Well, that's a kick in the pants ...

[Little David]

Hey, gang.

So, I haven't been posting here much. Right now, I'm on my netbook, and probably will be for a while yet. My main rig's currently unusable due to some unfortunate circumstances. I've got a question for the tech-savvy about that at the end of my post, but first, I need to vent.

I mentioned it before in a previous thread days after CoH ended for good, but out of the three months we had left I only got maybe six days out of it. I hadn't been able to play games since June due to my video card acting blooey (BSoDs and artifacts). Shortly after Thanksgiving I fixed the issue, but my OS drive (where most of my personal and project files are) promptly died on me, all data permanently gone as far as I know.

Thankfully I had spent the time just prior to that backing everything I cared about on that drive to a series of blu-ray discs, but I still had to spend several days installing my OS on a new drive, reinstalling all programs, drivers, and then copying those files back over to the new drive, which left me less time to play than I had hoped.

CoH and most of my other games were installed on a secondary drive, though. One that was also relatively new, bought to replace an older drive that started to fail earlier this year.

Well, just recently, that secondary drive also died on me. That would be dead drive #3 for 2012.

It happened rather suddenly, and SMART data didn't report any pending problems ... but the buzz-like series of clicks I'd heard was alarm enough to try and back up what I could. Not the usual "Click of Death" I've heard about. The second time it happened, the drive disconnected with a Delayed Write Failure, and that was it. I've tried hooking the drive up externally to another machine via USB adapter; at first, in Windows, it reported that the drive was unformatted. After that, I hooked it up while the machine was running a boot disc with partition and file recovery programs on it. The programs wouldn't launch if the drive was plugged in, and if plugged in after the programs were launched and refreshed, couldn't see the drive at all.

Backing up files on that secondary drive was a lesser concern for me, although I do have a USB external that does have a copy of the CoH folder as it was just prior to the final six days, as well as copies of other game folders I'd recently played. Combined with my previous secondary drive (if I can still access it) I might be able to reconstitute most of the drive data I've lost ...

But it still means I've probably permanently lost all the screenshots, logs, and demo recordings I took during the final six days of City of Heroes.

I want to feel like it's a cruel joke, but ... I feel horribly selfish thinking that in light of recent events. That and if it weren't for my paranoia over losing data, by now I wouldn't have anything left after three hard drive failures.

I don't know of anything else I can do to try and recover data from the drive (for example, I'm a bit wary of using the freezer trick on this drive since I don't know for sure how it failed), but I'm going to hang on to it in case an option presents itself at a later date. If anyone knows what may have caused the failure and if it's still possible to retrieve data from the drive, let me know.

I've never had drive failures as bad as I have this year. Prior to this I only experienced two drive failures in the past decade or so. The only thing I can think of that's responsible is waste heat from the video card that wasn't properly carried out of the case. The card sat below the installed drives, and card's fans pointed waste heat down at the bottom of the case, which has no vents or air to let the heat out. The secondary drive was pretty warm when I removed it (but not uncomfortably so).

The case was a mid tower, and had poor airflow all around; there were three case fans, one in front, one on top, and one on the side ... the top fan was blocked by cables and the front fan hasn't been working for a while (nor has the front USB hub). When I was taking my drives out of the case I'd noticed that a cable leading to the front panel was loose and frayed, so I think I may have found part of the reason ... and reason enough to get a new case (which I'm working on).

Venting out of the way now ... For any tech-savvy people here, which brands and models of hard drive have you had the best luck with? I want to get a replacement for my dead secondary, especially since I don't fully trust this Western Digital Caviar Green I'm using as a replacement for the dead OS drive.

[Victoria Victrix]

There may be a way to get that data off those drives.  Bear with me here.  There are a LOT of people who have had tremendously good luck in taking the drive, putting it in the freezer for about 4-8 hours, taking it out and trying to retrieve data from it.  I know, it sounds crazy, but it is worth a try.  I recovered everything off a dead laptop driver that way (google it, there are a lot of folks who will say the same.)

Take the drive, seal it in a plastic bag with a desiccant packet, and leave it there no less than 4 hours.  Overnight is fine.  Take it out and try your recovery.  When it stops working, put it back in the freezer.  Lather, rinse, repeat until it doesn't work anymore.

