hard drive recovery. not recognized in bios.

[cazwell220]

Hey all...

I'm stumped on this one...

I have a client that wants me to recover their hard drive. It's an ide seagate 160gb ntfs drive. I connected via a usb adapter in windows 7 it sees the usb device.. but doesn't assign a drive letter and requests that I initialize it.

Linux (fedora 10) via usb adapter doesn't even recognize the drive.

Tried it on another computer via usb with windows xp with no luck. I then put it on the idea as the master drive with a live cd, but it is not recognized in the bios.

The drive starts up and has power... spins up with no noticeably strange clicks. It does get very warm while sitting there...

Any ideas... I could really use a suggestion from the hive mind?

Thank you in advance.

Mark
www.geekmatrix.com
cazwell220 on twitter

 

[purple_minion]

My mythtv drive died on me after a simple reboot, I'm guessing it couldn't read in it's firmware off the platters and just refused to show up in the bios. Tried everything I could from different cables, different computers, the freeze trick, etc. Even tried wacking it on the side a bit and warming it up on the off chance maybe it was stiff. Some people report older drives like the 40 wire cables instead of the 80 wire. Is it a 2.5in or 3.5in? How old? Dropped at all? The usual things. If it gets really warm, warmer then normal, maybe the bearing is going out and that is producing more friction hence not getting up to speed. Have you tried seatools and see if it prods the bios a little more? I doubt if the bios doesn't see it then how would seatools but can't hurt to give it a minute. Perhaps a broken pin? Missing/wrong jumper?

 

[cazwell220]

My mythtv drive died on me after a simple reboot, I'm guessing it couldn't read in it's firmware off the platters and just refused to show up in the bios. Tried everything I could from different cables, different computers, the freeze trick, etc. Even tried wacking it on the side a bit and warming it up on the off chance maybe it was stiff. Some people report older drives like the 40 wire cables instead of the 80 wire. Is it a 2.5in or 3.5in? How old? Dropped at all? The usual things. If it gets really warm, warmer then normal, maybe the bearing is going out and that is producing more friction hence not getting up to speed. Have you tried seatools and see if it prods the bios a little more? I doubt if the bios doesn't see it then how would seatools but can't hurt to give it a minute. Perhaps a broken pin? Missing/wrong jumper?

Thank you for the reply...

The hard drive is not that old. It's a 3.5" drive. It's about 1 to 2 years old. Hasn't been dropped as far as I know.

I don't think it's the bearing as the platters are spinning plenty fast. I can feel that "gyro" effect when I pick it up, but I would say that it is getting warmer than normal.

I did try Seatools hoping it would see it, but my hope was not met with result. I haven't tried the freezer trick, but I'm thinking that could be a decent next step worth trying.

The pins are all intact and I have followed the pin out for the jumper to the letter.

I appreciate the ideas though... If I could only get it to be recognized in the bios then everything would be downhill.

I'm thinking, what if I just initialize it in windows 7 then use GetDataBack to recover the info? Anyone ever tried that? I probably won't do that as it's too destructive... but it's a thought.

 

[seedubya]

The freezer trick, as you call it, has one function only, to free stuck heads. What you are experiencing is either a firmware problem or controller problem. You might get around this by replacing the drive controller with a controller from a known good IDENTICAL drive. The likelihood of the drive initialising successfully in one machine when the bios of another won't even recognise it is very slim.

 

[jimman65]

This is very similar to a situation I had recently. In the end I had to use Getdataback to recover the information from the faulty drive and restore a new working drive.

 

[PcTestCard.com]

Try remove all jumpers at the back of the HDD and let the system auto detect it.

If no drive letter, you can use the disk management to assign teh letter manually.

Using the above suggested data back software may get the data back.
I use RecoveryPro application to backup the data.

Hope this helps!
Bill

 

[cazwell220]

Thank you for your suggestions on this...

I figured changing the controller from another drive would be a very possible solution, but what is strange is when I plug it in when I have it hooked up via usb in windows xp and windows 7... the drive is recognized, but has no file system information. If it was the controller circuit causing the problem, could it still be recognized via usb inside the OS?

I can't run GetDataBack or any other file recovery software because it's not being recognized as having a filesystem. So windows see's it, but the software does not.

I'm still trying to come up with possible solutions... and I appreciate all the ideas here... I'm almost ready to give up and let the client know they are going to have to send it into a dedicated data recovery service.

