1 00:00:00,530 --> 00:00:05,790 Deep spar located in Ottawa, Ontario is a data recovery company 2 00:00:06,110 --> 00:00:10,210 but not in the way that you might think like you can't just 3 00:00:10,910 --> 00:00:15,310 Drop in with your dead hard drive and ask them to fix it for you 4 00:00:16,490 --> 00:00:18,490 their business is 5 00:00:18,690 --> 00:00:20,450 developing the software 6 00:00:20,450 --> 00:00:27,530 Hardware and the techniques that are used by everyone from mom-and-pop shop technicians to huge 7 00:00:27,530 --> 00:00:34,750 data recovery houses right on the other side of this wall is actually the classroom that they use to train police and 8 00:00:35,030 --> 00:00:38,310 other government agencies on data recovery 9 00:00:38,850 --> 00:00:41,890 So they sponsored our trip out here to do two things 10 00:00:42,510 --> 00:00:48,310 one to show consumers how to save money on data recovery service and 11 00:00:48,890 --> 00:00:56,770 Number two to show repair shops how they can dramatically improve their chances of recovery for their customers 12 00:01:04,560 --> 00:01:08,960 Scenario one of hard drive failure is clicking and clacking noises 13 00:01:09,540 --> 00:01:12,260 Sometimes without being detected by the system. Oh 14 00:01:15,620 --> 00:01:20,380 Okay, this happens it's probably a physical problem 15 00:01:20,380 --> 00:01:24,360 So if the data is important don't do anything 16 00:01:25,220 --> 00:01:27,760 Unplug your drive and send it to a pro 17 00:01:28,280 --> 00:01:35,200 immediately or your chances of a successful recovery will be dramatically lower or possibly even zero a 18 00:01:35,200 --> 00:01:39,740 Pro is going to use tools like these and a laminar flow 19 00:01:39,940 --> 00:01:46,380 workstation like this one or even a full-on clean room like the one that I checked out over at drive savers to 20 00:01:46,660 --> 00:01:51,860 physically swap components into the failed drive from donor drives 21 00:01:52,260 --> 00:01:59,860 Sometimes this needs to be done multiple times in cases where one bad part is causing the other ones to fail prematurely 22 00:02:00,380 --> 00:02:04,580 They'll also use professional tools from companies like deep spar 23 00:02:04,580 --> 00:02:11,760 So I didn't know it, but I was actually using disk imager 4 when I performed my own head swap in that video 24 00:02:12,380 --> 00:02:18,340 Scenario number two is an electronics problem. Maybe a broken SATA connector or 25 00:02:18,900 --> 00:02:19,980 a 26 00:02:19,980 --> 00:02:25,520 Burned chip on the PCB could be causing the drive to not be detected by your system 27 00:02:25,860 --> 00:02:28,100 the most common symptom here is 28 00:02:29,200 --> 00:02:30,660 total silence 29 00:02:30,660 --> 00:02:36,420 Now sometimes you will need to call a pro for this and if the data is mission-critical 30 00:02:36,780 --> 00:02:43,420 You should always call a pro but for nice-to-have rather than need-to-have data 31 00:02:43,600 --> 00:02:48,220 this type of failure actually exposes the platters of the drive and 32 00:02:49,140 --> 00:02:52,040 Therefore the data to much less risk 33 00:02:52,080 --> 00:02:57,940 So in some cases you can attempt a home repair if you're handy with a soldering iron now 34 00:02:58,620 --> 00:03:00,620 DeepSpar had this whole demo 35 00:03:00,660 --> 00:03:08,820 Planned for me the the TVS chip actually handles over current protection for the drive and can fail due to an external issue 36 00:03:08,980 --> 00:03:10,160 so just 37 00:03:10,160 --> 00:03:16,040 Removing it from the PCB can possibly brick the drive because you're removing the over current protection 38 00:03:16,100 --> 00:03:22,140 But in many cases it can allow the data to be recovered if it's hooked up to a known good system and power source 39 00:03:22,140 --> 00:03:28,720 But we are throwing that demo out the window because I brought DeepSpar an unexpected present 40 00:03:29,040 --> 00:03:30,500 This is 41 00:03:30,500 --> 00:03:35,200 Terran's personal dead hard drive that failed when we hooked it up to a 42 00:03:35,540 --> 00:03:40,660 Faulty power supply cable while he and I were at the office late upgrading his system 43 00:03:40,980 --> 00:03:45,540 Now he would likely be quoted three to five hundred US dollars 44 00:03:46,020 --> 00:03:52,240 Minimum to fix this and since he's Terran and he doesn't remember exactly what's on it 45 00:03:52,280 --> 00:03:59,660 He's not gonna pay that much to get it back. So we're gonna see how we do with a DIY grade repair. Yarek, you ready? 