So you locked your backups away for years, huh? Allow me to introduce my colleagues, Brute, Force and Ignorance Welcome back to On Call, The Register's wall where readers inscribe the antics of users so those on the other end of the phone might consider their career choices. Today's tale sees the return of Register favourite Alessandro, whom we last encountered "taking one for the team" in a real sense as he fought to hang on to his …
*Taps tankards in toast*
That T-shirt looks good on you!
*Produces a messenger bag full of assorted hammers*
Wanna go void some warranties? =-Dp
*Brandishes my favorite Bonk-Squeak-A-Hammer-Of-Doom-+20*
Mine makes a happy squeaky noise when I whack stuff!
*Promptly wanders off whacking myself in the head, giggling at the squeaky noises, & trailing my Dried Frog Pills*
"Mine makes a happy squeaky noise when I whack stuff!"
Somewhere in a box, in the garage, maybe, I have a small plastic toy hammer from a trade show that makes the sound of breaking glass when you smack something with it.
The perfect tool when working on a Windows PC. It saves me using the the Bob Hartley axe handle framing hammer.
>> couple of twists of the disk and it miraculously returned
Wow, this takes me back.
A long time ago, one of my co-workers showed me how to manually spin the disk on one troublesome system when needed. It was usually required if it had been shut off for too long, cold to the touch. I believe it was an early IBM.
Seems like I remember about that same period... having to manually "park" the disk on some systems before shutting down.
That sounds like my comment whenever things are looking like being a little troublesome -
"I've got a 3lb forge hammer in the car if you need it..."
I ACTUALLY DO carry said 3lb forge hammer in the car but only for if I ever need to change a wheel since they have a habit of seizing onto the bit around the hub. Scared the life out of one user when I got it out of the car and brought it in (we were re-adjusting the shelves in the store room at the time, theirs was just an incidental job as I was passing).
Long time ago, in a Leeds far, far away.
Couldn't get a PC connected to a power dialler to boot.
Called PC support (or more accurately, the beating heart of the IS 5-a-side footer team).
Turns up.
Taps case in a few places.
Rolls up sleeve.
Power button -WHACK!
PC boots.
"Not just hitting it - it's knowing where to hit it..."
Quantum 105MB drives in early SPARC stations would stick after a while. The solution was to lift the front of the pizza box up about 2 inches (no more) and let it drop back to the desk. Then power on. Occasionally you needed an fsck to collect the loose bits afterwards...
Yeah, my first hard drive was a Quantum 80MB for my Amiga 2000. After a few years it started to not boot after the computer had been off overnight. I had to whip the case off, pull out the card (which had the drive bolted onto it), give it a good shake, plonk it back in and it would usually start up again. When I could afford it, it got replaced by a "huge" 540MB drive.
It was probably a 105 MB Quantum drive in my Amiga. When it started sticking I'd take it out of the Amiga and rotate it back and forth a few times along the drive platter's axis. More work but less drive stress. And not too much work when the drive first starts sticking.
Yeah, my first hard drive was a Quantum 80MB for my Amiga 2000.
The next page of comments gets into loose connectors. It reminded me of my first Amiga hard drive experience. I had a GVP A530+ (68EC030, 8MB RAM, and a 40MB Seagate SCSI hard drive) which would sometimes just not want to boot, throwing up the Kickstart "insert Workbench floppy" screen. I fussed with that thing so much, seating and re-seating everything including the SCSI cable. Not even swearing, begging, and praying would fix it.
TURNED OUT that the SCSI cable was the problem, but not because it kept working loose. This particular cable had something weird in the IDC connector which made it work only when it was NOT fully seated. I figured out putting the lid back on the accelerator would move the cable just enough to un-seat it into a working position. As this was a two-inch cable likely made specifically for the short distance between the SCSI header and the drive I really wanted and tried to make it work, but wound up replacing it with a longer cable onto which I eventually hung three other SCSI drives for a whopping 500-some MB of total storage.
Back in the day when I was starting to assemble computers for myself, a case seemed frivolous and costly. I managed to trade for a 286 board along with a 320 Meg HD. The advantage of having no case was that the hard drive had a bearing issue and needed encouragement to get going, so I had a screwdriver sitting on my desk next to things for when the computer got shutdown and the hard drive needed to be spun up (there was a slot on the end of the spindle.....). Was a budding young hobbyist at that point, but it didn't take long to realize I needed to upgrade that drive as soon as could be afforded. I got a lot of miles out of that drive (probably a year or more) and passed along to someone else who used for another 5 or 6 months before it finally gave up the ghost.
Then when I was in college, I came very close to losing my life's work when my IBM Deathstar (Deskstar) bit the dust with the click-click-click on boot after my semester ended before exams.... Luckily all projects had been turned in at that point and it was just a matter of downloading class notes again to cram for exams... Ended the semester on a high point academically, but if that drive had died a day or more sooner than that, I would have been screwed. We were told not having a backup was no excuse and they meant it!
No, you just need to hit it with EXACTLY the right amount of energy to excite it briefly to the higher level.This is an unstable state so it re-emits that energy as EM radiation (or "bloup", as Alessandro's customer would put it), and returns to its normal, stable state.
Or hit them with sufficient energy to cause them to emit both a functional and nonfunctional version of themselves, and then just collect the functional one.
Assuming you can catch it since the slightest error in the whack energy and it may be travelling at near lightspeed.
Headley_Grange,
"I assume that when we get real quantum computing then you'll have to hit them and not hit them at the same time."
Yes and No !!! :)
You could also, due to Quantum entanglement, just whack the 'twin' you keep in the office and avoid the journey to the customer. :) :)
Did this exact thing a few years ago.
Senior exec kept everything on a USB 3.5"external drive - no backup, obviously, because.
Sudden problem - drive not recognized.
Plugged it in, ear to drive - click, click click. Drive head stuck in "park".
