large capacity flash memory based devices are relatively new, the knowledge and tools required to deal with them are still rare, and as the last post says it's still a small part of the DR business — all those factors contribute to explain why the fees are higher in proportion of the device's capacity (which by the way must be weakly correlated with the amount of “active” work required for the whole task).
Quote:
This sort of recovery normaly is done by the use of R-Studio or similar to copy the data that can be recovered to another place.
And how do you explain that people are willing to pay such a high fee to get it done by a professional, if they have the possibility of doing it themselves by paying about half that price for an advanced software which they can keep for later use ? Or even using a free software, since something like Recuva (free version) is enough to get back deleted files in most situations (although it's not fool-proof as I said and can cause disasters by itself). On the other hand, I know for a fact that many people renounce paying for a professional recovery, even for important personal files, and prefer to either bring the device to 10 friends / relatives / colleagues who claim to be “good with computers”, or hastily read forums on teh Internetz and attempt some do-it-yourself method, then another, and so on... or directly open the drive hoping that it can be fixed like a car engine, y'know, tightening some bolt, putting some grease... at which point even a professional recovery becomes nigh impossible.
Quote:
The problem with head stack replacement is that you do need parts and often those parts will get damaged. You don't have to worry about that with firmware. With PCBs it's easier because even if you do have to buy/use a donor one it's not very often that the new PCB will get damaged and you can re-use it for the next job. Heads will last 3-4 jobs and will end up dying. Some Seagate DM will typicaly require 3 sets of heads to clone ...
Yes, I understand that this basically what you do, i.e. firmware fixing / PCB replacing, and I get why it's generally less tricky. But with the fees you ask, and the average amount of cases you get in a year, do you consider this activity “highly profitable”, “profitable”, or merely “sustainable” ?
Quote:
The more data you can provide about a drive like pre-amp, firmware, etc the more chances are for a "professional" to get the "match" that he needs. If you list a drive as "Samsung 500 GB" and you don't add extra info like firmware, pre-amp type, etc data recovery people trying to find a particular drive for PCB/HEAD swap will not know if the drive that you are listing is compatible for what they want or not ...
Alright, but that's still a large amount of work, as quoted above, and it requires large investments at the begining to deal with enough drives in order to present a complete enough catalog for professional to choose units according to their constantly changing needs. So I don't see this as a quick and easy way to get rich by exploiting people's ignorance and carelessness (even if those two things can be found in spades !).
Quote:
You can sell PCB of your broken drives and get some $$$ out of it !!! Looks reasonable. You can even offer ROM transference/adaptation fee and you are good to go ! Just try it !
I've seen people looking for PCBs on classified ads websites, but usually they expect to pay about 10€ for one. (And besides I suspect that most of them expect this to be the cure when their drive's actual failure is completely unrelated with the PCB. Sometimes I answer to such an anouncement just to ask them basic questions about their drive's symptoms, and imply that replacing the PCB might not be any useful.)
Quote:
There are firmware failures that are easy to predict, like when you start to see several pending sectors on some WD drives
Like, how many pending sectors should be raising alarms, and on which particular models ?
Quote:
or when you pick up a F3 ARCH drive you know that those will fail for sure !
Well, again, that's a bit exaggerated, since some of these drives work fine for years, and misrepresenting the global situation, since drives from all brands and models do fail no matter what, even when used in the best possible conditions. I now know that they are generally reputed as having an inferior design or being made with inferior materials, I will certainly check the ones I have regularly, just like I regularly check all my HDDs, I will take it into account when deciding which type of data goes to which drive (I have drives from almost every current manufacturer, except Toshiba — I've had one in the past, purchased new, which failed after just a few months, and the customer service was non-existant, I had to ask the seller for a refund, and was kinda lucky I could get it at all because said online seller disappeared shortly thereafter : that alone is a good reason to avoid that brand altogether), but I see no reason to just dump drives which are still operating perfectly, or re-sell them to people who most likely won't know anything about it (and I sure will have a hard time selling them if I say in the anouncement : “I'm selling all my Seagate hard disk drives because the overwhelming consensus among professionals on HDDGuru is that they are utter and complete crap, good luck with those P.O.J.” !).