46 00:04:01,810 --> 00:04:05,430 So what we're gonna do first is we're gonna verify the diagnosis now 47 00:04:05,430 --> 00:04:08,830 I'm already pretty sure that we're gonna power up this drive 48 00:04:09,590 --> 00:04:11,880 And we're gonna hear a whole lot of nothing 49 00:04:12,200 --> 00:04:17,040 So that means there's a good chance that this is an electronics problem then 50 00:04:21,200 --> 00:04:23,200 Okay, so we power down 51 00:04:25,540 --> 00:04:33,160 Now unfortunately swapping out a PCB on a drive isn't as simple as going to the hard drive PCB store and buying one 52 00:04:33,580 --> 00:04:34,080 Yeah 53 00:04:34,080 --> 00:04:36,080 If only life were that easy, right? 54 00:04:36,820 --> 00:04:38,820 Look at him. He's defeated. He's defeated 55 00:04:39,760 --> 00:04:45,000 So eBay and Craigslist are your friends now, fortunately in Terran's case 56 00:04:45,320 --> 00:04:49,680 He already had a donor drive that looks like it's gonna be suitable 57 00:04:49,680 --> 00:04:54,360 So there's a couple things to watch for you need the drive model to be the same 58 00:04:54,640 --> 00:05:00,600 so not just this drive model, but also this model number right here, so these look good and 59 00:05:01,480 --> 00:05:03,980 You even need the PCB review 60 00:05:04,080 --> 00:05:09,360 To be the same now these two are a slightly different color here, but that's not important 61 00:05:09,360 --> 00:05:14,100 What matters is that they're both revision a and they both have exactly the same model 62 00:05:14,100 --> 00:05:17,040 So these should theoretically be suitable 63 00:05:18,300 --> 00:05:24,340 Okay, now we're at the point where most amateurs would screw up this kind of a recovery. They'd go. Okay 64 00:05:24,340 --> 00:05:28,400 Well now we take the good PCB put it on the bad drive and bippity boppity off to the races, right? 65 00:05:29,080 --> 00:05:30,140 wrong 66 00:05:30,140 --> 00:05:34,140 data density on modern drives is so high that the 67 00:05:34,500 --> 00:05:40,480 Manufacturing tolerances are nowhere near good enough that every drive will just work out of the factory 68 00:05:40,580 --> 00:05:43,540 so they have to build in compensation and 69 00:05:44,340 --> 00:05:47,460 Calibration that is unique to your individual drive 70 00:05:47,580 --> 00:05:54,120 It's usually stored on a little 8-pin ROM chip that looks like this and that specialized data 71 00:05:54,380 --> 00:05:57,720 Has to be brought over. So you want to show us how to 72 00:05:58,580 --> 00:06:03,280 Desolder and resolder a ROM chip. All right, let's do it. Let's not screw these up. Which one's which? 73 00:06:04,340 --> 00:06:06,340 D on this one. Okay, we're good 74 00:06:08,780 --> 00:06:13,280 There's a pro tip right there my friends better to market first 75 00:06:24,010 --> 00:06:29,980 Now we're ready to put the PCB back on and we can go test it, right? All right 76 00:06:30,600 --> 00:06:36,220 Here goes nothing. It's not picking up. Do you want to have a look at it? 77 00:06:38,390 --> 00:06:43,650 Okay, so unfortunately sometimes when you hook up to a bad power source, it kills more than one thing 78 00:06:44,130 --> 00:06:46,370 So we fixed our PCB problem 79 00:06:46,690 --> 00:06:48,930 But we're still getting this zero capacity 80 00:06:49,450 --> 00:06:52,970 nonsense here and we're getting kind of like a chirping noise and 81 00:06:54,250 --> 00:06:59,630 Jarek like drive whispered with the horse whisperer for hard drives over here and 82 00:07:00,210 --> 00:07:04,870 Told me that what he thinks is that it's the pre amplifier chip. Is that right? 83 00:07:05,370 --> 00:07:11,210 So that's a component of the drive head, which means that we actually need to dig a little deeper on this thing 84 00:07:11,210 --> 00:07:13,930 We're gonna take it over to the laminar flow workstation and tear it apart 85 00:07:14,570 --> 00:07:21,150 So what this station does is it uses a really thick filter here to pull all the dies? 86 00:07:21,750 --> 00:07:27,130 and then you can take it out of the air and then you've got this constant flow of clean air so that none of the 87 00:07:27,130 --> 00:07:30,710 Contaminated air around us can get into the drive 88 00:07:30,790 --> 00:07:33,690 That's really important with how close the head is to the platter 89 00:07:34,070 --> 00:07:40,810 Even the smallest speck of dust could cause catastrophic damage to the platter as it's flying over it 90 00:07:41,210 --> 00:07:45,070 So now we're swapping the known good heads from the working drive 91 00:07:46,030 --> 00:07:48,540 into our bad drive 92 00:07:48,540 --> 00:07:52,140 And actually now that I come to think of it by known good 93 00:07:52,140 --> 00:07:57,070 I have not actually tested this drive. 