BOFH: When the Sun rises in the West and sets in the East, only then will the UPS cease to supply uninterrupted voltage BOFH logo telephone with devil's horns "Eeeeeeasy does it …" I say to the PFY, "almost there …" "What're you guys doing?" Richard asks, blundering into the Server Room without permission. That's Richard — not Rich, not Richey, and definitely not Dick — a new consultant, engaged to make the company more "agile". Dick is 26 …
I was told by an old friend of mine who used to work in manufacturing/milling ...
They had a lot of grinding/drilling equipment.
He was on the shop floor drilling as usual, with the "old reliable" (set to "3") when the CEO decided to tour with his young whippersnappers.
One keen eye'd kid saw the machine set to "3" and "instantly realised that if it ran at 10 that the whole system would be more productive and efficiency could be improved".
He informed the CEO, and berated the guy operating the machine about how inefficient he was being
My friend stepped in to replace the guy operating the machine. My friend tried to explain why it was only set to "3", but kept being cut short. The CEO stepped in, and the machine was indeed turned up to "10"
My friend (wearing all the PPE he could) started to use the machine as it destroyed itself in about 3 seconds, with metal bits flying all over the place and drills embedded in the wrong places.
Apparently old machines don't like to be run at their maximum, and the new kid cost the company a couple of months on that machine, and potentially a lawsuit, if my friend hadn't known what would have happened and made sure it was him operating it ...
the red faces were almost worth it ...
You can only hope they "live and learn"
My Dad used to work on kit like that. Designed so that 80% of capacity was ideal and would run without much maintenance at that level for months at a time without a stop.
So of course manglement wanted it to run at 100%(+) which resulted in different parts of the line not quite being able to keep in step and thus breaking after a couple of days. Usually taking a day to fix. Once a week....
You do the maths on the loss over unplanned maintenance vs planned maintenance every few months.
But noooo... "it's designed for 100% so 100% it shall be"
It's not even that they're designed to run at 80% (or whatever) of capacity, in a lot of cases. Different materials require tools to run at different speeds. Different tool heads require different power/speed combos, etc.
Just for example, running a drill too fast for the material generates excess heat, which makes the drill and workpiece expand, the drill seizes and/or welds itself to the workpiece, and things break spectacularly because the workpiece is firmly fixed to the bed of the machine, which is now trying to twist itself into a pretzel.
Could do with one of those now. Oh and bits that are less brittle than glass. And a cross head screw driver bit that actually fits into the screw I'm using without shredding itself and the screw when the screw is just proud enough to be a pain but impossible to get underneath the head without damage.
It's not even that they're designed to run at 80% (or whatever) of capacity, in a lot of cases. Different materials require tools to run at different speeds. Different tool heads require different power/speed combos, etc.
Exactly this.
A friend saved a mid-sized manufacturing firm ~£500/mo in cutting discs when he twigged that the machine was originally designed for a disc of a different composition and speed rating which had stopped being made. Somewhere along the line the current discs had been substituted and the speeds merely twiddled until they got a clean cut. They were running ~3x their proper speed.
Slowing them down involved tweaking the power supplies but lengthened tool life, saving both a decent chunk of money and improving consistency in the product.
@Steve Todd: One of the reasons Du Pont was so successful in producing nylon on an industrial scale, was that E. I. de Pont de Nemours was originally in the black powder and dynamite business and so knew how to handle potentially explosive reactions.
some cost accountant needs to factor in the COST OF INCOMPETENT MICRO-MANAGEMENT ON OVERALL PERFORMANCE when upper management tries to determine what speed settings to use on the equipment.
That vs the cost of an industrial "accident" that prevents such micromanagement from being implemented... [boss, the window is over here - check out the view!]
As far as i have seen, learning is not something that happens to manglement. Or as my experienced staff seargant said "its not the dumb officers that get you killed, its those who believe to be infallible."
The oldest and most questionable piece of hardware i saw was a I286 with a veritable ST225 20 MB Harddrive. As if Seagate hardware wasn´t bad enough, that data turntable was mounted top down.
Nobody dared turn it off, it provided the CNC programs for a FANUC (whatevertype) CNC Mill of quite antique design through a GWBASIC kind of... terminal program... (OPEN... FOR OUTPUT AS#1, that kind of stuff).