Mark

 

[cazwell220]

Also... I tried running TestDisk... but that won't find it either. TestDisk has come through so many times in the past... But it is not able to see the disk either... which again points in the direction of the controller. eh... I'm almost out of ideas on this one.

 

[purple_minion]

I'd image if the drive is directly hooked up then the bios doesn't see it as the drive is gone. It may be a problem with the controller, or maybe it can't read it's firmware, or maybe something else is wrong. When it's hooked up to usb I think it sees the usb controller, goes yeah something is plugged in but I don't see anything it must be a new drive.

I have no idea how much you know about linux so I appologize if this somehow insults you. I just started dabbling in it 4-5 years ago so I'm no expert. Personally I wouldn't use fedora, I'd use something geared more towards just the basics, I like SystemRescueCD which doesn't have all the fancy stuff. In some of those distro's they dumb it down, where as in the more robust ones like SystemRescueCD you can force mount things or at the least see some error message. I'd start up SystemRescueCD without the drive plugged in, once you're up plug in the drive and wait a bit, then do a dmesg command to see at the end of the log what it found for the usb and if it found a hard drive. If it finds one it usually prints up stuff about the size, location, etc. Perhaps try it with a known good drive if you want something to compare output.

Otherwise you can try swapping controller boards, no idea how hard it is to find an identical model or if it would work. It may be lost depending on if they want you to send it in to a data recovery service. I've read about people that have actually opened their drives themselves and poked the arm to unstick it to recover data, without a clean room. I have no idea why it didn't crash right away unless it is one of those things you can get away with for a few hours before the SHTF. I think you may be screwed sadly, but maybe you can sell them a good backup solution!

 

[cazwell220]

I did give knoppix a try with no luck... but SystemRescueCD is a good suggestion, I will give that a try... can't hurt at this point.

It does make sense that windows finds the usb controller rather than the hard drive controller... I am curious what would happen if I tried initializing the disk, almost tempted to try, but I will resist that urge.

Not sure if I can find this same hard drive to switch out the controller board, but I think I may need to give that a shot too.

As for the backup solution... I'm definitely going to get them set up on a backup.

Thanks

 

[cazwell220]

I'd image if the drive is directly hooked up then the bios doesn't see it as the drive is gone. It may be a problem with the controller, or maybe it can't read it's firmware, or maybe something else is wrong. When it's hooked up to usb I think it sees the usb controller, goes yeah something is plugged in but I don't see anything it must be a new drive.

I have no idea how much you know about linux so I appologize if this somehow insults you. I just started dabbling in it 4-5 years ago so I'm no expert. Personally I wouldn't use fedora, I'd use something geared more towards just the basics, I like SystemRescueCD which doesn't have all the fancy stuff. In some of those distro's they dumb it down, where as in the more robust ones like SystemRescueCD you can force mount things or at the least see some error message. I'd start up SystemRescueCD without the drive plugged in, once you're up plug in the drive and wait a bit, then do a dmesg command to see at the end of the log what it found for the usb and if it found a hard drive. If it finds one it usually prints up stuff about the size, location, etc. Perhaps try it with a known good drive if you want something to compare output.

Otherwise you can try swapping controller boards, no idea how hard it is to find an identical model or if it would work. It may be lost depending on if they want you to send it in to a data recovery service. I've read about people that have actually opened their drives themselves and poked the arm to unstick it to recover data, without a clean room. I have no idea why it didn't crash right away unless it is one of those things you can get away with for a few hours before the SHTF. I think you may be screwed sadly, but maybe you can sell them a good backup solution!

I gave SystenRescueCD a try. Turns out it says:



usb-storage: device found at 4
usb-storage: waiting for device to settle before scanning



I tried a couple other drives and after it finishes scanning gives all the details about the drive, where this drive apparently does not settle.



Any ideas what to do from here? Or is this a lost cause all together?



Thank you,



Mark

 

[purple_minion]

I gave SystenRescueCD a try. Turns out it says:



usb-storage: device found at 4
usb-storage: waiting for device to settle before scanning



I tried a couple other drives and after it finishes scanning gives all the details about the drive, where this drive apparently does not settle.



Any ideas what to do from here? Or is this a lost cause all together?



Thank you,



Mark

Well at least you've got something to google now. Googling the last line there gave me these.

http://ubuntuforums.org/archive/index.php/t-47805.html

Seems a lot of people have a problem with missing partitions and flashdrives that list this message.