94 00:07:57,070 --> 00:08:00,310 Presumably Taren checked it before he gave it to me. 95 00:08:00,310 --> 00:08:03,370 If it helps at all for the tension of this situation, 96 00:08:03,370 --> 00:08:08,650 I still don't know that this will actually work 97 00:08:08,650 --> 00:08:09,490 the way we want. 98 00:08:09,490 --> 00:08:10,990 I mean, we could have screwed something up, right? 99 00:08:10,990 --> 00:08:13,010 What's your confidence level right now? 100 00:08:13,010 --> 00:08:16,590 I would give it 80%. 101 00:08:16,590 --> 00:08:17,810 That's not bad. 102 00:08:17,810 --> 00:08:20,930 All right, let's go give it a shot. 103 00:08:20,930 --> 00:08:24,010 So this is it, moment of truth. 104 00:08:24,010 --> 00:08:27,070 I'm not superstitious, but I'm gonna cross my fingers. 105 00:08:27,070 --> 00:08:28,330 I'm not gonna knock on wood though, 106 00:08:28,330 --> 00:08:30,870 cause then we might like bump the drive. 107 00:08:30,870 --> 00:08:34,630 Hey, no, I think we're gonna get it. 108 00:08:34,630 --> 00:08:37,110 Now that's actually a really important point. 109 00:08:37,110 --> 00:08:40,510 Just because we are able to read data off of this drive 110 00:08:40,510 --> 00:08:43,570 does not mean that Taren can take it, 111 00:08:43,570 --> 00:08:46,290 pop it back in a system and start using it again. 112 00:08:46,290 --> 00:08:48,790 This is still a dead drive. 113 00:08:48,790 --> 00:08:51,630 So what we're doing right now is we're pulling 114 00:08:51,630 --> 00:08:53,890 a bit for bit copy of all of the data 115 00:08:53,890 --> 00:08:55,930 off and we're putting it onto a new one. 116 00:08:55,930 --> 00:08:59,710 We are imaging the drive at 97 megabytes a second. 117 00:08:59,710 --> 00:09:01,450 Fingers crossed. 118 00:09:01,450 --> 00:09:03,070 No, no, no, sorry. 119 00:09:03,070 --> 00:09:04,730 Sorry, okay, so what, I have to do this 120 00:09:04,730 --> 00:09:05,930 for another six hours? 121 00:09:05,930 --> 00:09:09,240 Six hours. 122 00:09:09,240 --> 00:09:12,680 Scenario number three is a firmware problem. 123 00:09:12,680 --> 00:09:16,080 Now for unusual issues, deep knowledge 124 00:09:16,080 --> 00:09:19,680 of hard drive firmware design and a professional tool 125 00:09:19,680 --> 00:09:22,960 like a PC 3000 is necessary. 126 00:09:22,960 --> 00:09:24,840 But the fees that you can expect to pay 127 00:09:24,840 --> 00:09:26,180 to someone who can interpret 128 00:09:26,180 --> 00:09:28,660 this matrix level stuff right here 129 00:09:28,660 --> 00:09:33,660 would be anywhere from $500 to easily thousands of dollars 130 00:09:33,760 --> 00:09:35,620 depending on the complexity of the problem. 131 00:09:35,620 --> 00:09:39,100 The good news is that many common firmware issues 132 00:09:39,100 --> 00:09:42,320 can actually be solved with pretty much a button press. 133 00:09:42,320 --> 00:09:47,320 And that is where DeepSpar's tool, the RapidSpar, comes in. 134 00:09:58,390 --> 00:10:01,200 Wow, this is taking a really long time to detect. 135 00:10:01,200 --> 00:10:03,560 Drive still works. 136 00:10:03,560 --> 00:10:05,220 Oh, here we go, well, hold on a second. 137 00:10:05,220 --> 00:10:06,060 It's really, those are just- Yeah. 138 00:10:06,060 --> 00:10:07,020 We've got three partitions. 139 00:10:07,020 --> 00:10:09,160 We've got three partitions off of it already, though. 140 00:10:09,160 --> 00:10:10,000 And then we're just waiting on- 141 00:10:10,000 --> 00:10:11,880 The partition is still loading. 142 00:10:11,880 --> 00:10:15,050 What a frustrating and tedious line of work. 143 00:10:15,050 --> 00:10:18,270 But you get to deliver people their data back. 144 00:10:18,270 --> 00:10:19,880 It's our customers. 145 00:10:19,880 --> 00:10:21,660 Right, but someone gets their data back. 146 00:10:21,660 --> 00:10:22,500 Someone. 147 00:10:22,500 --> 00:10:23,520 So it only took five minutes. 148 00:10:23,520 --> 00:10:24,340 There we go. 149 00:10:24,340 --> 00:10:26,300 Okay, so now then we're good. 150 00:10:26,300 --> 00:10:28,600 It's just that every time we have to plug in this drive, 151 00:10:28,600 --> 00:10:30,400 it takes a long time to detect. 152 00:10:30,400 --> 00:10:31,240 Oh, sure, yeah. 153 00:10:31,240 --> 00:10:33,640 Let's see what happens when we try to copy files. 154 00:10:33,640 --> 00:10:36,240 So wait, so everything is slow like this. 155 00:10:36,240 --> 00:10:37,080 You're right. 156 00:10:37,280 --> 00:10:40,060 Funny story, I was actually here two weeks ago, 157 00:10:40,060 --> 00:10:44,000 but we lost part of the data from that shoot. 