When the Hardware finally died and was replaced with some pentium Win98ish box, the "terminal program" would no longer connect to the FANUC Mill. Adding a few adaptive delay loops to the data transfer subprogram fixed that. As vile as that hard- and software heap might sound, it was reliable once running, much unlike its new and shiny counterpart.
The new and shiny CNC had such bad firmware that you had to declare drills as milling bits because once a single drill was declared, most of the more complex milling functions simply refused to start anymore.
Several visits from the company serviceman did not fix this. Poor guy knew exactly thet his visits were futile, as he could only do his checklist and had no access to even update singular software modules.
"2nd LT with a map and compass."
Any officer will do, not just a 2nd Lt. My dad got to Colonel and could lose the family on a walk at will using a map. Mum was merciless, which was hilarious because she left the WRAC as a Captain. On the other hand she had eyesight twice as good as normal and really good spatial awareness. Dad wore bottle ends. Good team.
the most dangerous thing in the world is a 2nd LT with a map and compass
The night I was commissioned a 2LT, when I got the traditional first salute from the RSM I got the advice that at that point I was qualified to lead 33 men to their deaths, and to listen to advice from my Sergeant.
It seems that removing part of brain occurred with removing the Sergeants stripes I had worn for 4 years and had been an OR (EM for the leftpondians) for 4 years before that.
Only happened a few times that I can recall, but I would always get in writing when the person wished me to do <this thing> and it would be recorded that I suggested not to do said <thing>.
Occasionally person would back down when it was obvious whose name was going to appear in print but at least twice the result was not good.
Places don't like when data can't be retrieved from now non-existent backups that I suggested wasn't a good idea to get rid of despite aaaaaaall those cost savings you said would occur.
"Live and learn" - we've heard of it.
Of a carefully curated room of old equipment required to rebuild old software and hardware for certifiable systems - think aerospace, nuclear, various transport, medical, etc. with up to 30 year support contracts. And the nitrogen store for the really delicate stuff and old components.
New owners of company replaced it all with new management offices over a long weekend. They had been careful to get in the shredders who could take a whole PC at a time!
Quite true but even if there is no survivor, this ensures that the culprit wont do that misake ever again.*
Sadly, i have seen too few individuals who seem to be able to learn from the first mistake, let alone a really grave one.
*Terms and conditions like kingdom come, zombie apocalypse and other resurrection based problems apply.
To be fair.. They gave the young lad every possible bit of warning Not To Meddle without actually breaking out the CattleProd™ itself...
Clearly Darwin Award material, that lad...
You start wondering who else he had already ticked off to rate the .....suggestion... he should have a look at the efficiency of the IT department...
I think part of the problem is that youth(s) never seem to check that they're properly grounded, before grasping lots of things.
Nothing delivers 10kV of Experience™ quite as effectively as a "Here, hold this!" trustingly complied with. (How else are they gonna learn? The ones that survive, I mean.)
I remember working at a place with a server where nobody knew what it did, except we all knew its fan made a lot of noise.
Two years after I started, the decision was taken to turn it off and deal with the consequences. We turned it off. The room went silent. And for the first time I realised that my Mac wasn't actually 100% quiet, it's just that I had never heard its fan over the noise from the server.
We never heard of anything failing without that server.
We had a server that belonged to a customer which had a bad fan for years. It was so loud you could hear it from the other side of the room, but this wasn't a small server room. This was a medium size hall in a colo center, filled with other people's kit, and yet, as soon as you opened the door, the loudest noise was the tiny abused HP fan in a rack on the far side of the room, buzzing away at a frequency that ground into your ear drums.
I can only imagine how happy the other customers at that colo centre were when we finally managed to decommission that machine.
Every server my old company built used cheap fans, they all went noisy or seized within a year. I saved a lot of time by using a syringe with a secret mix of (herbs and spices)* injected injected thru the rear seal while still running. I learned to check them even if no one complained and I was on site for other reasons. * 50/50 wd40 and SAE30 non detergent oil
I usually just use 3-in-1 oil on all my fans large or small (probably similar to your formula).
You can often get PC fans that seem to be completly dead going again with drop or two and a few manual spins.
