Linux.com

On drives where the motor's going, you can get it started buy shaking the thing. It won't work very long (usually untill you turn it off) but it does get it going for generally long enough to get the data off<nobr> <wbr></nobr>:)

#

The problem you're trying to get around is head sticktion. The manifestation is that your disks don't spin up to speed fast enough, and the startup process times out. You can actually hear the disks trying to spin up repeatedly.

Removing the drive, rotating it back and forth quickly on the axis of rotation can free the heads. Best way to do this is to hold the disk flat in your hand, and rotate your wrist quickly. Sort of as if you were, um, polishing a knob.

The key to drive recovery is to try the least-damaging options first. Eg: restart, spin, freeze, mild shock (1"-2" drop onto a padded surface), etc. Some of these actions can actually damage the drives (head crash sucks), so start conservative.

I used the hand-spin method to revive a set of 6 year old drives on a system which had sat unused for nearly a year, following a move. Running fine to this day.

#

One system I had running for around 5 years that ran pretty much 24/7 I had replaced and set asside. I decided to bring it back to life with Linux. When I powered up I noticed the hard drive wouldn't spin up. I decided to try rotating the system box and doing that I got it to spin up. I suspect that the bearings were marginal and/or the motor wasn't strong enough to break the initial friction.



Also back when there were disk packs (2' diameter Hawk packs) tempature could mean the difference between a pack that was readable and one that was not. I was installing software from a pack and in the middle it stopped reading. Some experimenting showed that as the pack was spinning in the drive it was warming up and getting out of alignment with the system.

#

If nothing else has worked and your only resort is to throw away the drive or rma it. I would highly recommend throwing it hard and flat against a hard carpeted surface so that the circuit board is facing up. Often times smacking the drive like this can fix some "head" problems. Especially if it stopped working after you dropped it in the first place. But again, only do this as a last resort. It has worked for me several times. Don't use the drive for long after that though, you should just grab your data off of it and then RMA it or chuck it.

#

I can also vouch for these findings. I have used that freezing trick to recover information from several failing drives. It doesn't bring them back to life fully (ie. don't try to boot off them), but I was able to mount them with a Linux LiveCD and dump what I needed back to another empty drive.

As for the laptop drive on a desktop, if you need something quicker, CompUSA does carry them, but they are about twice the price as the ones at Cables to Go. However, when you need that report ASAP, you might not be able to wait for shipping!

#

Many old drives suffer from a phenomenon known as "stiction". The lubricant built into the coating of the platter surfaces becomes, after considerable time, sticky instead of slick. when the drive is warm then spins down and sits til cool it can cause the medium and the head to stick togehter tightly enough to prevent the drive motor from spinning the platter. Manually giving the drive a spin (around the axis of rotation of the spindle) can sometimes break it loose, due to the inertia of the rotating platter/spindle mechanism. I've had it work many times.

I've also had the read/write head get ripped off the actuator arm, too!

#

Stiction! My oh my, I haven't heard that term since my BBS days. Which Seagate drive was famous for that, the ST-251 or something?

#

I do remember one of those<nobr> <wbr></nobr>:-)

I didn't solve the stiction problem by rotating the drive (didn't know about that solution back then), I just opened it and loosened the platters by rotating the spindle by hand.

Once I got some fingerprints on the platters.. slightly panicking, I just cleaned them with glass-cleaner (windex?) and it worked fine until the next time the heads got stuck.

I don't know how long they managed to use the machine after that, but it was end-of-life anyway<nobr> <wbr></nobr>:-)

#

You don't need to do that -- you're risking a big scrape on the drive surface, and consequent data destruction. Instead, just tap the center of the drive (over the spindle) as it first tries to power up. Start gently, but if the firmest fingertip tap you can muster doesn't do it, then switch to tapping one side (again, start gently). The object is to get it spinning again, NOT to initiate a head crash!<nobr> <wbr></nobr>:)

With this routine, I've never had a stictioned drive fail to spin up.

I have one old (850mb!) HD still in use that was stictioned but good when I got it... the routine went thus:

[taptaptap]<nobr> <wbr></nobr>... rrrr
[TAPTAPTAP]<nobr> <wbr></nobr>... RRRRRR
[WHACK!]<nobr> <wbr></nobr>... and it fired right up and has worked fine ever since (about 4 years now). I don't trust it with real data, but it's perfectly okay for testing motherboards and suchlike.

#

Mount the failed hard drive in an enclosure. Hotplug the enclosure and browse. Disks, that have MBR corruption or write errors and will not boot, will sometimes function mounted as read only.