158 00:10:44,000 --> 00:10:45,880 Last time we were able to show 159 00:10:45,880 --> 00:10:48,000 that the data was transferring at such a speed 160 00:10:48,000 --> 00:10:50,920 that it would take infinitely long to get it off, 161 00:10:50,920 --> 00:10:53,140 meaning that the drive would be guaranteed to be dead 162 00:10:53,140 --> 00:10:55,660 by the time we actually managed to recover anything from it. 163 00:10:55,660 --> 00:10:57,620 But it's going even slower, why? 164 00:10:57,620 --> 00:10:58,820 It's just getting bogged down, 165 00:10:58,820 --> 00:11:00,940 doing something that it's not supposed to be doing. 166 00:11:00,940 --> 00:11:02,860 So instead of giving us the data, 167 00:11:02,860 --> 00:11:04,720 it's just kind of running in circles. 168 00:11:04,720 --> 00:11:07,380 Eventually it's just not gonna identify whatsoever. 169 00:11:07,380 --> 00:11:08,680 But we'll still be able to fix it, 170 00:11:08,680 --> 00:11:10,060 because it's only a firmware problem. 171 00:11:10,060 --> 00:11:13,140 So we are going to stop this process, 172 00:11:13,140 --> 00:11:15,980 and we're gonna connect our drive. 173 00:11:15,980 --> 00:11:19,160 And this is completely custom hardware and software, 174 00:11:19,160 --> 00:11:20,000 is that right? 175 00:11:20,000 --> 00:11:22,260 Correct, we rebuilt everything from this from scratch. 176 00:11:22,260 --> 00:11:23,760 It's made in Canada as well. 177 00:11:24,980 --> 00:11:27,480 We coded everything from scratch in assembly code. 178 00:11:27,480 --> 00:11:30,360 The reason is that off-the-shelf operating systems 179 00:11:30,360 --> 00:11:32,880 make certain assumptions about the hardware, 180 00:11:32,880 --> 00:11:34,880 namely that it works. 181 00:11:34,880 --> 00:11:36,860 So as soon as the hardware doesn't work, 182 00:11:36,860 --> 00:11:38,180 it starts running into issues. 183 00:11:38,180 --> 00:11:41,400 And that's why any software product 184 00:11:41,400 --> 00:11:45,500 is going to get stymied by certain drive malfunctions. 185 00:11:45,500 --> 00:11:48,120 Whereas when you build your own thing from the ground up 186 00:11:48,120 --> 00:11:50,860 that expects problems, 187 00:11:50,860 --> 00:11:52,580 well, you can have a different experience. 188 00:11:52,580 --> 00:11:55,140 So what Serge is doing right now in their software 189 00:11:55,140 --> 00:11:57,940 is he's running a diagnostic test on the drive 190 00:11:57,940 --> 00:12:01,060 to determine what exactly the firmware problem is. 191 00:12:01,060 --> 00:12:03,240 So this process takes a couple of minutes, 192 00:12:03,240 --> 00:12:04,840 and what it's doing is it's going through 193 00:12:04,840 --> 00:12:06,820 and it's finding any discrepancies 194 00:12:06,820 --> 00:12:08,680 in the firmware of this drive 195 00:12:08,680 --> 00:12:10,700 compared to what's supposed to be there, 196 00:12:10,700 --> 00:12:12,200 and then it's overwriting them. 197 00:12:12,200 --> 00:12:14,740 And that doesn't make it so the drive loses any data. 198 00:12:14,740 --> 00:12:15,580 No. 199 00:12:15,580 --> 00:12:16,420 No, okay. 200 00:12:16,420 --> 00:12:19,160 So now all that's left is power down the drive. 201 00:12:21,800 --> 00:12:23,790 Touch interface. 202 00:12:23,790 --> 00:12:24,750 Nice touch. 203 00:12:26,080 --> 00:12:30,280 Then, theoretically, we pop this guy 204 00:12:30,280 --> 00:12:32,590 back into our drive dock here. 205 00:12:32,590 --> 00:12:37,040 And now, oh, I thought I heard. 206 00:12:37,040 --> 00:12:38,680 Sorry, okay. 207 00:12:38,940 --> 00:12:40,960 We're on the drive. 208 00:12:40,960 --> 00:12:42,440 Hey, there it is. 209 00:12:42,440 --> 00:12:44,720 Oh, that's way faster. 210 00:12:44,720 --> 00:12:49,740 Of course, though, the acid test is a file copy. 211 00:12:49,740 --> 00:12:52,480 Can we get data off of this drive now? 212 00:12:52,480 --> 00:12:55,760 Now that discovery process is going faster. 213 00:12:55,760 --> 00:12:57,300 Hey, there we go. 214 00:12:57,300 --> 00:12:58,940 This is much better. 215 00:12:58,940 --> 00:13:00,420 Awesome. 216 00:13:00,420 --> 00:13:01,260 Here we go. 217 00:13:01,260 --> 00:13:02,240 There we go. 218 00:13:02,240 --> 00:13:03,540 Fantastic. 219 00:13:03,540 --> 00:13:04,920 So with the rapid spar, 220 00:13:04,920 --> 00:13:08,080 we went from not being able to copy the data 221 00:13:08,080 --> 00:13:09,140 at any kind of reasonable speed, 222 00:13:09,140 --> 00:13:12,440 to now we are copying all the data off this drive. 223 00:13:12,440 --> 00:13:17,440 Now, one of these guys does cost about $2,000. 