The big one circulating air in the room goes in the shower every few years for a clean an then gets oiled and put back on duty when it is somewhat dried out (I like to live dangerously, plus it dries faster that way).
Sometimes I only remember to re-lube it when it starts emitting a high pitched squeal.
It has been running for about 15 years continuously now other than cleanings; not bad for an el cheapo fan from a big box store.
My tried-and-true, decades-of-results-to-back-it-up method of cleaning nearly anything electrically inert (either by its nature, or temporarily rendered that way) is to shower with it.
Fans, both environmental and device-extracted? Check. Clogged vacuum cleaner hoses? Checkity-check. Window blinds, window screens, aircon filters, sink mats? Check, check, check, aaaand... check. Keyboards? 1000% check, any other method is just farting around and wasting time. (It helps to have a spare on hand you can swap it out with, for the drying period.)
They all get a nice, soapy shower (and a good scrubbing and exfoliation from Mr. Retired Old Toothbrush), then they get a perch in front of the box fan on highest-tolerable speed, positioned at an optimal drying angle for ~12-36 hours (depending). ...Unless what I was cleaning was the box fan, then I too like to live dangerously.
My shower drain's always a little bit gross, but the rest of the place is pristine!
I'm just reminded of how damned glad I am I recently installed my new NAS and (eventually) powered it through a UPS.
I got caught out a couple of years ago when a power outage trashed my PC and six HDDs. So when I built a new one I got a small UPS to protect it in future.
You'd think I'd have learned, but when I bought the NAS and set it up (sans UPS at the time), several days later I accidentally pulled its plug and the interface regaled me with dire warnings of how that could damage the disks/data. So I bought a better UPS to power things on that side of the room (one that the NAS could talk to).
Then a couple of days ago there was a 'thunk' and we lost power for about 5 seconds due to a problem with Western Power. The NAS duly reported it was switching to battery, then that it had switched back to AC. And the PC didn't shutdown, either. So at least they've been hot-tested now.
All I've got to do is find put where I plugged the damned router in somewhere under the desk and junk, because that did go down.
I remember worked at a place with an ancient SUN server that nobody knew what was. One day a mail from Corporate IT went out that the server would be decommissioned in one week.
Very shortly after the network flatlines with 'rsync' and 'torrent' traffic: It was the corporate off-colour backup server, that is, the global repository for Warez, Pr0n, ripped dvd's, ....)
"The year Richard was born I set a company record for how fast a VAX 11/780 would roll down the road from our building. I could've got a world record if it hadn't been for that cyclist."
I did my MSc thesis work on a VAX 11/780. Big hulking beast, and everyone complained their vi or emacs editors slowed to a crawl whenever I was running thermal simulation software on it. I actually found my shiny new 80386 at 25 MHz with a Cyrix 387 co-processor and a whole 8 MB of RAM outpaced the big hulking beast by quite a margin.
Lovely episode. It was inevitable Dick was going to go for the practical exam
I actually found my shiny new 80386 at 25 MHz with a Cyrix 387 co-processor and a whole 8 MB of RAM outpaced the big hulking beast by quite a margin.
I bet that's not true.
I'm sure your pooter got the simulations done in less time. But when it came to rolling down a hill, my money's still on the lumbering old VAX.
Also copes better with the inevitable interactions with cyclists...
As a fellow anonymous South African coward, I'm surprised you didn't note the above remarks about the virtues of running vital equipment at 100+% capacity indefinitely and what that has done to our national power grid.
So at the pub have one for me, provided loadshedding or unscheduled power failures don't get in the way...
One of my first dreadful Amiga memories was implementing a bigger power source for all those interesting peripherals. All went fine until the 5 volt line regulator in the power supply went haywire and delivered 12 VDC. Apart from the successive little bangs of overloaded capacitors, the only victim was a literally exploded Hitachi HA13445 IC in the floppy drive, responsible to create the pseudo 3phase for the floppy drive motor. The Amiga mainboard still ran without problems, even without replacing the burnt out capacitors. Yes, i still remember the IC, you do not forget your first piece of silicon with a literal crater down to the chip surface in it.
The commercial slideshow for (state interfered) cable TV in germany also ran over a decade on some Amiga, proven by the occasional guru meditation that replaced the advert slides every two years or so.