Made a big production of holding it *precisely* 6" above the desk (measured with a ruler for best effect) and slammed it down.
Drive kicked into life. Copied everything back onto his PC and then threw it in the bin.
Later heard that the reason they guy was sweating like a bison was the drive held a draft financial report he'd been working on for days.
User education ensued.
I did that for a colleague whose PC wouldn't boot. Waited until the disk should spin up and dropped it. Presto, the resumption of disk spinning noise.
"I can't believe you did that!"
"Well it wasn't going to make it any worse. I suggest you copy your data off while you can."
The old 10Mbyte drive attached to our (6801 and M68000) processor emulator at Racal in 1985 would occasionally do the same thing, and a swift half-inch drop onto the bench would deal with it.
The Sun engineers were a little more wary of the new 1.3G 5.25" full height disks in about 1991. Apparently the oil in the bearings could get a little too viscous if they were left powered down in a cool room for too long, so the advice was to lift the box and *gently* rotate it to start it spinning up.
Its the old joke, guy goes to mechanic with broken car. Mechanic sucks teeth for a minute, then pulls out a hammer and walks around the car. Finally he hits it once, hard and everything works again.
"£1000 please"
"What? But you just hit it once with a hammer!"
"£10 for hitting it with a hammer, £990 for knowing where to hit it"
My late father-in-law was a TV repairman in the 60s and 70s, and he used to have the apprentice distract the owner whilst he gave the box a smack to break the tin whiskers! Often he could then just furtle around drinking tea for half an hour and declare it fixed (got to make them think they are getting value for money).
When we were designing laser printers the optics would get out of alignment so I used what I called an "optical sledge hammer" to realign the various mirrors, modulators etc. You did have to know what you were doing as a random whack would just throw the thing further out of alignment.
The service personnel were not allowed to touch the laser "sled" but mailed it back to the factory for realignment. It was then mailed back out (probably destroying the alignment again).
And of course for those devices and drives which are truly frustrating and reluctant to submit to the mercies of percussive maintenance, it's not just knowing where and how hard to hit them, but also to know when to stop doing so.
The age-old balance between releasing pent-up anger at the thing and actually getting it to ever work again...
I once called my mechanic and told him my car wouldn't start - engine turns over but just won't "catch". He asked me if I had a rubber mallet, and explained exactly where to hit the bottom front of the gas tank, while someone was turning the key, to temporarily unstick the fuel pump long enough to drive it to his shop. There's a REASON we're repeat customers...
Used to work in the desert, pumping cement down oil wells. The pump trick had two large Cummins diesel engines to drive the pumps, and a smaller engine to drive the hydraulics. Got called on site in the early hours to be prepared for a pumping job and the smaller engine refused to start. Quick call to the Head Mechanic who lived about 50 miles away. He turned up about 40 minutes later without a tool box, just a hammer and WD40. Got is up and running in about 5 minutes with judicious application of force.
Most V8 and some V6 Ford starter motor solenoids are notorious for failing at inopportune times. A good wack with a tire iron will usually unstick it in an emergency. (You often have advance warning that permanent failure is immanent because the car fails to start when hot, but starts fine when it cools down & the tow-driver gets in and casually starts it ... fortunately starters are an easy R&R.)
Come to think of it, Granpa used to call a 5lb ball-peen "Ford tool #1" ...
There's nothing peculiar to Ford, starter motors or engine configuration here, most parts of all American cars are notorious for failing fullstop.
Your grandpa called it his,"Ford tool," somebody else's called it his,"Chevy tool," and yet another called it the,"Dodge tool." Hilariously.
high school, circa 1970, 'earth sciences'' teacher drove this ratty mid 60s Jeep Wagoneer in which we did many field trips. It frequently wouldn't start when hot, he'd pull the rubber mat up near the gas pedal, there was a strategic 1" hole in the floor, through which he'd bang the starter a couple times with the crowbar kept under the seat, vrooom, chugchugchug...
> "What? But you just hit it once with a hammer!"
I needed to remove the stuck viscous fan on my Land Rover. The manual insisted all kinds of special tools were necessary for the job, so I took it to our local garage. The mechanic and owner took a look, went inside and came out with an air ratchet, to which he fitted a percussion tool. He aimed it a nut on the fan and gave it a short "prrrp". The he simply spun the fan off its mount. It had taken five seconds. I was deeply impressed. "That took five seconds and twenty five years," I said. He smiled, understanding me perfectly.
Many, many years ago, I attended a BBC Micro that steadfastly refused to boot. Took the lid off, all LOOKED fine. Tried reseating all the chips, most of which were socketed in those days. Felt a light CRACK! of static as I touched the CPU. System springs into life (ready to go in under a second - those were the days!).
"Do you by any chance have nylon carpets?"
*shakes head*
"A predisposition to wearing shell suits?"
"No."
"An amateur radio enthusiast nearby?"
"Actually, I have a ham radio rig just here in the cupboard alongside the desk. I put the computer next to it a couple of weeks ago so I could use it for logging callsigns and stuff."
"And this cable which I see now goes out through a hole in the window frame and which I assumed was for the TV is the feeder cable for the aerial, yes? The one that is bundled up with these other cables some of which connect to the computer? You DO have a static bleed resistor fitted on that aerial, right?"
"Erm... "
I can do you one better than that. I used to play guitar and bass in a covers band. One gig, we all got back up after a break, and I found my guitar sound had gone strange. Lower strings produced noise, albeit muffled, but higher strings didn't do much, and all the level was down. I checked all my gear to see what settings had got changed, but everything was fine. I assumed it must be an FX pedal gone faulty and went round swapping them out, but no joy. Me and the other guitarist both played bass and guitar, so we finished the gig handing instruments between each other, and I went fault-finding at the end.