224 00:13:17,700 --> 00:13:20,300 So end users aren't expected to buy one, 225 00:13:20,300 --> 00:13:23,040 but since it's so simple to use, 226 00:13:23,040 --> 00:13:26,720 if you, the end user, can find a shop that has one, 227 00:13:26,720 --> 00:13:29,440 you should expect to pay about 300 bucks 228 00:13:29,440 --> 00:13:31,480 for a recovery that it can handle. 229 00:13:31,480 --> 00:13:33,420 So, I mean, maybe that's still not worth it 230 00:13:33,420 --> 00:13:36,020 for photos of that night of binge drinking or whatever, 231 00:13:36,020 --> 00:13:38,920 but if your tax records and your will 232 00:13:38,920 --> 00:13:42,320 are on your hard drive, it's worth considering. 233 00:13:42,320 --> 00:13:43,820 In scenario number four, 234 00:13:43,820 --> 00:13:45,340 we're gonna see another solution 235 00:13:45,340 --> 00:13:47,180 that you can actually try at home, 236 00:13:47,180 --> 00:13:50,220 and this time without any specialized tools. 237 00:13:50,220 --> 00:13:55,220 All we need is a USB drive dock, our drive, and a computer. 238 00:13:55,300 --> 00:13:58,760 So this is an example of logical corruption. 239 00:13:58,760 --> 00:14:01,140 We're gonna go ahead and plug in our drive and power it on, 240 00:14:01,140 --> 00:14:02,180 and you're gonna see something 241 00:14:02,180 --> 00:14:03,420 that you might've seen before. 242 00:14:03,420 --> 00:14:06,820 So this can be caused by accidental formatting, 243 00:14:06,820 --> 00:14:09,060 viruses, or Windows errors. 244 00:14:09,080 --> 00:14:09,920 See this? 245 00:14:09,920 --> 00:14:11,400 You need to format the disk in drive D, 246 00:14:11,400 --> 00:14:13,200 but hold on a second, I had data on this. 247 00:14:13,200 --> 00:14:17,780 Okay, step number one, don't click format, cancel. 248 00:14:17,780 --> 00:14:20,420 Then we're gonna use a software called RStudio, 249 00:14:20,420 --> 00:14:22,280 but there are actually free options out there 250 00:14:22,280 --> 00:14:23,940 that are somewhat similar. 251 00:14:23,940 --> 00:14:26,300 We're gonna go ahead and we're gonna click on our drive here. 252 00:14:26,300 --> 00:14:28,560 We're gonna double-click empty space, 253 00:14:28,560 --> 00:14:32,260 and if it's not actually a hardware issue, 254 00:14:34,100 --> 00:14:37,020 so no bad sectors, and look at this. 255 00:14:37,020 --> 00:14:38,420 Here's all our files. 256 00:14:38,420 --> 00:14:41,180 They are ready for recovery. 257 00:14:41,180 --> 00:14:42,400 Fantastic. 258 00:14:42,400 --> 00:14:44,780 This type of software can also help us 259 00:14:44,780 --> 00:14:47,540 diagnose scenario number five. 260 00:14:47,540 --> 00:14:51,000 So the thing is is that unless you drop it or something, 261 00:14:51,000 --> 00:14:54,560 a hard drive doesn't generally go from working perfectly 262 00:14:54,560 --> 00:14:57,710 to completely dead in an instant. 263 00:14:57,710 --> 00:15:01,330 Read instability almost always comes first, 264 00:15:01,330 --> 00:15:05,070 and these types of drives, these mostly dead drives 265 00:15:05,070 --> 00:15:08,010 that haven't fully failed yet make up the majority 266 00:15:08,450 --> 00:15:11,210 of the cases that get sent in for data recovery. 267 00:15:11,210 --> 00:15:13,450 The symptoms of these drives can include 268 00:15:13,450 --> 00:15:15,630 not being detected in the BIOS, 269 00:15:15,630 --> 00:15:17,710 preventing the computer from booting, 270 00:15:17,710 --> 00:15:19,130 spitting out random errors, 271 00:15:19,130 --> 00:15:21,250 or just getting so slow 272 00:15:21,250 --> 00:15:23,430 that you can't do anything with them anymore. 273 00:15:23,430 --> 00:15:27,790 You shouldn't have any noticeable knocking or clicking yet, 274 00:15:27,790 --> 00:15:29,330 and that's exactly the situation. 275 00:15:29,330 --> 00:15:32,860 So we're gonna fire up RStudio here, 276 00:15:32,860 --> 00:15:34,950 and oh, that's interesting. 277 00:15:34,950 --> 00:15:38,400 Our computer is completely frozen up. 278 00:15:38,400 --> 00:15:39,900 One of these partitions has shown up, 279 00:15:39,900 --> 00:15:42,430 but this one's just not populated. 280 00:15:42,430 --> 00:15:45,930 Here's a little Windows hack that you might be able to use 281 00:15:45,930 --> 00:15:48,430 to get access to a drive like this. 282 00:15:48,430 --> 00:15:51,270 All you need to do is start up the command prompt 283 00:15:51,270 --> 00:15:52,670 as an administrator, 284 00:15:52,670 --> 00:15:55,670 start up Windows' built-in partition tool. 285 00:15:55,670 --> 00:15:58,310 Oh, I gotta, right, the computer's completely locked up. 286 00:15:58,310 --> 00:15:59,810 I gotta power down this drive first. 