Cant vouch for the later "Amiga Inc." hardware as too many components were left out, especially in the latest A1200 series but the old A500/2000 Hardware sure was reliable and survived many "electronic tinker mode attacks" of my freinds and mine.
Ah, the Amiga power supply! Happy memories. One of ours died, so after making sure that a PC power supply would supply the right set of voltages (if wired up correctly), I nipped down to Computer Services at work and "rescued" a condemned (because of old age) PC from their stonehenge of machines that they kept meaning to get rid of...
Once I got it home, I discovered that this predated modular power supplies and the gubbins was just plonked in the case as a set of components that needed the whole case to work... It was the weekend, the kids wanted their Amiga back and so I just shrugged and persevered. The wiring went beautifully, but I had to tie an old PC desktop case (with nothing in it except power supply) to the back of the desk with a bit of plastic clothesline I had lying about.
Probably mentioned this before, but slightly related. Built a home designed guitar amp intended to run off a 20V DC supply (i.e. laptop power supply to avoid messing with mains), push-pull design (to drag out the most power possible think I was aiming for 20-30W), this normally needs AC, instead used a quartet of beefy audio transistors, and a little 2N5550 transistor balancing ground between the two halves.
Simulation says this transistor takes almost no current, just there to set the bias right. I say little, but they're supposed to be 0.6W. Worked with a test load (not going to blow up a real speaker). However, something about the design was not quite stable, so on the first real test it worked for a couple of seconds, then output stopped, there was a brief pause and then a very loud bang. Half the transistor casing had blasted off as it went from 0.5V to 20V and lost its short but valiant struggle. Changed a couple of resistors and tentatively tried a new transistor... the second one survived. It's been rewarded by sitting around for several years waiting for a cabinet.
printing out the new cockroach coat of arms (2 crossed dead keyboards on a mound of dead keyboards), the BOFH has given me another good idea for making sure the irritating production 'engineer' has an accident.
Just need to google up "Earthed anti-static mats" and know a suitable sparky who can be bribed into doing less than stellar work behind the filing cabinet in said engineer's office.
Followed by a monday morning request "Could you just get that drawing out"......
If some annoying individual like "Richard" keeps pestering the whole company, it might be wiser to allow for some work accident plausible deniability than simply letting the issue vanish.
Much more if those responsible for the timely demise are well know to have a special kind of... tolerance.. for individuals with management asperger syndrome.
Once worked with a guy who was keen to project a quintessentially "British country gent" image: Summer attire often featured cricket white jumper, flannels etc., tea in a cup, on a saucer - etc., etc. His name was Charles but he preferred - much preferred and with an irritating insistence - to be called "Chas".
I wondered how much we could get his name to rebound. Cruel, but entertaining. After a very little deliberation, we baptised him "Chuck" - both in conversation and to his face.
Oooooh, and you could hear his teeth grind at every mention.
The 'frame room' lacked documentation. No idea which pair went where. Employed some guys to trace and document what they could. Disconnected the rest and awaited the outcome. Personal FAX here. Personal FAX there. And the Director's personal phone line so the offspring could demand a pizza was collected on the way home.
Don't start me on faxes.
We bought a fax server and ran it 5 years. For that it was used occasional for people to send mortgage forms off and to receive car lease spam.
But as it was on the headed paper, we MUST keep it.
At least they stopped paying the maintenance fees after 3 years.
Old stuff will keep running even if you need a Virtual Machine to keep it running.
My local library, and by local I mean is three miles away on a direct line, crossing a river, runs their very very very old MADE IN DOS 2.0 BASIC "program to keep track of books" in a version of Dosbox..Of course since Dosbox doesn't include printer support, the old library cards and so on have to be FILLED BY HAND and they don't change a thing because Printers are a scam and the ink is terribly expensive.
Of course any machine over twenty years old will have a few hardware problems, hard disks can be replaced by scavenging parts after an "upgrade" of other stuff but if the bios or the video card dies and a battery replacement/part replacement is not possible the machine is as good as dead.
Are you sure? Was it an APC? Because you only have to unwrap a sealed lead-acid within sight of an APC logo for it to boil off all the electrolyte.
For the curious: APC float voltage calibration
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