Turns out it was a cable fault. Somehow, and I don't ask anyone to explain the physics of it, a cable had failed *just enough* to still pass a signal but turn the cable into a low-quality low-pass filter. Lower notes came through with distortion; higher notes simply didn't.
I've seen plenty of dead cables, and plenty of intermittent cables, so if it had been anything like that then I would have immediately started swapping cables. This is the first (and so far only) time I've seen a cable turn into an audio-frequency filter.
> This is the first (and so far only) time I've seen a cable turn into an audio-frequency filter.
That can happen with squished cable. If squished well enough the isolation between the wires gets so thin it starts to work as a capacitor. But this is the first time I heard of it when it comes to guitars, must have been very close to a short out. It happens more often with networking, monitor, WLAN cables and other high-frequency connections.
"It happens more often with networking"
Called to a customer site where the complaint was the network was running slow. 10BaseT network. The co-ax from the server started the run to the office PCs by going under the carpet before running along the wall. It was crushed flat and didn't look in the least like it might have had a rounded cross-section at some point int past. No, that is NOT where the cable was when it was installed. They'd re-arranged the office and re-routed the cables themselves. We marked it up as user damage and charged accordingly to replace and re-route the cable with proper trunking, ie like when we originally installed it but using their new routing and office layout.
Yes! I had that happen myself. A tenant moved in upstairs from where my family had their business. My bro and crew were wired up at the time with 10BaseT coax. We ran the cable up into the ceiling and along the wall in our storage closet and down the other side. We had not choice and couldn't run the wire in the wall due to the walls being made of brick due to the space being an old textile factory, otherwise, that would've been the better option.
One day things didn't work as you said stuff was a bit slow. I got called at work to go take a look. The new tenant had taken over my family's storeroom without asking and put his file cabinet on top of our network cable even though the cable was place tightly against the wall. The cable was no longer round and looked like an old TV aerial that would be tacked to the wall.
I had a similar thing except with video. I was using a Varityper Epics 20/20 typesetting system which had a minicomputer, terminal, and output device. All three devices were connected together with a proprietary cable that connected up the three units that made up the system. The system was used in home office. Our cat liked to chew on plastic, and had an affection for computer cables.
One day the terminal had blinking dots on the screen that would come and go, flicker, and then everything was fine. Outputting stuff to the 6830 was awfully slow as well. Instead of taking a minute to output a full advertisement, it was taking up to 10 minutes. Then everything was fine... Accessing the mini to save jobs was slow as well. Then again I was writing to a 20 MB, yes a 20 MB, Miniscribe full height drive so that was not necessarily the fastest device.
I checked the connections to the devices and everything was snug, and I went as far as to open the terminal and re-seat the memory and the multiple CPU's in the terminal, which was all socketed. This was all on multiple boards covered with 4164 DRAM chips and 68k processor boards. In the troubleshooting process, I did find a bent pin on one of the DRAM chips and fixed that, but all that effort didn't seem to do anything different. Occasionally still there was a flicker and blinky dots and lines.
I left it alone since the 6830 outputter decided it was going to work, and the minicomputer part was working fine until one day I saw the culprit. In walked my rather fat and quite old cat. He sauntered over to the fat RS422 cable and had a chew... I chased him away, and sure enough the cable was completely eviscerated with holes all over it, some of which were deep enough to puncture the foil ground sheathing underneath!
A call to the company and $750 later, the cable was replaced. The blinking lines and dots went away, and all the other devices were back to their high speed again.
I've seen this before.
PC, loose power cable not making contact.
Support walks around PC, tapping it etc. Unseen while they get your attention with the tapping, they push the plug home.
Smack the case, then hit the power button
I'm sure I recall a similar case, where for the "big reveal" the techie did a truly apocalyptic bit of percussive maintenance, but what they were actually doing was walloping something nearby while giving the impression that the computer was taking the force of the hit.
Of course, next time the problem occurred, the user decided to repeat what they thought they'd observed the techie doing, and hit the PC so hard that it actually did break it for good.
I'm sure I recall a similar case, where for the "big reveal" the techie did a truly apocalyptic bit of percussive maintenance, but what they were actually doing was walloping something nearby while giving the impression that the computer was taking the force of the hit.
I have done precisely this trick on the office coffee vending machine, which was powered by a small computer onboard computer to process the options etc on the buttons. It crashed every so often, and the engineer from the supplier said just to reboot it if the machine locked up.
In time I discovered that there was a cutout switch on the door, which you could trip by pulling the door slightly open at the top, even though it was locked shut (the lock was in the middle). The trick was to lean against the machine, pull the top open and restart it, and then give it a good thump in the middle just as the BIOS in the machine inside hit the POST stage.
I'd do this frequently in the mornings for drinks instead of phoning out an engineer since this workplace had banned kettles. One morning going for my morning drink I discovered two people methodically beating the crap out of the front of the machine with varying forces of hit in different places. I did my little trick, only to discover that two service engineers had been watching.
After chasing the users off, the senior chap sent the junior engineer back to the van for a part he didn't need specifically to get him out of the way so he could get me to teach him how to do that trick.
I stopped doing it about the time that somebody actually broke a hole through the front of the machine trying to recreate my fix. :/
Of course, next time the problem occurred, the user decided to repeat what they thought they'd observed the techie doing, and hit the PC so hard that it actually did break it for good.
I learned over the years that certain customers should never witness or have explained to them how you fixed a specific problem. Then any time there is a problem they will attempt the same procedure in an attempt to avoid calling (thus spending money on a visit.) Ultimately the call comes, usually after a long bit of down-time and stewing in frustration.
I call this the "man with a hammer" problem, hearkening back to the proverb, "To a man with a hammer, everything looks like a nail."
Of course, there's the dreaded evil twin of the loose power cable: the power cable that's only just in and which falls out when you adjust a nearby cable.