287 00:15:59,810 --> 00:16:00,650 There we go. 288 00:16:00,650 --> 00:16:02,310 Now everything, see, now everything comes back. 289 00:16:02,310 --> 00:16:04,030 Then we just need to run a command prompt. 290 00:16:04,030 --> 00:16:05,230 Oh, there we go. 291 00:16:05,230 --> 00:16:06,630 We've run a couple of commands, 292 00:16:06,630 --> 00:16:09,490 auto-mount disable and auto-mount scrub. 293 00:16:09,490 --> 00:16:11,750 With our tweak, what's gonna happen here 294 00:16:11,750 --> 00:16:15,190 is that Windows isn't gonna try to load a drive letter, 295 00:16:15,190 --> 00:16:16,990 and that's actually a good thing 296 00:16:16,990 --> 00:16:20,210 because it can hang the entire operating system 297 00:16:20,210 --> 00:16:22,490 while Windows tries to mount the file system, 298 00:16:22,490 --> 00:16:24,310 and it'll try and then time out, 299 00:16:24,310 --> 00:16:26,730 and then it'll restart and do it forever, 300 00:16:26,730 --> 00:16:29,850 doing more damage to what is already a damaged drive 301 00:16:29,850 --> 00:16:30,730 in the first place. 302 00:16:30,730 --> 00:16:32,730 So now that we've done this, 303 00:16:32,730 --> 00:16:35,230 we can open up Disk Management, and boom! 304 00:16:35,230 --> 00:16:36,390 There it is. 305 00:16:36,390 --> 00:16:38,330 There's our 300-gig drive, 306 00:16:38,330 --> 00:16:41,550 and we can actually launch our RStudio software now 307 00:16:41,550 --> 00:16:43,850 because the system isn't completely locked up. 308 00:16:43,850 --> 00:16:48,390 In cases where the drive doesn't have a ton of bad sectors, 309 00:16:48,390 --> 00:16:51,910 we actually can attempt a recovery. 310 00:16:51,910 --> 00:16:53,830 That's looking pretty... 311 00:16:53,830 --> 00:16:58,400 Oh, this is taking longer than it should. 312 00:16:58,400 --> 00:17:01,520 So this is the point where, as a data recovery technician, 313 00:17:01,520 --> 00:17:05,160 your gut feeling should be to stop what you're doing, 314 00:17:05,160 --> 00:17:07,720 because we're only 17% of the way through, 315 00:17:07,720 --> 00:17:10,700 and this drive has many bad sectors, 316 00:17:10,700 --> 00:17:15,680 which are taking anywhere from one to three minutes 317 00:17:15,680 --> 00:17:17,940 to detect each time. 318 00:17:17,940 --> 00:17:19,120 Now, the problem here is that 319 00:17:19,120 --> 00:17:21,560 this is a typical software behavior 320 00:17:21,560 --> 00:17:23,640 where the drive encounters a read error, 321 00:17:23,640 --> 00:17:26,560 and it just keeps trying and keeps trying and keeps trying. 322 00:17:26,560 --> 00:17:29,160 This puts a lot of strain on the drive 323 00:17:29,160 --> 00:17:32,760 and can cause it to fail right in the middle 324 00:17:32,760 --> 00:17:34,860 of this type of recovery attempt. 325 00:17:35,160 --> 00:17:37,540 We are gonna switch over to the RAPIDS bar 326 00:17:37,540 --> 00:17:40,260 and talk about that method. 327 00:17:40,260 --> 00:17:43,340 So all we're gonna do is plug our drive in here, 328 00:17:43,340 --> 00:17:44,720 go ahead and power it up, 329 00:17:44,720 --> 00:17:46,780 and then there are a couple of things 330 00:17:46,780 --> 00:17:50,620 that might make us think that this, the RAPIDS bar, 331 00:17:50,620 --> 00:17:52,960 is gonna work better than a standard computer. 332 00:17:52,960 --> 00:17:56,360 For one thing, if it encounters a bad sector, 333 00:17:56,360 --> 00:17:58,580 instead of reattempting and reattempting, 334 00:17:58,580 --> 00:17:59,920 wearing out the drive, 335 00:17:59,920 --> 00:18:02,660 it will actually cut off its reattempts 336 00:18:02,660 --> 00:18:04,380 after a couple of hundred milliseconds. 337 00:18:04,380 --> 00:18:05,340 This dramatically reduces the rate of error. 338 00:18:05,340 --> 00:18:07,440 This dramatically reduces the stress on the drive. 339 00:18:07,440 --> 00:18:11,580 For another, it actually borrows a lot of DNA 340 00:18:11,580 --> 00:18:13,040 from their higher end products. 341 00:18:13,040 --> 00:18:15,480 So this is the disk imager. 342 00:18:15,480 --> 00:18:18,580 So when we set our source drive parameters here, 343 00:18:18,580 --> 00:18:21,740 we select the brand and the interface, 344 00:18:21,740 --> 00:18:25,720 and what it's doing is it's actually taking information 345 00:18:25,720 --> 00:18:28,680 about the behavior of our drive as it scans it, 346 00:18:28,680 --> 00:18:31,100 and it's uploading it to DeepSpar servers 347 00:18:31,100 --> 00:18:33,440 where it compares it against a database. 348 00:18:33,440 --> 00:18:35,940 Their server then sends our RAPIDS bar 349 00:18:35,940 --> 00:18:38,700 some recommendations for how to deal with it 350 00:18:38,700 --> 00:18:40,440 to help accelerate the recovery. 