Yes!!!

I often tell people about the time I worked supporting sub POs and one particular office had a Unix box that failed three times. Each time i had sent an engineer out to swap the box out (Unix boxes don't fail via the OS - it is usually hardware - so faffing about onsite wastes valuable engineer and PO time).

After the third swapout I managed to secure authority from management to have the engineer stay on site as the boxes always failed after a only a few hours.

The sub Post Master's wife made a cuppa for the engineer and while he and the PM chatted, she set about festooning the Unix box with fridge magnets which she had removed from the old box prior to the scheduled engineer's visit...

Good to see I'm not the only victim of that!

I also had an encounter with a magnet issue. One of the wife's friends had a son who called me one day about his Mac laptop. It would seem that it wasn't starting any more. I alerted him to the fact that I didn't know all that much about the Mac software environment, but agreed to go take a look.

When I got to his room, I started checking out the laptop and asked him when the issue had started. His answer was to point to a large lump of metal, about the size of my fist, in the form of a cylinder and said that it had fallen on the keyboard. I took one look at the thing, with its nails, paper clips and other assorted "decorations" clinging to it and told him that his disk was wiped and probably dead. He looked at me as if I was telling him the sky was green and asked me why.

That's when I had to explain to him about magnetic storage and how it doesn't do well with ginormous magnets in the immediate vicinity.

So yeah, not knowing about cassette tapes is likely a cause of ignorance of the issue.

Thanks for the massive spoiler in the thumbnail on the front screen el Reg!

At least I made the connection quicker than the two desktop engineers

(although i admit thats a little like watching 'You've been framed' thinking "this'll go wrong" )

You would be surprised at the actual strength of an external field that is needed to upset a hard drive - even back in the days of the noughties and GMR media I cannot see a thousand fridge magnets upsetting things. The field strength on the write heads is enormous, and thought the read heads have a weaker response, there is a drive case shielding it, and a main PC case, and then the distance they are away..... plus the fields from internal PSU's which are not inconsiderable and don't bother the average PC one jot...

Well not to doubt you, but we have at least a few tales here in the comments of exactly this happening, so it does seem like its real.

Notice that none of them speak about damage to the harddrive itself, just the occasional bit here and there being flipped, and causing disk corruption, which certainly seems plausible. Whilst the read and write heads are pretty well protected, the platter itself isnt overly EMC protected...

The platter isn't protected because there is no need - the coercivity of the media is enormous. Also there is a pair of field coil magnets right next to the platter - that are waaaaay more powerful than a fridge magnet if you have ever dismantled a HDD....

I've tried to deliberately corrupt several types of magnetic medium using rare-earth magnets, degaussing coils and the like, and never managed to succeed, even with floppy disks.

For hard disk drives, you not only have the steel case of the computer itself (it needs to be steel or another ferrous metal for the magnets to stick, yes?) but the metal chassis and cover of the drive itself. Even non-ferrous metals tend to disrupt magnetic fields due to induced eddy currents.

The other thing I would like to point out is that if it were single bits being corrupted, the error-correction in the disk controller would easily fix them. If it were multiple bits, more than the ECC could correct, then the read would be retried, and if that failed, the sector would be marked bad, and the error would be fed up the chain to the OS and the requesting program. The chance of a silent read failure is exceptionally low, measured in the one in billions (this is a guess, it is documented for most drive models) of read operations.

Hard disk drives suffer from bit-rot all the time, so the ECC is pretty robust.

It seems incredibly unlikely that the fridge magnets were the cause of the problem, although I'm not going to try to suggest an alternative.

Not to mention the fact that there's a couple of honkingly powerful magnets inside the HDD casing.

Exactly. This is where this story always falls down - the weak fridge magnet outside can totally trash a drive but the rare earth neodymium one inside has no effect at all....

It might be corrupted before getting to the disk. Lots of wires with weak currents flowing through them, lots of magnets all over the place, somewhere a bit gets flipped.

I've seen PCs running with zero problems in places with rather strong magnetic fields. At SLAC, for example.

Yeah I'm suspecting a case of cogito ergo sum. I salvage the impressively strong magnets from inside the HDD whenever I dispose of one (they're terrific for finding screws and therefore studs in a wall for example).

While there are more examples in here of it happening, consider how many unreported examples there are of fridge magnets where it doesn't cause issues, because it's not worth reporting or noticing. Here's mine - for several years, I had a mini network switch with magnet feet clinging to the side of my case without data corruption. Here's my other one - all the fan dust filters on my current machine are held in place by virtue of being giant fridge magnets.

I also doubt that this user is unique. Fridge magnets have been put users' PCs in an office environment since I can remember.

Not at all as bad as losing a drive. At work I use MacOS and Windows daily and on separate hardware. A few laptop generations back, I had run out of space at my desk, so, when I was done working on the Dell laptop, closed the lid, plopped the MacBook on top to continue work there. Flip open the lid, blank screen. What? Close lid, reopen. No change. It was working seconds ago! Must be the display brightness...

Jab middle finger on to power button to force reboot. Boots up, Mac happy sounds pling from the speakers. Blank screen. Fine. Grab the MacBook and trundle up to the IT walk-up support station to whinge about crappy logic boards build quality etc. Flip open screen... ta-da! All hunky dory.

Back at the desk, MacBook back on top of the Dell, flip screen... blank!