I've encountered it several times before, including just a few days ago whilst moving power cables into a new UPS, my hand brushed the input for the old UPS.
Cue a slight crackling noise, and the unsettling *clunk* as it tried to power everything from a very poorly battery, ran out of capacity in half a second, and clunked offline.
A quick shove and the power cable was seated properly again (seriously, how did it last 10+ years with the power cable only barely plugged in?), but the damage was done. One old, single PSU'd, machine was already rebooting, causing all sorts of issues.
"(seriously, how did it last 10+ years with the power cable only barely plugged in?)"
Most likely, it didn't. It was probably only a little loose originally and then there were probably temperature changes over the years and the expansion/contraction cycle pushed it further and further out.
Just remembered a war story with PA equipment. Old-school 2kW power amp, heavy old linear power supplies. A total sod to cart around, but it was utterly reliable. Except one day where the sound just started fading away quieter and quieter. I couldn't figure out why. Turns out the power cable had got loose - but the amplifier power supply had such large smoothing capacitors, and the amplifier itself was efficient enough without much signal, that it took a couple of minutes before the capacitors finally ran out of juice.
@baldrickk Support walks around PC, tapping it etc. Unseen while they get your attention with the tapping, they push the plug home.
Thinking of all that tapping and mystique then using sleight of hand to distract the user while pushing the plug in reminded me of this quick sketch by Smith and Jones years ago. Shows the lengths you can go to, if you have the time and a suitably gullible user!
Recently I unlocked my car with the key fob, opened the door - no light. Hmm. Got in and tried to start it - no reaction whatsoever. Hmm. On a hunch, I stepped out, and (really, truly, for troubleshooting only, I really wasn't angry) SLAMMED the door. Opened it up, light came on, car started fine. Issue was a loose battery cable (since fixed).
The central locking is a bit dodgy on my knackered 206, and recently every time I tried to lock the car, the locks would pop back up again after a second. Turns out the fix is an easy one, simply SLAM the passenger door closed hard, and it works again.
Weirdly, the problem crops up even when I've not used the passenger door.
Funny.
My Bosch fridge actually has a power button inside the main door. While it does have the standard "line-in-a-circle" power symbol on it, not a lot of people know what that means in this context...AND, it gives no indication of its state (it's a momentary push button)
The cleaning people accidentally pushed it once, and, yes, it does turn off the refrigeration.
So, there's now a label next to it: "DO NOT PUSH THIS BUTTON"
This post has been deleted by its author
In the 80ties, I worked as a student at the campus Apple center, and later at the big Apple center in town.
We has harddisk issues as well. The Sony 20MB used in Macintosh SE/30 had lubrication that got thicker over time. So at some point, you could not let the computer stay without power over the weekend, or it would not boot monday morning.
First trouble shooting was to try to turn the whole computer 180 degrees back and forth more or less around the center of the harddrive. If that did not work, open the computer, out with the disk, and then in one hand twist it back and forth almost like making a milkshake. Then try to boot again.
It is unbelievable how many customers continue with the defective disk instead o just buying a new one. But SCSI disks where expensive at that time. And if not powered off, they could run for decades.
One of my first encounters with SCSI was in the late '80s with a Mac where some genius had set the SCSI ID for the boot drive to 7, instead of removing all the jumpers and leaving it at its default 0. This meant that when they added a device (in this case a SyQuest 44Mb drive), even if switched on without a disk inserted then the computer would repeatedly try and boot from that ID before eventually geting to ID 7.
20 seconds to fix the problem and 10 minutes to make it look very complicated indeed.
Ahhhh! External SyQuest SCSI drives! With those little buttons to choose the SCSI ID? Just the sort of thing to press with your finger when moving the only drive in the office (cheap bastards) between workstations.
If I had a quid for every time I fixed one by telling them to change it to anything but 0 or 3 or 7 (IIRC) then restart, I'd probably have the same amount of money I have now due to ------->
Oh the other thing with those Syquest drives...
If there happened to be another one lying around, the platters from one would not work in the other. The discs tended to work only with the device they were formatted in, causing some rather frustrated customers who brought their artwork to the printer for printing only to find out that the printer couldn't read their discs!
So much for portability!
Hammering isn't necessarily the best option here. Giving the box (or drive if really bad) a good twist in the disk's horizontal plane was usually the best option.Had a 486SX box as a work machine that ran for a year or so like that.
We did have one really stubborn drive that, as a last resort with the data lost anyway, we took the top off and manually spun up the platters to get them moving. It survived long enough to retrieve the required files.
I had to manually spin up a 2.5" external hard drive with the top cover removed recently (1Tb).
It had been dropped while running and the power had disconnected at the exact moment it hit the floor meaning that a combination of g-forces and lack of power left the heads on the platters rather than retracted.
when it tried to spin itself up it couldn't retract the heads (because they weren't floating) and it couldn't turn the platters because the heads were stuck.
A well timed spin of the spindle got the platters moving (with a screech) and the drive came back to life with everything apart from a couple of files intact.
once copied the drive was destroyed with a hammer.
And they normally would, unless the drive is powered until the moment it hits the floor. The impact can whack the heads against the platter hard enough to lose the air cushion and stick them in place. Some (mostly laptop) drives now contain a accelerometer that parks the heads if it detects a 0G situation (freefall) to protect the drive
you could actually hear a computer boot: first...
Even up to these days, sequence can be crucial in problem solving. Recently, on a unofficial side job, I had to extract an Exchange mailbox in the dark hours. My only access was the client computer with restricted user permissions. Clever admins prevented pst files and I knew no other feasible means to get the data out. I couldn't wait till support hours because I needed the data urgently and, just as importantly, didn't want to answer inevitable questions of why.
