Gnarly Error Messages

Veeru writes "In my career, I have run across some whopper error messages, but a call from the mainframe sysop one night beat them all: 'We are experiencing MVS processor spin loops, the programs are running while holding a disabled CPU. This is causing XCF communication delays to the point where we are losing VTAM RTP routing, are suffering OSPF adjacency failures on TCP/IP dynamic routing and MIM VCF failures. Whatever this code is, it should NOT be propagated to production or we run the risk of losing the development plex if XCF signaling is adversely impacted by processor disabled spin loops'. My friend once got an error message 'Error 2 while trying to report error 2'. I would be curious to hear from the Slashdot community on encounters with other bizarre error messages."

Comments

Illegal Operation

[stephenisu]

Working in Technical support for a government website frequented by technophobes with college aged children, I can't count the times I have had people scared to death because their computer had encountered an illegal operation. One woman started yelling at her kids for putting that &#*!ing nappy (napster I am guessing) thing on their machine. It took me 15 minutes to explain the situation to her.. after the 10 minutes of telling her to calm down.. at least she wasn't one of the criers.

Apple's MPW C compiler famous for its error msgs

[Cerlyn]

Apple once put out a C compiler famous for its error messages. Who else would make a compiler that states "This label is the target of a goto from outside of the block containing this label AND this block has an automatic variable with an initializer AND your window wasn't wide enough to read this whole error message"?

Searching for Apple compiler error messages on Google picks up dozens of sites with the error messages from this compiler, as well as spreads out the slashdot effect.

Doing a search for Eudora humor error messages on Google shows Eudora to have a similar sense of humor as well ("Memory is tight-Live Dangerously").

SGI message - has anybody else had this?

[fantomas]

First time my boss went away and left me in charge of everything, our baby, the SGI Indigo2 ( this was a few years ago) decided to die big style. I am not a full blooded geek so scuse me if I don't describe this right, but...

...screen filled with text, went up the screen rapidly filling the whole thing, I think it was like when you start up and all the boot stuff goes past. Finally the screen flashes then does a sort of blue screen of death and the only text on the screen in the top left is DON'T PANIC.

I swear I saw this, if I hadn't seen this with my own eyes, I wouldn't believe it, but there I am, the boss is away for the first time on holiday and the computer is saying 'DON'T PANIC' . I knew things were very, very bad.

Can somebody tell me about this error message, how SGI got to put it on their machines, and why?

(end note is boss was cool as ever and the engineers fixed it and we got our data back, but boy, was I afraid to touch that machine again...)

Errors covering errors

[Ektanoor]

Many years ago, one of my colleagues fell into a weird situation. He was quite good in Assembler and wrote some quite long program. When he finished, he said that he doubts that the program could work. "I should have done some checks before finishing it..." He compiles the program, gets ready for some long debugging and... the program works... He stares at the screen.
"Something is wrong here..."
"What?" I ask.
"The program works...".
"Well it should doesn't it?".
"No, it shouldn't, no one can write Assembler in such volume and avoid errors..."
"But does the program give the right result?"
"Yes, but that's impossible! I nearly guessed how to do it. How can it work?.."

So he starts checking the program. Finds nothing. Debugs it, all seems to work. Then he starts to doubt that the results are correct. So he makes two three checks by hand. Then he writes a small segment of the program and things go nuts.He gets back to the whole program and starts debugging it, step by step. In the end, and after taking four times more what took him to create the program, he approaches me with some clear relief.

"There were errors..."
"So the result was wrong..."
"No, the result was absolutely right!"
"!?!"
"Well, the fact is that I did one offset wrong but in other section of the program, another error in made returned the values to normal. That's why the program worked fine..."

How many such programs exist?

Values of beta will give rise to dom!

[BitwizeGHC]

This story from Dennis Ritchie tells of an error message in old versions of Unix that was actually sort of a Bell Labs version of "All your base".

From personal experience, one that sticks out in my mind is from Microsoft's Flight Simulator. If you auger into the ground, it says "Crash". If you bellyflop into Lake Michigan it says "Splash". But if you make a perfect landing, forgetting the minor detail of putting down your landing gear, it'd say "Crash! Lower your gear next time!" This message dates all the way back to MFS 1.0.

Re:Amiga Error

[ewhac]

Everyone likes to malign the Amiga system crash dialog, simply because it bore the term 'Guru Meditation'. "Ha ha," they joke, "see how primitive and useless the error message was."

You have to understand that this was a massive advance forward. Prior to that, the major systems were first-generation Macs (which displayed a certain number of bomb icons and nothing else); and Apple ]['s, Commodore-64s, and MS-DOS-running PC clones -- all of which displayed nothing; it just (if you were lucky) silently locked up.

Carl Sassenrath, designer and author of the Amiga's 'kernel', thought this state of affairs sucked, so he did something about it. Amiga's Guru Meditations, cryptic though they were, told the programmer which task was responsible for the crash (first hex number), and what exception it generated (second hex number). You could then hit the right mouse button to drop into a very primitive serial debugger to get more information. While these numbers were useless to 95% of the users out there, it was information the user could give to the vendor, helping them track down the problem more easily -- information they never had before.

Meanwhile, everyone just happily tolerated Windoze BSODs, even though they were, and still are, no more informative than Amiga Guru Meditations.

Schwab