#

If the disk is beyond frozen recovery Ibas kan probably fetch the data out. However they are very expensive so it's probably not worth the money for saving your ~/.

www.ibas.com

#

IF you have a magnetic force microscope handy, u can use that to recover data from the HDD by scanning it<nobr> <wbr></nobr>:).

#

Swap the platters out from the dead drive to the good drive, and pray. (If it works, throw them both away.) I've done this once, and it worked. The tough part is to see if the firmwares on the drives matches up. Ebay has "ask seller questions" for a reason I suppose.

--Adrian

#

Leave the HD's were they are and swap the PCB. Just make certain they match up. e.g. same familiy, equal or greater size.

#

Most times, its the PCB that's fried. Changing the PCB will allow you to at least keep the new drive after the recovery.

#

There's a reason drives are sealed. You'd really need to be desparate for the data to follow this route.

#

Want to look really cool on the day when your disk drive fails?

Have it mirrored, or in a RAID array. And keep a backup copy offsite.

You can still whack things with hammers or whatever, if that's what you like. I don't know whether you actually get points for style, but I'm sure you'd score higher if you did it out of joy than out of desperation.

#

in the case of a laptop - would you go through the pain of getting the hard drive out, or would you feeze the entire laptop which, I would imagine, could damage the screen/display/LCD?

#

Someone took a stupid pill this morning.

#

your intellectual superiority does not need to be demonstrated by your lack of courtesy. a simple and short answer would have been enough.

#

Common courtesy and respect is so easy, why do people need to take the low road?

To answer the question, take the hard drive out of the laptop and freeze it. Make sure it's lightly wrapped in a cloth to prevent condensation when you remove it.

It does not need to be frozen for a long period, just make it very cold so it can remain cold while you reinstall. Make sure you are setup to get the data off the drive FAST.

#

thank you very much for your answer. as for your question, maybe the previous poster is under the mistaken impression that all NF posters are skilled in hardware removal and experienced computer user. I suppose my question proves him wrong.

thanks again!

#

or would you feeze the entire laptop which, I would imagine, could damage the screen/display/LCD?



I sure hope not, or I can't leave the house anymore. It's -26 Celsius outside this morning. I've left my laptop in the car overnight on many -30 nights, and it's never cracked. I wouldn't recommend it as standard practice for sure, but they seem to tolerate the cold MUCH better than heat.

#

-26 Celsius?!

You have my sympathies! Were I live it is also 26 Celius today, just with a plus sign in front of it<nobr> <wbr></nobr>:-)

seriously, if laptops can take freezing cold, then what is the point of going through the hassle of taking the drive out, then feezing it, then putting it back in in a hurry so it does not get too warm (according to some posts above) when sticking the entire laptop in the freezer would do just fine?

(-: in case anyone wonders whether I have taken my stupid pill this morning - yes I have. I take it daily<nobr> <wbr></nobr>:-)

#

You have my sympathies!



None needed. We wear our frostbite scars like badges around here.<nobr> <wbr></nobr>:)



what is the point of going through the hassle of taking the drive out, then feezing it, then putting it back in in a hurry so it does not get too warm (according to some posts above) when sticking the entire laptop in the freezer would do just fine?



I have no idea, I'm wondering the same thing. The thing I worry about is bringing it in from the cold, and having frost collect on it, then melt. That's more of a concern than LCD damage. What is the liquid component of LCD's? And what's its freezing point?

#

Well I honestly don't know if your just lucky or what, I live in South Dakota and work at a university here that leases laptops out to it's students, and I can't count the number of times people left their laptops out in the cold and it's cracked the LCD not to mention hardware falure due to condensation.

#

Well I honestly don't know if your just lucky or what, I live in South Dakota



I'm not lucky, I just buy <A HREF="http://www.ibm.com/thinkpad/" title="ibm.com">Thinkpad</a ibm.com>'s.



I'm straight North of you, just over the US/Can border.<nobr> <wbr></nobr>:)

#

Hahaha, I should have linked to <A HREF="http://www.pc.ibm.com/us/thinkpad/community/legends/index.html" title="ibm.com">this</a ibm.com> page instead.



Check out <A HREF="http://www.pc.ibm.com/ww/thinkpad/community/legends/stolen.wmv" title="ibm.com">this</a ibm.com> story of one man's thinkpad enduring an entire winter in the cold. (As well as other ones getting run over by a <A HREF="http://www.pc.ibm.com/ww/thinkpad/community/legends/crunched.wmv" title="ibm.com">truck</a ibm.com>, and set on <A HREF="http://www.pc.ibm.com/ww/thinkpad/community/legends/crunched.wmv" title="ibm.com">fire</a ibm.com>)<nobr> <wbr></nobr>:)

#

If the data was important enough to backup, and all your applications are of the freely available open source variety, and your music is already on your MP3 player, go ahead and smack the hard drive as hard as you can three or four times with a Hammer. This will prevent you from screwing around with it for any longer than necessary. After all, new hard drives are cheap and unreliable ones should be quickly put out of their misery.