Yeah... turns out I am PEBCAK idiot. Both laptops were of similar design (same Foxconn factory) and both utilised a magnet for their keep screens closed feature. Magnets that are good enough to trigger the sensor on the top laptop that despite screen is physically in open position, the sensor indicator still indicates closed, prohibiting the screen from displaying anything. Lesson learned.

I have a classic 160GB iPod with a spinning drive. It's old but I was using it in the car back when I did regular commutes. I had a dashboard magnet, designed for holding a phone, and it would have been perfect for holding the iPod. I knew it probably wasn't going to like it, but I couldn't resist the temptation to try. Sure enough, it made a horrible noise, and the screen glitched out. For a few moments, I thought it was totally buggered, but it eventually sprang back to life with seemingly no permanent damage.

Suspect that it was the read heads being physically moved by the magnet. Also suspect it was a much stronger magnet than one designed to only hold the weight of the magnet.

I've managed to corrupt a classic iPod (to accelerate a warranty claim that was intermittent), but relied on it being a really strong neodymium placed directly on the iPod case (itself only 1-2mm away from the platters and read heads) and left to play for 12+ hours before it failed.

There's a lot to be said for making PC casings out of plastic or aluminium.

Just a thought - are fridge magnets all or predominantly magnetised with the same orientation? If they're random or better still each ha two poles with opposite orientation (like a horseshoe magnet) then the net field on the other side of the cover should be low to nil.

Just checked a dozen or so that I have here in the office ... They are oriented such that either the N or the S is towards the "fridge". It seems to be truly at random.

Many moons ago, I witnessed a field engineer open the back of a piece of equipment, pull the diagnostic floppy (8", just to date myself) off the inside of the door where it was affixed with a magnet ... and the fucking thing still worked! Observing my surprise, he just shrugged and said "I know. I don't get it either. They did it this way for years before I got here. I don't ask questions, I just go by their playbook and collect my pay." He claimed to have seen several tens of these things, and the disk was only dead once ... and that was caused by a couple of rather obvious staple holes.

Magnetic media is a lot tougher than most people think. I would be absolutely astonished if something as weak as a fridge magnet could affect a hard drive. Even several tens of fridge magnets. I'd be even more shocked if a fridge magnet or magnets attached to a desktop PC's case could affect data in transit between CPU and drive media ... especially if the magnets were immobile.

With that said, I have heard this story many times over the years. But it has always been a FOAF telling it, not the person it supposedly happened to. And for the record, we tried on numerous occasions to cause the so-called corruption, with weak little fridge magnets all the way up to rare earth magnets. We observed no data corruption at all.

If anyone can tell me how to replicate these results I'll be more than happy to believe it is possible, but until then this is in Urban Legend territory.

> And for the record, we tried on numerous occasions to cause the so-called corruption

I sometimes dismantle old, dead, HDDs for the magnets. I have several of them stuck on the side of my PC as I write this. They have never caused any pr0q2w38iv dzzzzz that I have seen.

The story went around in the 80's that trravelling on the tube i London would corrupt any magnetic media you were carrying - apparently the fields from the motors were rather strong. It was actually not tue, but we could use this excuse to travel by taxi, rather than tube, when working in the capital and get paid our expenses fr it.

The magnetic fields in a railway traction motor are intense, but they are also confined to specific closed paths within the motor. Otherwise the motor would be less efficient than it should be. However, underneath each motor coach of an electric multiple unit is control gear for the motors. Resistances and contactors in older units, operated by electromagnets, and making/breaking quite large currents (a thousand of amps or so at 600v - 750v). Arcy/sparky. This could have an interesting effect on electronic equipment a metre or so above, especially if the train car has/had a wooden floor, as many did. In 1963 aged 11 I went to Headquarter and General Supplies in Croydon and bought a battery-operated tape recorder for £4.19.6 and on the way home by train, tried it out. When I played back the recordings I made, I could hear track sounds, the motors starting after each station stop, etc, but also kind of hissy clicks, (like distant lightning on an AM radio) which I decided must be the EM interference from the control gear switching out the starting resistances for the DC motors. For any rail buffs, I was in a 1946-1951 built BR Southern Region 4-SUB multiple unit, but all electric trains up to around the 1980s used this technology.

I must admit it sounds more like an urban legend to me, too, but being an open-minded commentard I've got to ,,, well ... make a comment...

The disks in the story did not fail, did they? Rather, just some files were mangled (Word files - probably just because that's all the user was doing). Maybe it takes a lot less to flip an occasional bit, and possibly only when things change (so that doc contents get corrupted while Word and system libraries and such keep working). The SLAC comment was interesting and sounded relevant indeed, but then maybe even a strong uniform background component (let's not forget the Earth's field, either) would not have the same effect.

In short, a proper experiment is in order. No, I am not volunteering.

[*] Yes, I know, I suspect I might have mixed a TV show and a movie.

It probably happened, multiple failed disks isn't unheard of. Or perhaps the third engineer also replaces the IDE/SATA cable as standard where the first two engineers doesn't.

Some word files mangled, but (say) pagefile.sys carried on regardless? If every million bits was getting flipped, I'd expect to see random BSODs before I saw a bunch of .doc files corrupting.

"I found it a trifle worrying that two 'engineers' hadn't made the mental connection between magnets and magnetic media."

Possibly because they weren't engineers, but actually technicians? Or even, fitters.

But what if the fridge magnets slightly moved the recording tracks so they were off-centre?

ISTR a Dilbert cartoon where his boss was wearing a magnetic bracelet which then buggered up his computer. Maybe some astute Reg person can find it, because I buggering can't.