APLawrence.com -  Resources for Unix and Linux Systems, Bloggers and the self-employed

Troubleshooting: Try something different

newspaper. They had a mixed network of Unix machines, a Novell server, and quite a few Macs. I worked mostly on the Unix side, but one morning I was asked to run down to Editorial because there was some major problem. The person doing the asking was the President, so I suppose he wasn't really asking. He looked upset.

Newspapers are very concerned with deadlines. Small newspapers are even more concerned than the big ones - there's no "Stop the presses" for them, because they are usually renting press time from someone else. They've bought the use of that press for a certain time on a certain day, and the copy had better be ready to print when those presses are ready to roll. Consequently, any computer problems that interfere with the production of content are taken very seriously. VERY seriously. A problem in Editorial is much more important than a problem in Accounting. Even Sales would be far less important. The paper must get out, every week, on time. Everything else depends upon that happening.

I went down to the Editorial department on the first floor. This was a large room of curving desks and a lot of networked Macs. Two or three Mac servers were on a shelf at the opposite end of the room from where I stood. People sat at their desks and most of them had pained or annoyed expressions on their faces. They weren't typing at their keyboards, which I knew couldn't be good, because Deadline was only half a day away. Over by the servers were three very unhappy looking people in suits. One of them was the resident Mac guru, but I didn't recognize the other two. I asked the person at the desk closest to me what was going on.

"I don't know. All our machines keep crashing" He indicated his screen, which indeed showed a Mac trying valiantly to boot but failing miserably, only to try again, and again.. I didn't know much about Macs then, but obviously this wasn't good.

I looked over at the three people standing near the servers. They were talking to each other, and they obviously looked worried. No doubt the President had already expressed his opinion directly to them. He was well known for expressing opinions colorfully and pointedly. He didn't like problems, and especially not deadline threatening problems. I was vaguely thinking about that when I suddenly found myself pointing at the leftmost of the server boxes and shouting "Shut that server down"

The three suits turned toward me with obvious surprise. Why was this sloppily dressed Unix guy telling them what to do with their server? Honestly, I had no idea why either. But I was still shouting as I moved toward them. "Shut it down. Yes, THAT one. Shut it off. Shut it off NOW!".

I must have sounded insane, and perhaps dangerous. One of the suits reached up and flipped the switch. The server shut down. There was a very pregnant pause as we all looked at each other expectantly. OK, the crazy guy wanted us to shut off the server. We did that. What's next, brandishing of weapons? Who IS this madman, anyway? Then happy sounds started coming from the desks:

"I'm up!"
"Me too!"
"All right, let's get to work!"

That last from the Editorial manager, who had probably been off in a corner sobbing softly. Yes, the Mac network was back up.

And I, of course, am a genius. My deep and thorough grounding in Mac network protocols had enabled me to immediately see the problem that the darn fools in suits had been struggling with for an hour or more. Yes, I am that good! You only WISH you were this good. Big touchdown dance, let's hoist me up on grateful shoulders and have a little victory parade. I am GOOD!

I'm also full of horse manure. To this day, I have no idea why that server was causing everything to reboot. I have no idea why I yelled at the three Mac experts (as it turned out, all three were Mac consultants) to shut that server down. Maybe I subconsciously knew that the server was new; as it turned out, it had been added to the network just a few days earlier. Maybe there was some sub-sonic sound it was making that some part of my brain picked up on. Maybe I just don't like people in suits working on computers. I really don't know. It was just intuition, and I'm sure it just as easily could have turned out with me being that crazy Unix guy who made them shut off a perfectly good server for no reason.

But intuition has fixed problems for me other times, too. I'm not going to get all Zen on you here, but your brain is full of stuff you don't even know that you know. Maybe I had overheard someone mention that there was a new Mac server in Editorial some other day. Maybe someone else had said it sounded louder than the other servers, or that his machine was a little "funny" since that new server was installed. If so, I had forgotten entirely, but maybe some lower part of my brain had not. That kind of thing has happened to me more than once, so when I get a hunch or a "feeling", I pay attention.

Of course you have to realize who or what you are paying attention to. The lower parts of your brain may have enviable retention skills, but they are called "lower" for a reason.

A Dream

I am taking a test. It is an important test, and I and all the other students know this. There is an air of nervous apprehension as we open the first page and begin. Ahh, this isn't so bad: 7 + 5 + 4. An easy one, that's 14. The next one is more difficult: 9 * 12. I have a moments hesitation as I calculate. There, done. My pencil writes "81" and I look to the next question. It's a logic question. Good, I like logic.

"If a cumulous cloud has a radius of 14, and the diametric is opposed to the vertical, what is the arc radian?"

I see that some of my fellow students have also reached this point. They are puzzled, some showing early signs of panic. Not such an easy test, is it? But I turn my razor sharp mind to the problem. Aha! It is opposed diametrically! I smile and write down the only possible answer: two bananas.