No problem, right? Quick search through the registry and the right key is found. Change the reg key, start Outlook and - nothing. The bloody thing (the reg key) is reset every time the bloody thing (i.e. Outlook) starts. Obviously, the reg key had to be reset before it was read and applied again. Solution: frantically re-applying the reg key to the desired value while Outlook starts. And the mailbox's GBs started flowing.
Samsung HDDs. They'd sometimes make a strange clicking noise and the BIOS wouldn't find them.
If you rapped them on the top with your knuckles, or gave a swift tap with a screwdriver handle, they'd spin up and you'd be able to copy the contents across to a replacement drive using Ghost.
Happy days.
SWMBO uses a shoe as a hammer and a butter knife as a flat bladed screw driver. She says that if they don't work then it's too technical for her and I have to fix it.
As she doesn't wear stilettos and most screws are now cross head, I deem her antics become safer as time progresses and everything gets smaller.
Let me make sure I understand this, not only did you get the machine back up, using a technique which was frequently employed, the customer then decides that he DOESN'T want the person who sorted it out in the first place?
There is zero logic there, he fixed it, thank him and get him back!
"Alessandro" here: the boss ended up sending me back anyway having told me to explain said accountant that my daily job was in software engineering, with an occasional side job in system engineering. I also explained that if any other engineer in a 30Km radius would have been sent, the verdict would have been:
Sorry, all of your data is lost. Buy a new computer and enter all data again...
A long time ago, I had heard that putting a disk in the freezer was sometimes effective for data-recovery. I thought is was a good spoof for ordinary citizens to open themselves to ridicule.
In desperation, after multi-plane, sub-destructive percussion had failed, and with no-one watching, I put a troublesome disk into the freezer compartment of the fridge. Like the disk, I was stunned, but we both recovered for long enough to extract the data files.
And as the extension/variant of that: Once the drive is chilled and the backups are inevitably taking bloody ages because there's nine zillion tiny files that are being copied because you're trying to get everything possible and the customer isn't quite sure exactly where/which files are needed...
Spraying an inverted can of air duster at the thing while it chugs and tries to warm up.
There are some tasks that I ensure I'm not being watched while I perform:
-Most percussion fits that bill
-Anything that looks like a pagan ritual (using a freezer while chanting [usually "Work you f@#ing thing" repeatedly], and the like)
-Doing things that are clearly not to be imitated (scraping off a "no user servicable parts" sticker off a PSU to get the top open, swapping disk drive PCB's, running a motherboard in the sink to clean enough nicotine off to catch a buzz, etc).
I used this technique to recover data from a dying hard disc from an Acorn RiscPC. Drive attached to a USB adapter and a long-ish lead, I had it in a couple of zip-lock bags, let it cool down then just dd'ed the thing to my Linux machine. It ran long enough to get an error-free image of the entire disc, now that image lives on with an SD card and an IDE adapter, and the RiscPC has never run so fast! It's also much easier to back the thing up now as the SD card is easily accessible to be imaged.
They were getting away with providing the customer zero service and not doing any preventive maintenance even though everybody knew what happened inside microcomputers in the disgusting office environments of those days. It is not a good look for the company.
I once had to tell a designer that his brilliant idea of carving up some MDF in his office to make himself an ergonomic desk had destroyed both the graphics card and the cpu of his computer, as the MDF dust had stopped both fans and the relevant bits had overheated. On that occasion it was the disc drive survived.
I used to get calls all the time from friends of friends (who wanted cheap home IT support) who had a tower system at home that 'just stops working after 5 minutes'
My first question was "Do you have carpets?"
In almost every case the user had a floor mounted tower on wooden floors in a corner somewhere that collected the dust and fluff and clogged the CPU fan causing a shutdown on overheat. Carpets collect the dust anf fluff, wooden floors somehow shove it all in a corner.
Quick vacuum and an air blower and I accepted my payment in beer or whisky.
Sadly my free beer supply dried up with the switch to laptops and tablets.
It's called "stiction".
First unplug the hard drive then plug it back in again. That's the power and both ends of the data cable(s). Then turn the box on, see if it boots. If it does, it was poor termination and you've temporarily fixed it. Back up the complete system before doing anything else.
If the above doesn't work, turn it on, wait until the drive just starts to spin, rap the computer case with your knuckles parallel to the plain of the platter(s).. Sometimes a good hard slap is required. If that works, it was stiction and you may have just booted it for the final time. Back up the complete system before doing anything else.
If that doesn't work, I recommend (and use!) Drive Savers. Spendy, but they'll get your data back if it's at all possible.
I should add that the drive in question can no longer be trusted. You'll need a new one. Preferably a completely new computer system. However, in the case where your old software won't run on modern equipment, there are reputable companies which sell known-good vintage replacements. I use and recommend Excess Solutions in Santa Clara, now that they have taken over Silly Con Valley's legendary HalTed.
Preferably a completely new computer system.
Indeed. Some twenty years ago I replaced one PC's crashed hard drive a few times. I was suspecting its user of performing unjust percussive maintenance procedures on the machine. But it turned out to be a dodgy HD controller on the mainboard, which needed to be replaced.
Indeed. Some twenty years ago I replaced one PC's crashed hard drive a few times. I was suspecting its user of performing unjust percussive maintenance procedures on the machine. But it turned out to be a dodgy HD controller on the mainboard, which needed to be replaced.
Had a client we delivered a brand-new Netwre 3.12 system to. Said server had a VLB IDE HDD controller, and was mirroring to two HDD's (Netware's SFT II)
Anyways, client phoned later (I was on standby) and complained about data getting hosed. So I did a reinstall, restored data, and all was well - until a few hours later.
Took the machine to the office, ripped out the fancy VLB ISA IDE card, plonked in a standard ISA IDE card, reinstalled, restored, and delivered to customer.
No more issues.
I left that buggr'd VLB IDE card somewhere, somebody snaffled it later on, and I chortled with glee.