Perhaps try a lsusb command (or lsusb -v for more info) and see what happens. How long did you wait for it to settle? Maybe grab a drink or something and give it a few minutes, or maybe linux just gives up. I hope this is an externally powered drive and not powered from the usb cable. You could try "modprobe -r ehci_hcd" before you plug it in, which will remove the usb 2.0 module and force everything to be usb 1, which is slower but it worked for some crappy file transfers that didn't work in winblows. Granted I could see the drive then so it might not work for you.

Again, I had the same issue, would not show up in bios (I don't have a usb enclosure so never tried that, maybe the bios didn't wait long enough, who knows.) It was a drive that was in constant use 24/7 always recording or playing back something for a few years, I believe HD's are rated for something like 8 hours a day for the regular non-corperate drives so I assume I shorted it's life by a factor of 3. So from 3-5 years to 1-2 years? Oh, also if your drives are warm/hot they die a lot quicker, I now have a front fan that blows on them just for this reason. Google did a study and just a wee bit warmer killed them faster, here mine were WARM so no doubt they died.

Saddly in my limited opinion, if the drive never responds or even communicates then you're screwed, as how can you do anything if it won't talk??? Try a controller switch-a-roo, or data recovery service... if it's important enough to the client for several thousand dollars. When you set them up again perhaps get an enterprise drive, they are rated for double the MTBF I think, usually not too much more expensive but more reliable. Also go with good airflow and a fan if you can, as a few of my towers have a spot for a fan in front, or even rig up something with a spare fan and zipties or whatever, I've done that for some PCI cards and North Bridge chips. And then there is always the backup plan. Sorry I didn't have any magicly answer. Good luck!

 

[cazwell220]

@purple_minion

I appreciate the extra perspective you have offered me. I will look more into the commands I can use from linux. I think I'm beating a dead horse at this point, but it sounds like spending another hour with this and a few extra commands won't hurt anything.

I will update if anything comes of it. Thank you for your time. I will sell them a backup solution in the meantime. I'm thinking doubling up carbonite for remote and syncback for local. I'd love to do incremental DriveXML backups, but as of now I don't think it can do that right?

Mark

 

[purple_minion]

@purple_minion

I appreciate the extra perspective you have offered me. I will look more into the commands I can use from linux. I think I'm beating a dead horse at this point, but it sounds like spending another hour with this and a few extra commands won't hurt anything.

I will update if anything comes of it. Thank you for your time. I will sell them a backup solution in the meantime. I'm thinking doubling up carbonite for remote and syncback for local. I'd love to do incremental DriveXML backups, but as of now I don't think it can do that right?

Mark

Well if you want to caulk it up to a learning experience then sure spend more time on it, but I have to agree, you're beating a horse that's been dead for a while. You just tell them what can you do if the drive doesn't respond, give them the option of a professional data recovery service, of course you'll be the middle man to help them with any technical communications etc.

PS. Maybe see if you have some usb drivers here for your enclosure.

http://www.usbman.com/index.html

There are generic ones there too, but you have to figure out your enclosure's chipset.

 

[purple_minion]

Hehe, how BRAVE are you.

http://hddguru.com/content/en/articles/2006.02.17-Changing-headstack-Q-and-A/

Oh, maybe you want to ask them there, they have a forum's section. I mean with a website like HDDguru they have gotta be good!!!!

 

[cazwell220]

LOL... not that I don't appreciate the extra mile here, because I certainly do, but I have to say.... not even on my best day

I'm just going to give it back and let them know a data recovery center is the best bet... and help them back up their data so it doesn't happen again.

Thank you for that resource though... maybe I will work on some old dead drives and that can be a niche for me someday.

Mark

 

[purple_minion]

http://datacent.com/datarecovery/hdd/seagate

Seagate drives operate under special firmware microcode that could also fail sometimes. Typically hard drives with corrupted firmware spin up normally, do not click but still fail to initialize. Such drives could have one of the following symptoms:
HDD is not found in BIOS at all
shows up with wrong S/N or capacity,
fails to read any data or boot up operating system.
If you attempt to boot up from such drive or read any data from it you would get "Primary Master Hard Disk Fail", "No operating system found", "Drive Mount Failure", "Disk boot failure. Insert system disk and press enter", "USB Device malfunctioned", "S.M.A.R.T. Capable But Command Failed" or some other BIOS hard drive error message on boot.
Seagate's new family - Barracuda 7200.11, specifically with FW version SD15 has especially high firmware failure rate. In such cases hard drive either doesn't show up in BIOS at all or identifies with 0 capacity. Unfortunately it is impossible to fix such drives without special equipment capable of accessing and repairing firmware modules in the hard drive System Area.