Quote:
Also as i stated just buy a lot of drives, like for example 20 drives for $40 or $50 and then refurb those drives and sell them for $20 or $30 each ... same for PCBs if you can't "recover" the drive .... Now advertize some data recovery work, get something like R-Studio and start to charge people lololol
Well, those are tempting ideas, I'll think about it !
But advertising is definitely not my strong point, and it seems to be a crucial part in such a venture...
Quote:
Of course that if you already have something like an IT business you can for example start to offer "basic" data recovery services. Just clone the drives with hddsuperclone and extrat the data with R-Studio and you will manage to get at least some cases.
I already did that a few times, but so far it's been excruciatingly painstaking tasks for the money I asked. Last summer, I found a guy who had a 1TB HGST drive which had failed all of a sudden, he said that it was clicking, which sounded bad, I didn't think that I could do anything, but he was adamant that he would not pay the high fee of a “true” professional recovery (
here are the prices list of the only data recovery company in the whole region where I live if you're interested), so he gave me “carte blanche” for the 50€ service fee I proposed ; and to my surprise, when insisting a few times, unplugging and plugging it again, the drive would stabilize, I could run ddrescue and start imaging it, albeit very slowly (after reading a ddrescue guide in french mentioning that writing to a NTFS partition was not recommanded and could by itself cause slowdowns for large volumes,
I asked about it on SuperUser, proceeded to create an Ext4 partition and resume the cloning, saw only empty data in the preview but didn't know yet that it was perfectly normal, started all over again, surprisingly there were no skipped area at all in the first 150GB or so, as opposed to the first attempt, but then it slowed down to a crawl way before it did on the first attempt). It started to become more unstable, and I gave up after about four days (and I could barely do something else as it was very hot, I tried to keep the drive cool with a laptop computer cooler and cold packs...), it had recovered about 250GB, but by luck the client had only about 200GB worth of data on it, and when I analyzed the output with ddru_ntfsfindbad from ddr_utilities, I found out that only about 130 files in personal folders (i.e. excluding system and programs files) were partially or totally corrupted. So I extracted all the files with R-Studio, and then, by my own initiative, and without charging anything more for it (well, I thought of it as practice, and a compensation for my rookie mistakes, without which I may have been able to get a slightly higher recovery ratio !), noticing that there were many duplicated files on the drive, I set about fixing as much of the damaged files manually, by first detecting them with DoubleKiller (scanning the whole recovery for files with the same size, disabling the CRC analysis), then opening the two or three instances in WinHex, and seeing it the corrupted area(s) from one instance could be copied-pasted from another, and the other way around : in most cases, indeed, either one instance was complete, or two instances were damaged the corrupted sectors were in different spots, so I could rebuild the original complete file. At the end of this quite daunting task, there remained about 20 corrupted files, among which six low quality DVDrip movies, a bunch of unimportant Web pages elements, so only a handful of personal videos were partially or completely unreadable (I also repaired a few MP4 files with a missing index using Grau Video Repair). And then as the dude was dumb enough that he didn't find the recovery, although I had made a complete report with detailed explanations taking into account his very basic knowledge level (he expected to find his personal folder or Windows' “desktop” right away, just like when turning on his computer, not nested deep in the directory tree), I spend an extra hour showing him through remote assistance where his files were, and trying to show him a few tools and tricks to backup them and organize them, which he probably forgot right away...
Quote:
Even if you can't fix firmware and damaged heads you can still charge $100 or $200 for your logic recovery and pay the taxes and be on the legal size without extra worries as you do already have on IT shop open to the public, etc ...
I currently don't have a legitimate business to begin with, which is indeed making things more complicated... But in my area, I've seen data recovery fees by computer repair shops or freelance technicians (for logical recovery obviously) closer to 50€ than 100-200€.