Yeah I recall the stiction problem with some IBM (I think) drives in some NCR storage back in '94-95. I was working for NCR at the time and knew a few of the field engineers (mainly as they kept borrowing items from my benchmarking centre stash). The (maybe?) documented procedure was for the engineer to remove the offending drive, "lightly tap" it against their knee, then replace the drive.
I recall, in the dim & distant past, that I acquired a utility that allowed the users to "park" the heads close to the spindle - rather than anywhere else on the "Viagra" disk's surfaces. [40MB MFM Seagate]
Not so much effort to break the "stiction" if the heads are parked up close to the spindle.
So many years ago that I completely fail to remember if my acquisition of this utility was ever strictly necessary, or whether it was a "just in case" belt, braces, and a long bit of string" to "save my trousers hitting the floor". <Grin>.
Yup, sounds like a good old Seagate ST-238 starting up..it would spin up the motor, and then move its heads up and down a bit as part of a self-check..
Those were indeed the days.. they were notorious for locking up if you powered them off for a while. Always had a small plastic hammer in my toolbox :-)
I was once presented with a Dell PC that wouldn't boot and had a strange sweet, oily smell.
It turned out that "a friend" had told the owner that it wouldn't boot because the hard disk was sticking.
The owner had therefore sprayed WD-40 into every opening in the machine's case to "free it up".
The PC actually survived this treatment and a small amount of ... er ... gentle tapping on the hard drive, restored it long enough to extract the data.
WD-40 isn't a lubricant, it's a solvent so using it on sticky fans will work to get crud out of the bearings. It's not really for use on electronics since it's bad for some plastics which aren't petrochemical-resistant unlike the materials used in car engines. I use a can of spray-can of isopropyl alcohol to fix problems with fans and computer mice etc. but if WD-40 is what you have to hand then go for it.
Icon 'coz WD-40 was invented for rockets.
The sticky residue is actually the intended function, not a bug, in WD-40.
It's a water displacement product, intended to be sprayed onto seams where it'll seep in between the material and form a residue there or for covering a surface and form a protective film, preventing corrosion. The aforementioned rockets (The Atlas specifically) were one of those applications.
What you usually want is "penetrating oil" -- something like "Liquid Wrench" or "PB Blaster" or "Kroil"
All three are basically light hydrocarbons (check the MSDS) and will loosen up gummy lubricant. Some of them leave a lubricating residue. PB Blaster seems to work better than the reformulated Liquid Wrench, which used to be my go-to.
Not spinning rust but I was once summoned to a customer site to fix a computer that was randomly stopping. This was pre-pc days and the box in question was an ISC3651, an 8088 based colour desktop. It had a big proprietary parallel port on the back for which we had made an interface box.
This machine had been running an engine test bed for a couple of years before we got the call.
Pitched up onsite and started to investigate. Everything seemed normal. Took machine apart for a close look and... Nothing.
Rinse and repeat.
I'd just put everything back together for the umpteenth time and realised I'd forgotten to connect the interface box. Never mind, it wouldn't object to being hot plugged, so I leaned across the machine and... It stopped.
Top off again and a really, really close look at the main board.
The OS on these boxes was in PROM and it turned out the machine had been running happily for two years with one PROM datapin never having been soldered to the PCB, just pressing on the side of the plated through hole.
30 seconds with a soldering iron and all was well.
Two years though. Wow.
But I always explain to the customer the internal mechanics of the drive and the reason for whacking it. Along with explaining that I've carefully assessed it and know that it is this particular problem this time, so don't try it when I'm not around.
I have also resorted to taking the top off and spinning the platters manually to get them moving - obviously didn't reuse the disk once I'd (successfully) copied the data.
This post has been deleted by its author
My Uncle Hughie (actually my father's cousin) had been a heavy-horse farrier in his youth. His motto was "always use a big hammer to start with." His reasoning, and I could never fault it was that there was no limit to how gently you could hit something with a big hammer but there was a limit to how hard you could hit something with a small hammer.
Crayon munchers brought in a macintosh se FDHD to show off how long he'd been using Apples.
Naturally he plugged it in to see if it still worked only for angry noises and no desktop.
Out of idle curiosity I ask what he's doing listen to the explanation, then the nostalgia and whilst he's wittering on, proceed to just hit the back of the erstwhile machine, to which it promptly finds the disk and starts.
He looks at me, I look at him and then wander off for coffee whilst he continues to stare. The look he had that I not only had the belligerence to strike his beloved childhood friend but also that it turned it on like a PHB in a bondage parlour...
Fun times...
I'm surprised nobody has mentioned the classic "turn it upside down" treatment. I've lost count of the times I've restored a failing drive to temporary serviceability by simply turning it upside down. Stiction can occur simply because the lubricant has very slowly moved under gravity and a simple gravity reversal can give you full serviceability for at least a few days (never tried it for longer).
During my time at Time Computers I regularly got the help desk staff to talk customers through removing hard drives, giving them a sharp rap on the floor and replacing them.
It would free up "frozen" read heads and get the drive working long enough to back up data.
Quite a common problem, especially on 80gb Seagate and Hitachi/IBM drives
Years ago I had a customer whose computer refused to boot,she called up (they were about 1000 miles away) and we talked about a few possibilities for getting it up again. A day later she had it up and running, called and said that the problem was the hard disk and I walked her through the backup and replacement procedure. After it was all done I asked her how had she managed to get the system up and running again?
She told me that she'd listened to the system starting up and realized that the disk was not spinning, so she had undone the screws on the top cover of the disk, removed the cover and rebooted the system. Then she put her finger on the disk and gave it a little push to get it going! The system booted and she kept it running while the backups were made!
I haven't seen anyone mention the all-time classic Tale Of The Stuck Drive.