[The Fifth Horseman]

Don't try the freezer trick UNTIL you've exhausted all options. It's a last ditch solution with a high likelihood of destroying the hardware.

There is still one program you might want to consider trying: DMDE. (http://softdm.com/) If the drive is recognized but claimed as unformatted, the partition table - or even only the boot sector - may have been damaged. Had that a couple years ago with the boot sector on one of my drives.

There's another hardware solution (or rather a few) that can be tried, but I'm short on time so will explain later.

[Profit]

Yeah, most efforts at this point will be last ditch. Freezer does work, so does dropping it from standing height onto carpet sometimes.

I would try some data recovery software, i have easus data recovery pro and it works extremely well. Love it. I would give that a try.

Also, keep in mind, once you do the freezing or dropping tricks, sending your drive to a data recovery specialist is out of the question. So if that option is even on the table, do it first.

Also also, I do this for a living, and yes, I scoffed at the dropping trick as well, until I watched a guy do it three times on three different drives and it worked. I trust this guy with repair and recovery, he's brilliant, so I can confirm that dropping does work at least sometimes.

[corvus1970]

Little David, I am very sorry to hear all of that. I went through something similar myself back in May, when my PC crashed, and crashed hard.

My main HD became inaccessible: the platters still spin when power is applied to them, but nothing I tried will recognize the drive's presence at all. I tried accessing it on a Windows machine, and also on a Linux-box, with no joy. Trying recovery software did nothing. I even used SATA to USB adapters and so forth, anything I felt I could try to get the drive to be recognized. Zilch.

I've come to the conclusion that I will end up having to send that drive to a recovery specialist. I wish you luck!

[The Fifth Horseman]

Okay, since I'm back and have the time.

The two alternate hardware solutions I've mentioned:

1. Replacing the drive's PCB with a compatible one from same manufacturer. The donor drive does not neccessarily have to be the exact same model, as there is a certain amount of cross-compatibility, but you will need someone with very fine soldering skills (and gear) to transplant the ROM chip from the original PCB to the new one. This might be a solution for Corvus' issue as well.

2. With a bit of electronics know-how and access to correct schematics, it is possible to build a service terminal and potentially use that to recover the HDD's contents. A bit hit-and-miss, since nobody can guarantee that the neccessary schematics and software will be available in the first place.

[corvus1970]

Thanks FH. I have suspected that the PCB on that drive may indeed be damaged, since the platters spin. I may look into replacing that board and see if that helps. Thanks for the suggestion!

Edit: Like a goof, I never did searches for known issues by the drive type. Well, when I did searches specific to the Barracuda 7200.11, I found out that there's a known firmware issue that can cause exactly what I am experiencing.

I'm going to research some more!

[Little David]

There is still one program you might want to consider trying: DMDE. (http://softdm.com/) If the drive is recognized but claimed as unformatted, the partition table - or even only the boot sector - may have been damaged. Had that a couple years ago with the boot sector on one of my drives.

Alright, I'll give DMDE a shot. Looks promising since it runs natively in the OS and doesn't have to be run from a boot disc or something. By the way ... hooking the drive up externally with a USB adapter shouldn't be causing a problem with a boot disc OS or recovery programs recognizing the drive, should it?

Take the drive, seal it in a plastic bag with a desiccant packet, and leave it there no less than 4 hours.  Overnight is fine.  Take it out and try your recovery.  When it stops working, put it back in the freezer.  Lather, rinse, repeat until it doesn't work anymore.

Hrm, desiccant packet ... didn't think of that! I know I still have some from the boxes the hard drives came in. If I wind up trying the freezer option I'll do that. I know one site stressed that the drives should be double-bagged with the kind specifically meant for freezers ...

Yeah, most efforts at this point will be last ditch. Freezer does work, so does dropping it from standing height onto carpet sometimes.

Judging from the noises this drive was making, what do you think happened and which would be the better last-ditch option to try first?

I'll look into the easus data recovery program, too, if I can get my hands on it and DMDE doesn't work.