351 00:18:40,440 --> 00:18:41,780 Pretty cool, right? 352 00:18:41,780 --> 00:18:44,400 So we can double click partition one here, 353 00:18:44,400 --> 00:18:46,860 and boom, we're running a scan. 354 00:18:46,860 --> 00:18:47,820 So you can see here, 355 00:18:47,820 --> 00:18:49,860 we already hit one of those bad sectors, 356 00:18:49,860 --> 00:18:52,500 but instead of trying for two minutes, 357 00:18:52,500 --> 00:18:54,600 it's only gonna try for a couple of seconds 358 00:18:54,600 --> 00:18:55,700 before it moves on. 359 00:18:55,700 --> 00:18:57,440 Now, we're never gonna get back any data 360 00:18:57,440 --> 00:18:58,640 that was on those sectors, 361 00:18:58,640 --> 00:19:01,920 but at least we can get back whatever's on either side. 362 00:19:01,920 --> 00:19:02,780 So check this out. 363 00:19:02,780 --> 00:19:04,260 We didn't get everything. 364 00:19:04,260 --> 00:19:06,440 You can see there's some corrupted stuff here, 365 00:19:06,440 --> 00:19:11,500 but even though this drive had hundreds of bad sectors, 366 00:19:12,380 --> 00:19:15,260 instead of taking two minutes 367 00:19:15,260 --> 00:19:17,320 every time it hit a bad sector, 368 00:19:17,320 --> 00:19:20,960 this whole scan only took us about seven minutes, 369 00:19:20,960 --> 00:19:22,460 much less wear and tear. 370 00:19:22,460 --> 00:19:24,320 And there's some other cool stuff too. 371 00:19:24,320 --> 00:19:25,400 For this trick, 372 00:19:25,400 --> 00:19:28,340 we need a drive that is in much worse condition. 373 00:19:28,340 --> 00:19:32,440 Let's go ahead and power this on. 374 00:19:32,440 --> 00:19:33,780 That's the hard drive. 375 00:19:33,780 --> 00:19:34,860 That's less fine. 376 00:19:34,980 --> 00:19:36,380 That's not fine at all. 377 00:19:36,380 --> 00:19:40,180 Ah, ah, turn it off. 378 00:19:40,180 --> 00:19:44,000 But the RapidSpar has another borrowed trick up its sleeve. 379 00:19:44,000 --> 00:19:46,140 So all we gotta do is plug this in, 380 00:19:46,140 --> 00:19:48,360 but we're not gonna power the drive on right away 381 00:19:48,360 --> 00:19:50,600 because we don't want it to start thrashing. 382 00:19:50,600 --> 00:19:53,660 Instead, we're gonna let the software interface 383 00:19:53,660 --> 00:19:54,940 power on the drive 384 00:19:54,940 --> 00:19:58,360 and then start issuing commands to it immediately 385 00:19:58,360 --> 00:19:59,660 so that it doesn't get a chance 386 00:19:59,660 --> 00:20:04,520 to try to click itself to death here. 387 00:20:04,520 --> 00:20:05,720 So it's powering it on, 388 00:20:05,720 --> 00:20:08,120 and then hopefully we're not gonna hear a ton of clicking 389 00:20:08,120 --> 00:20:12,660 because that would mean the potential for our dead head 390 00:20:12,660 --> 00:20:14,260 to cause our other head to fail. 391 00:20:14,260 --> 00:20:15,620 That's something that's really common. 392 00:20:15,620 --> 00:20:17,580 Once you've got one head gone, 393 00:20:17,580 --> 00:20:20,270 the other ones tend to follow suit. 394 00:20:20,270 --> 00:20:21,290 Smart failed. 395 00:20:21,290 --> 00:20:24,210 Okay, that makes sense. 396 00:20:24,210 --> 00:20:25,110 Now in a moment here, 397 00:20:25,110 --> 00:20:27,050 it's gonna start checking the heads and media. 398 00:20:27,050 --> 00:20:29,070 We're probably gonna hear a couple of clicks here 399 00:20:29,070 --> 00:20:30,830 because you can't check if the head works 400 00:20:30,830 --> 00:20:35,480 without trying to move it. 401 00:20:35,480 --> 00:20:36,600 Diagnostics are done now. 402 00:20:36,600 --> 00:20:38,120 So 19% of our tested sectors do contain data. 403 00:20:38,120 --> 00:20:39,060 So 19% of our tested sectors do contain data. 404 00:20:39,060 --> 00:20:41,260 There's something to recover here. 405 00:20:41,260 --> 00:20:42,680 As we expected, 406 00:20:42,680 --> 00:20:46,620 one of our three heads is at 46% health. 407 00:20:46,620 --> 00:20:48,540 Now we're not getting back anything 408 00:20:48,540 --> 00:20:50,380 that's on that platter with the bad head, 409 00:20:50,380 --> 00:20:53,100 unless we do a head swap and put a new head in the drive. 410 00:20:53,100 --> 00:20:55,020 So if this stuff isn't mission critical, 411 00:20:55,020 --> 00:20:57,560 what we're gonna do is we're gonna build a head map here. 412 00:20:57,560 --> 00:20:59,640 Then we're gonna disable that head 413 00:20:59,640 --> 00:21:01,800 and see how much of the data we can pull off. 414 00:21:01,800 --> 00:21:03,760 All we gotta do is click this, 415 00:21:03,760 --> 00:21:05,980 skip anything on head one, 416 00:21:05,980 --> 00:21:07,060 click apply, 417 00:21:07,060 --> 00:21:09,060 and now we're getting a new sector map. 418 00:21:09,060 --> 00:21:10,060 All right, heads. 