The A-6 Intruder was an interdiction/strike attack aircraft used by the United States Navy and Marine Corps starting in the early 1960s. It was an all-weather (that means it could fly at night, night was a weather condition to aviation types at the time) two-seat twin-engine medium bomber with state of the art electronics, including the most advanced bomb-aiming computer in the world, complete with that magical new invention a drum drive. (Those of you who have encountered drum drives may alrfeady knopw where this is going...) The A-6 was heavily used in the Vietnam War, inspiring among other things the book and movie Flight of the Intruder. The crews of early versions discovered that the drum drives didn't like being subject to high gee forces, such as violent evasive action to avoid anti-aircraft fire and SAMs, and would stick, and the bombs couldn't be dropped without the advanced bomb-aiming computer, so all that trouble getting to the target went for nought. One frustrated Marine kicked the drum... and it unstuck. So standard operating procedure was to kick the drum whenever it stuck. Yes, folks, you're sitting in the most advanced attack aircraft in the world, flying over Hanoi with lots of 100mm guns and SAMs and MiGs all after your blood, and you're kicking the computer to make it work so you can drop the bombs and get the hell outta there. Gotta love military tech, you really do. The drum was replaced in later versions, and a simple bomb release which didn't depend on the computer added, just in case. The A-6 kept flying for the USN and USMC until the early 1990s; a specialised electronics warfare version, the Prowler, showed up in the early 1970s and only retired last year.
As a PFY, I worked in research for a company developing microwave oscillators for a combat aircraft.
One day I approached the test bench with a powered device attached and accidentally kicked the bench as I got seated. Result : level on power meter goes AWOL, frequency goes off analyser screen.
Approach boss to point out that this may not be a "marketable feature" for the particular aircraft: his solution was "Don't kick the desk".
It convinced me that a degree involving the study of structures was a very good choice.
Never ever take on your accountant as a client, the bastard will be able see how much money your making & will always always try to cut you down. I made that mistake once, never again. I have an understanding with my current accountant that should he need to change his IT Support company I will gladly help him find a replacement, but it won't be me. If he decides he wants me to do it then I will have to find a new accountant.
Back on Topic, many many moons ago working in a PC repair shop I was tasked with getting the data off a failed drive. I got to the point of sending it to a specialist as the customer was happy to spend the £600 or so quoted, filing out the form it asked if the failure was mechanical or electrical, so powering up one last time I heard a clunk clunk, to this day I don't know why but I gave it one almighty whack on the workbench, it burst into life just long enough to ghost it to another drive.
One happy customer and we made an extra £600 that day :-)
Do you remember the days of using QIC tapes as backups???
One day, after a catastrophic hard drive crash, I had to rebuild a system from scratch. First load the operating system on a new drive, and then restore the system with the most recent backup.
I put in the O/S install tape, did a re-tension, and the fucking band snapped. Stopped dead in the water because of a 1/8 inch wide tension band.
Luckily for me, another company locally used the same hardware system, and I was able to use their install tape to load the O/S. Saved my ass!
After that, and making sure that manglement understood where I was 'coming from', we started purging all QIC tapes older than 3 years from routine backup use.
In an early project we were working in a cold portacabin which had removable (Bernoulli) drives which had to be taken out and stored in the locked filing cabinets overnight. The only heating, a portable heater had to be turned off. During a very cold winter spell the drives got so cold that they wouldn't boot until they had defrosted slightly. As the junior my job was to be in first, turn the heater and PCs on so that by the time everyone else rolled in they might be in a position to start work.
Back 30+ years ago when I worked for a small software house & PC dealer we made a rule that any new systems in winter had to sit in the office for a minimum of 24 hours before being set up. Typically they'd arrive from the carrier freezing cold, would format and install OK but then the HDs would fail the next day requiring them to be reformatted.
Just for contrast - I was starting out as a computer support tech in the mid-90s (Windows for Workstations 3.11 running on MS-Dos 5.x). Went to a support call in our HR department, this 'critical' PC wasn't working. After talking to the end user, found out that it wasn't really that critical, and that in fact they just wanted it taken away.
So, unplug it, take it back to the shop to see what it really is (they didn't know, and it was a typical 1990s beige box). First of all, find out it's a 286 that someone assembled at home (as far as the other staff was concerned, anyway...) Secondly - there was this mysterious power cabled device in one of the bays, that was way too small to be a disk. It turns out that it was a small heater, specifically designed to sit in a 5.25" disk bay, to make sure that a hard drive was warm enough to start.
The saddest part in my opinion? The hard disk it was supposed to warm was gone by that point! So they effectively had a PC acting as a heater plugged in under this desk for at least 4-5 years... I really wish now that I'd kept that device just to show it off.
Had an old XT that stopped booting at school. Same thing; a frozen drive. So while everyone was running around with their hair on fire (Lots of homework on it apparently.), I picked the drive up, it was disconnected, and whacked it on it's side against the table. Everyone froze. The blood in everyone's face left for other areas. Everyone's jaw dropped. I was amused, after all, there was nothing left to loose. I plugged it back in and turned the computer on, it spun up and booted.
"Now then, back it up and go get a NEW drive." I said and walked away.
stories of hitting stuff to get it moving... please remember the number 1 rule of hitting stuff with a hammer to get it working
never ever tell the crayon eaters what you are doing, how you a doing it and why you are doing it.
Because a light tap to free up a stuck bearing becomes a 3 day job to remove said part after the crayon muncher has twatted the thing with a 5lb lump hammer "because he saw you use a small plastic mallet to give it a light tap"
Icon.... for your expression when you see the wreckage.....
Jeffrey Nonken,
"The "crazy hammer guy" who got your PC running and saved all your data? That's gratitude for ya."
Not been in IT long have you ???
Gratitude is the thing that tends to reduce with rank of person you deal with or difficulty of the fix/job you actually did.
The people at the 'workface' always thank you, but the working through the night or over a weekend tends to be not noticed by many more senior people.
Sad but true !!!