My main HD became inaccessible: the platters still spin when power is applied to them, but nothing I tried will recognize the drive's presence at all. I tried accessing it on a Windows machine, and also on a Linux-box, with no joy. Trying recovery software did nothing. I even used SATA to USB adapters and so forth, anything I felt I could try to get the drive to be recognized. Zilch.

That's exactly what happened to my previous OS drive after I fixed my video card. I could feel the drive motor spinning up, but nothing would recognize the drive itself; not the motherboard, not Windows on another computer, not the boot disc OS.

That drive didn't have any telltale signs of mechanical failure, so I can't begin to guess what went wrong with that one.

[The Fifth Horseman]

Alright, I'll give DMDE a shot. Looks promising since it runs natively in the OS and doesn't have to be run from a boot disc or something. By the way ... hooking the drive up externally with a USB adapter shouldn't be causing a problem with a boot disc OS or recovery programs recognizing the drive, should it?

Bootable toolkits... might, depending on whether or not they have USB support in the first place. Recovery software most likely depends on your OS' drivers, so should not have issues.

[GuyPerfect]

I've always liked Western Digital for hard drives, but Seagate is a pretty big name too. You can order them from NewEgg.

[Codewalker]

I've always liked Western Digital for hard drives, but Seagate is a pretty big name too. You can order them from NewEgg.

I had several WDs die in quick succession somewhere around 2002-2004ish, swore them off, and have gone all Seagate since then without issue.

I've read posts from other people who have had the exact opposite experience as me.

I suspect the phase of the moon and the alignment of the stars may be involved somehow.

If something is really important, I recommend a RAID1 or RAID5* NAS, and of course making regular offline backups. RAID won't save you if the files are overwritten.

* You really need RAID6 for any disks >1TB, but that isn't very practical for a home setup

[GuyPerfect]

I suspect the phase of the moon and the alignment of the stars may be involved somehow.

I was talking with a co-worker when I was working IT for a school district about how sometimes, you go in to troubleshoot a problem, and everything looks like it should be fine. When you test it, the problem has magically resolved itself. We called the troubleshooting ritual the "dance to the server gods." Perhaps the server gods did not smile upon your hard drives in those days.

[dwturducken]

OK, this is to be taken with a grain of salt, since we all kind of only know as much about each other as we see on here, but I worked in a computer repair shop for a while when it was owned by the ISP that I used to work for. This was during the last few months of 2008. Some of the things I picked up:

Seagates were great until they bought Maxtor. Nothing changed at the former Maxtor manufacturing plants was the labels on the drives, so you had what amounted to a crap shoot as far as quality (from a batch of 8 that we had ordered from Newegg, four failed basic hardware tests). Western Digital, around the same time period, started producing higher quality drives, but that was hit or miss for the first couple of years. Now, you seem to have a color scheme to follow. I used to know what it was, but I think it's black for the best, blue for the middle range, and green for the cheapies.

The PCB trick is fickle. I've only had it work twice out of a half dozen attempts. Once was an identical model as the donor, and the other was not. I can't remember specifics of the drives, but I do know that I never had it work on a Western Digital drive. It's also important to have as clean an area as you can possibly manage. Cleanroom conditions are ideal, but I never had that available.

If the drive is still recognizable, Nucleus Kernel was the software that we had the best results with. The thing is, the drive is now to a point where you have an extremely limited number of tries before it fails completely. No one can know how many. It could be four, or it could be none. FWIW, I successfully have used each of the techniques described above, just not as often as they failed.

[SerialBeggar]

OK. I need enlightenment.  I can guess that the dropping of the hard drive technique is based off the old hit the TV until it works method.  But I can't think of why freezing the hard drive would help.  Is it a physical tweak, ie the cold making the metals shrink?  And another thing, even with desiccant packets, as soon as you open the baggies, condensation will form on it and maybe in it.  Although I don't know how well sealed a drive is.  Won't this  have a more adverse affect on the drive?

[corvus1970]

OK. I need enlightenment.  I can guess that the dropping of the hard drive technique is based off the old hit the TV until it works method. 

AKA: Percussive maintenance.

But I can't think of why freezing the hard drive would help.  Is it a physical tweak, ie the cold making the metals shrink?  And another thing, even with desiccant packets, as soon as you open the baggies, condensation will form on it and maybe in it.  Although I don't know how well sealed a drive is.  Won't this  have a more adverse affect on the drive?