419 00:21:10,060 --> 00:21:11,600 Here comes our files and directories, 420 00:21:11,600 --> 00:21:13,380 and boom, there it is. 421 00:21:13,380 --> 00:21:14,660 File tree, baby. 422 00:21:14,660 --> 00:21:15,500 Love it. 423 00:21:15,500 --> 00:21:17,100 And through all this accessing, 424 00:21:17,100 --> 00:21:19,060 you don't hear any clicks anymore 425 00:21:19,060 --> 00:21:21,480 because that head has been parked 426 00:21:21,480 --> 00:21:24,100 as long as we're using the RapidSpar. 427 00:21:24,100 --> 00:21:26,040 So this thing is really cool, 428 00:21:26,040 --> 00:21:29,900 but to be clear, the RapidSpar is also not magic. 429 00:21:29,900 --> 00:21:33,000 So of the jobs that come into your shop, 430 00:21:33,000 --> 00:21:35,620 some of them are gonna be solvable with software. 431 00:21:35,620 --> 00:21:37,620 And then of the ones that remain, 432 00:21:37,620 --> 00:21:40,660 about half of them can be dealt with with the RapidSpar. 433 00:21:40,660 --> 00:21:41,900 And then the other half 434 00:21:41,900 --> 00:21:44,460 are still gonna have to be outsourced to a professional 435 00:21:44,460 --> 00:21:46,520 that's armed with more powerful tools 436 00:21:46,520 --> 00:21:48,080 and probably a clean room. 437 00:21:48,080 --> 00:21:49,800 So DiskImager 4, 438 00:21:49,800 --> 00:21:51,980 along with the software that accompanies it, 439 00:21:51,980 --> 00:21:53,700 allows very fine control 440 00:21:53,700 --> 00:21:55,680 of things like the reattempt threshold. 441 00:21:55,680 --> 00:21:57,620 It has a ton of configuration knobs 442 00:21:57,620 --> 00:21:59,100 that a professional can tune 443 00:21:59,100 --> 00:22:00,820 and can be used to work on drives 444 00:22:00,820 --> 00:22:03,580 that the RapidSpar can't even interface with yet. 445 00:22:03,580 --> 00:22:05,620 Like for example, a PCI Express NVMe SSD. 446 00:22:05,620 --> 00:22:06,620 The RapidSpar is a very powerful tool. 447 00:22:06,620 --> 00:22:07,620 It's a very powerful tool. 448 00:22:07,620 --> 00:22:09,740 The RapidSpar can handle a regular SATA SSD, 449 00:22:09,740 --> 00:22:13,120 which by the way, can have bad sectors as well. 450 00:22:13,120 --> 00:22:14,560 So that's pretty much it. 451 00:22:14,560 --> 00:22:17,320 So hopefully whether you're an IT manager 452 00:22:17,320 --> 00:22:18,740 or you run your own shop, 453 00:22:18,740 --> 00:22:20,140 or you're just an end user, 454 00:22:20,140 --> 00:22:21,200 you've gained some knowledge 455 00:22:21,200 --> 00:22:23,000 about the data recovery business 456 00:22:23,000 --> 00:22:25,060 and how it works behind the scenes. 457 00:22:25,060 --> 00:22:27,080 Now, of course, DeepSpar, 458 00:22:27,080 --> 00:22:28,460 just wanna throw this in, 459 00:22:28,460 --> 00:22:30,160 DeepSpar values your business, 460 00:22:30,160 --> 00:22:32,320 but any video about data recovery 461 00:22:32,320 --> 00:22:34,740 would be incomplete without a reminder 462 00:22:34,740 --> 00:22:36,940 that in a perfect world, you wouldn't need it. 463 00:22:36,940 --> 00:22:38,780 Please back up your data 464 00:22:38,780 --> 00:22:41,440 because the DeepSpar guys would love nothing more 465 00:22:41,440 --> 00:22:44,000 than to live in a world where no one loses their data 466 00:22:44,000 --> 00:22:46,860 and they can just sit on a beach somewhere drinking cocktails 467 00:22:46,860 --> 00:22:49,770 instead of hanging around with me all day. 468 00:22:49,770 --> 00:22:50,770 So thanks for watching, guys. 469 00:22:50,770 --> 00:22:52,830 If you disliked this video, you can hit that button. 470 00:22:52,830 --> 00:22:55,790 But if you liked it, hit like, get subscribed, 471 00:22:55,790 --> 00:22:57,790 and maybe consider checking out where to buy the stuff 472 00:22:57,790 --> 00:23:00,310 that we featured in the video description. 473 00:23:00,310 --> 00:23:05,190 This would be a great addition to any tech shop out there. 474 00:23:05,190 --> 00:23:06,030 While you guys are at it, 475 00:23:06,030 --> 00:23:06,910 we've got our merch store, 476 00:23:06,910 --> 00:23:08,570 a list in the video description as well, 477 00:23:08,570 --> 00:23:10,270 and a link to our community forum, 478 00:23:10,270 --> 00:23:12,740 which you should totally join. 479 00:23:12,740 --> 00:23:15,020 You look very satisfied right now. 480 00:23:15,020 --> 00:23:16,700 It's like, yes, this drive is broken 481 00:23:16,700 --> 00:23:19,420 exactly the way I wanted it to be for this demo.