Of course there are exceptions but those are remembered because they *are* exceptions !!!
"Alessandro" here: the boss ended up sending me back anyway having told me to explain said accountant that my daily job was in software engineering, with an occasional side job in system engineering. I also explained that if any other engineer in a 30Km radius would have been sent, the verdict would have been:
Sorry, all of your data is lost. Buy a new computer and enter all data again...
We had an old Sun 'pizza box' workstation at Boeing that died. Just short of its weekly backup. I was walking by as the IT guy was explaining 'stiction' to the poor user, who was not thrilled with having to re-do a week's worth of work. I stepped up and said, "Well, there's nothing to be lost." Picked up the box, flipped on the power switch and gave the whole thing a violent twisting jerk. The disk spun up and I suggested that they run the backup right now.
My dad has an old HVAC blower motor that wouldn't start spinning when power was applied, but give it a little turn by hand and it would spin right up. Useless for an under-the-house motor, but he put a grinding wheel and power cord on it. That was when I was a kid; still running strong.
Our old 8088 and then 8086 used to have a batch file we ran to "park" the heads on the disks prior to shutdown. But we started with a 20MB "hardcard" (and a Hercules b&w graphics card, along with DOS 2.11...)
But later we had a dead hard drive, that we could restart with a pencil in the drive spindle, releasing it. We did back it up then
Some years ago during the original tablet fad we had one of our VPs bring us a MS Surface (Pro 2 or 3 maybe?) that would power up for a second, then shut off. The user's data was of course not backed up and was urgently needed as he was "flying out to a customer in four hours". I tried a little of everything to revive the @#$%! tablet and had no luck. (was never fond of these) Not looking forward to peeling the self-obsoleting glued-together device like a banana to try and extract the hard drive, I discussed the situation with a colleague. We agreed that we'd like to just take a hammer to it, which gave me the idea that with nothing else to lose, as I'd likely have to destroy it anyway, I'd beat the thing on my desk a few times and see what happened.
After detaching the pizza box keyboard, I gave the formica top of my desk a stout whack with the tablet, then another. Nothing. Feeling my frustration seething, I did it a few more times, with increasing force. By this time someone hesitantly asked if I was alright, coworkers were nervously looking towards my desk, things were beginning to fall off my shelves, and the person I'd discussed the problem with was laughing maniacally. Success! The device came on by itself during the last assault and stayed running. I quickly copied the data off it over the LAN. I prepared another tablet for the user with time to spare. Everyone lived happily ever after, and my colleagues had a new respect for me, possibly born of fear.
In the early 1990's a small video-training company I worked for had a 2 1-GB Seagate SCSI drives RAIDed together at RAID 0 for video editing. I know it's RAID 0, but these weren't for permanent data and only for video editing before the content was written to tape.
Anyway. The drives would heat up and sticktion would kick in when the drives were turned off. I told the user not to turn the drives off, but he would and then the next morning it was my job to get the RAID working again.
I would repeatedly power up the drives and turn them off and bang the box on the floor a few times while they were spinning. Eventually, there would be a scrunchy sound and the drives would come to life and everyone was happy until the process was repeated again. The drives were replaced eventually with much bigger 9 GB drives and these new drives never had that problem.
I had an old Compaq laptop given to me, in the mid nineties - I was still at school and used it for learning to program and school work. The hard disk started to fail on startup, you could hear it trying to spin, but couldn't - a wack (from my fist) used to sort it out...for a while. Then it stopped. While I kept some important stuff backed up on floppy disks, I didn't have enough disks for everything. So I proceeded to take the disk apart, and manually move the heads out of park. Believe it or not, this actually worked and continued to work for some months..pop the top off, flick heads out, pop the top back on and slot it back in, by the end I wasn't even putting the screws back in. Oh, and I did this with the power on!
I don't recall the make of hard disk, but the PC was a 286 MOD purchase of indeterminate origin. On discovering that the disk was stuck and refusing to spin up, we discovered that a comb was just the right size to slip into a slot in the front of the disk and jemmy the thing into movement. Worked every time.
As computers became sophisticated, Solid State drives replacing the (un)beloved Hard Drive, and of course, cloud computing, the reliability we get from our devices are very high that we get complacent on having backups.
One thing I will give it to the older systems was that we (users) had the power and control to figure out what went wrong and try to fix them, in today's systems we (and so do the OEMs) have NO idea on how they work, and what went wrong.
A business software developer customer of mine bought an 80 MB QIC24 backup drive for $3000 from me. He tried to sell it to his client who owned an asphalt plant, but the guy thought that his backup floppies were all he needed. So my customer went and demo-ed it to his client anyway, because he had spent some time there developing and there weren't any compatible media to take his work home with. In this case 'demoing' meant a full image backup.
A few days later, the asphalt plant burned down. The heat was to intense that the only recognisable thing left was the Faraday shield around one of the VT100 terminals.
The client proudly went to his fireproof box to get his backups, and found them all melted! Fireproof boxes do not double as Thermos flasks!
Naturally, with the client having $250,000 in receivables to collect from his general contractor clients, my customer made a killing selling him his own data back!
Back in the late 1970s I serviced an aging mainframe system at a large metro newspaper. The combination device for reading and punching cards as well as printing was an even older device. The technology of the "printer" was based on numerous relays. The printer was having startup problems. I pulled out the box that controlled power, it was completely filled with relays. After cleaning the relay contacts it came back to life. I noticed the box had a large black mark on the outside. While I was attempting to clean the black mark the computer operator remarked to not bother, that was where they would give it a swift kick when it got cranky.
So that PC XT must have had either a RLL or MFM drive, and possibly full height.
Due to the weight of these 5mb hard disk drives, a hammer would be the only tool to use for recovery and "interactive" purposes.
Hopefully the hammer technician checked to see if the terminator plug was still attached......
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