Keep in mind that with both of these solutions, they are designed to get the drive working long-enough to get your data off it, and nothing more. Once a drive has failed to the point tat you have to resort to such methods, its generally a good idea to replace it, regardless of whether or not you can retrieve your data.

[GuyPerfect]

I can guess that the dropping of the hard drive technique is based off the old hit the TV until it works method.

Never drop a hard drive. Ever. The moving parts inside are calibrated with a precision smaller than a particle of smoke, and shock is a sure-fire way to ruin it for good. Drive failures are usually caused by the mechanics wearing out in the first place, resulting in the magnetic heads coming into contact with the platters, and bouncing it around will only expediate the damage. The nature of electric motors may result in some short-term improvement of functionality after concussive force, but this is akin to loosening the lid of a glass jar by hurtling it down a flight of concrete steps.

But I can't think of why freezing the hard drive would help.  Is it a physical tweak, ie the cold making the metals shrink?

That's exactly what's going on, and it's a very short-term solution. Like mentioned above, when the magnetic read heads of a hard drive cross that grain-of-smoke distance and contacts the platter, the charge is dispersed through the platter and you can't get a good read. When this first starts to happen, a freezer can cause the metal to contract to the point that it's no longer touching together. But there's a fairly small window for that.

Freezing electronics can also have an effect in the case of power transformers. Power cords and other capacitors, for a reason I'm not totally clear on, can be repaired with cold temperatures if they ever overheat. I've been able to revive multiple Nintendo 64 units this way: they'd refuse to turn on (the red light on the front would stay off), even after resuming to room temperature, but ten minutes in the freezer would make them work again. I did the same with a laptop power cord once when it was buried in the couch cushions for too long. I can't explain why it works, but I know there's some kind of actual science behind it.

And another thing, even with desiccant packets, as soon as you open the baggies, condensation will form on it and maybe in it.  Although I don't know how well sealed a drive is.  Won't this  have a more adverse affect on the drive?

I'm not totally clear on what you're asking. Hard drives aren't air-tight: they have ventilation holes to prevent overheating. A desiccant will not cause water to form on a hard drive, since their very purpose is to absorb moisture from the air before it can condense anywhere else. Having a hard drive or any form electronics spend some quality time with desiccants is pretty much always a good thing.

[FatherXmas]

I was talking with a co-worker when I was working IT for a school district about how sometimes, you go in to troubleshoot a problem, and everything looks like it should be fine. When you test it, the problem has magically resolved itself. We called the troubleshooting ritual the "dance to the server gods." Perhaps the server gods did not smile upon your hard drives in those days.

Programmers call those things Heisenbugs.

[SerialBeggar]

I'm not totally clear on what you're asking. Hard drives aren't air-tight: they have ventilation holes to prevent overheating. A desiccant will not cause water to form on a hard drive, since their very purpose is to absorb moisture from the air before it can condense anywhere else. Having a hard drive or any form electronics spend some quality time with desiccants is pretty much always a good thing.

That's what I was saying.  Desiccants can only absorb the ambient humidity in an enclosed environment, like inside the baggie.  However, once you take the hard drive out of the baggie to hook it up, it being ice cold will now cause it to sweat with condensation, on the internals in particular, which you've confirmed is not sealed off.  That's got to be bad for any moving parts.

[dwturducken]

I did have to send in a failed drive, once, for RMA, but the data on it was deemed "sensitive." I sent a different, similar drive and was told to make sure that the sensitive drive was unusable. I dropped it on a bare concrete floor. Forcefully and repeatedly.

[GuyPerfect]

However, once you take the hard drive out of the baggie to hook it up, it being ice cold will now cause it to sweat with condensation, on the internals in particular, which you've confirmed is not sealed off.

Ah, I see what you mean. Don't use the desiccants and the freezer at the same time; forcing the moisture to condense on the device will be counter-productive to using the desiccant in the first place. You can freeze the hard drive with the desiccant after it's had a chance to do its absorption thing, but allow it to resume room temperature before taking it out of the enclosure.