#

Pretty much any hard drive manufactured in the past seven years has S.M.A.R.T. capabilities -- that's self monitoring analysis and reporting technology. This won't help you revive a long-idle dead drive. It may give you advance warning of an impending drive failure.

You can access the SMART features under Linux with <tt>smartmontools</tt>. They key facts are these:

• A short test (runs in two minutes) detects most errors, but has a higher false-positive rate.
  
• A long test takes a while (nominally 20-30 minutes) has a very low false-positive rate.
  

Most distros allow for installing the SMART monitoring tools. Check your logs for any test errors. You can plan your drive replacement while data are still accessible.

Details: <A HREF="http://www.linuxjournal.com/article/6983" title="linuxjournal.com">Monitoring Hard Disks with SMART</a linuxjournal.com>, <A HREF="http://smartmontools.sourceforge.net/" title="sourceforge.net">http://smartmontools.sourceforge.net/</a sourceforge.net>. The linked whitepapers at the latter give a good overview of the stats associated with the tests, though the long/short test info is most of what you need to know.

#

question from a newbie:

is there any risk is using smartmontools? does it only mount and read drives, or does it actually perform any writing tests on them?

thanks!

#

The authoritative answer will be in the whitepapers, linked at the Sourceforge site in my previous post.

The tests are AFAIU not based on any destructive tests to the drives. My understanding is that most of the tests are checking historical registers (the drives log errors / issues), and checking performance to see that it's within spec (spin up, head movement, etc). A long test is good for some interesting sound effects....

If you've got an already dodgy disk, there's probably a possibility the the tests themselves will result in failures, so your first step should probably be to ensure your backups are current and complete.

There is a performance impact, particularly for the long test, and default installations tend to run this in what should be off-hours (your own lousy sleeping habits are not a developer problem.

There's a <tt>WARNINGS.gz</tt> file in my Debian install, which desribes a parade of horribles that may happen. Essentially: you can hang your OS fi calls are made on poorly designed hardware.

My own experience includes a drive which was operating far slower than spec, by an order of 150 times. Running the long test was simply not feasible -- I think it ran well over 12 hours before I terminated it (typical is ~20-40 minutes). But I took that as a strong indication that the disk was, in fact, not 100% dependable.

#

i once had a case of a SCSI HDD that wouldn't spin up.

i decided before I would ditch it I would try replacing the PCB with another drive.

just make sure the Drives are EXACTLY the same.. its easy a few screws and a few flat cables to swap...

#

Most hard drive manufacturers have hard drive test and repair software freely downloadable from their websites. This usually gets dumped to a bootable diskette or CDROM. You boot from this media and run their software. Sometimes you can even repair your problem in this manner. Often though it may only tell you what sector is damaged. But even this data is useful with Linux utilities such as smartmontools, hdparm, e2fsck, tune2fs, debugfs, and others. Not all is lost - USUALLY!

Larry Gagnon, A+ certified tech.

#

In the "old days" hard drive stiction happened a lot more often. Macintosh Plus machines came with 40 meg drives that would often stick as they got older.

Any computer with a horizontally mounted hard drive can often have stiction safely solved by doing the following:

Place the computer on your office chair (tie it on if you're a nervous type), give the chair a good spin and let it spin for a second - then grab the chair stopping it immediately.

The cetrifugal (or rather, centripidal) force will set the platters spinning again.

#

Chair is fine as a first step, but is likely too gentle. Just hold the drive with one hand by the edges and give it a very quick in-plane rotation. Repeat as necessary.

A truly stiction-stuck drive (any kind with surface lube, like those used in old Amigas) can be safely opened up in a "clean room" consisting of a clear garbage bag. Scrub hands, arms and screwdriver well beforehand, and tape the bag shut over both arms. Bags with lots of static are good; they attract and hold dust away from the drive. Once open, there will be an area where you can press a thumb against the edge of the disc and give it a shove in the direction of normal rotation (along the read head arm).

Once I used a black garbage bag, and did the whole "repair" by feel. In that case we remembered to run park _always_ before shutting down, and never got stuck again (remember "park"?). The drive worked for another whole year after that as a temp drive, with no bad spots, luckily.

Hooray for desperate foolhardy experimentation.

#