Looking toward the front of the room, I see our beautiful instructor. Her silken hair falls to bare shoulders. When I finish this test, I will approach her and suggest some "extra credit" work. She will melt in my arms.. but just then, the door bursts open. A bare chested man rushes into the room waving a sword, definitely bent on mayhem. The other students cringe and cower together. I stand firm, fixing him with a steely gaze. My lip curls in scorn, and with a banshee yell, I launch myself upon him, knocking him to the floor. More blood-curdling roars rush from my throat as my fists pound him into submission. I smash him again, and..

"Wake up". My wife is poking me. "Wake UP!". She's annoyed that my screaming has once again pulled her from more pleasant dreams. "What were you killing this time?". She's very familiar with my dreams. I mumble something and we fall back to sleep.

Dreams. The window to the unconscious mind. Maybe. I think dreams are just our subdued mind's way of making sense of bits of data that drift by as other parts of our brains process and store the days lessons or even just fire off random neurons because there is nothing else to do. Assigning meaning to incomplete information is something we do very well; it's how we get through every day. If there is any meaning to be had from dreams, I think it is only in looking at what our minds conjure up and how we react. From the dream I related above, and other similar nighttime dramas, I know my subconscious mind is aggressive, confident, reactive, and no more intelligent than most squirrels. It always thinks it knows the answer, always knows the best course of action, and isn't hampered at all by doubt or fear. It would surely serve me very well if I were a prehistoric hunter rather than a modern computer consultant. If it had full rein in our modern world, I'd more likely be in jail, or at least get my face slapped regularly.

On the other hand, brash confidence isn't always a bad personality trait. I'm sure part of my success in life comes from that base attitude which confidently assures me that I will make the right decision. But I'm also aware that the same under-brain is pretty sadly equipped when it comes to more complicated thinking. I am therefore not one who gets all starry eyed over the hidden cognitive powers of my unconscious mind. Sometimes it has a good idea, and sometimes its ideas are incredibly dumb. It's up to me to apply rational thought to whatever it is urging me to do.

But if whatever it suggests isn't dangerous or apt to dig me into a deeper hole, doesn't involve dynamite, strangling the guy over in the corner, or jumping up on the table and tearing off all my clothes, I might try it. Especially if I have no other useful idea. Doing something, no matter how useless it may seem, can help us move toward a solution. I'm not suggesting that a printer that won't print is going to get fixed by stepping outside and washing your car, but that's probably not what your subconscious is suggesting. Or maybe it is, but what it really means is that it thinks the printer won't print because something is dirty and needs cleaning. Maybe it's right, maybe that is the problem. Or maybe that's just another "two bananas" answer.

Maybe your subconscious has more sense than mine. Rather than advising you to charge ahead, it advises caution. You might sometimes find that the nagging doubt interferes with your ability to focus and solve problems.

Remember that those inner urges are just that: urges, tendencies, suggestions. It's up to you, the rational part, to analyze the whole picture and make the best decision. So what if your subconscious doesn't like some of those decisions and wants to churn up your stomach? That's the same part of your brain that can't add up three numbers. Ignore it.

cartoon

The Guru Speaks

When you are stuck, you need to get unstuck. Gee, that's great advice, isn't it? Somebody ought to put that on a plaque so more people can benefit. Definitely Nobel prize material, I think. OK, maybe not. But part of problem solving is being able to recognize that it is YOU who is stuck. You have all the facts (or you hope so, anyway), but they don't add up to a solution. Maybe you'd like some more facts to mull over, and possibly you can run some experiments that will give you more data to work with. But maybe not. Maybe the set of observations you have in hand right now is all you get. Like the spiral thread of a screw, you follow the logic downward, and it twists ever smaller, downard, smaller, smaller and.. vanishes into thin air. No answer, no solution, no ideas.

Sometimes it helps to list what you don't know, rather than what you do. I don't really know that print job ever got to the spooler, do I? OK, spoolers have logs, so I can check that. Come to think of it, do I really know that user is not printing to some other printer? Maybe there's a big stack of paper in the output bin of the printer down on the next floor. What else don't I know? How do I know that the printer isn't in fact using white ink? Exactly the right shade of ink to blend perfectly with the paper, I bet. Not only that, but because it is a very smart printer, when it sees it has printed an entire page in white ink that no one will be able to read, it doesn't bother to eject what would be a useless blank piece of paper.

Silly thoughts? Yes, but silly thoughts can be good. You're stuck anyway, and nothing is working, so indulge yourself in fantasy for a minute. No, it's not white ink, it's Bit Bandits in the network! These ravenous critters are living in the cable between the switch and the print server, and are gobbling up all the printer data we try to send. Ugly little things and the only way to purge them is to tear out the cable and replace it.. hmm, maybe that's not a bad idea. It could be a bad cable, couldn't it? Or.. back to that white ink again, maybe it is printing, and just isn't ejecting. Or maybe..

That's the point of letting fantasy and imagination loose for a little bit. You should have thought of the idea of a bad cable, but for some reason you didn't. OK, yes, that was a tremendous oversight. So sue me. But your printer is printing now, isn't it? So there.

Or maybe it isn't. Oh well, it was worth a try. We don't want to over do this imagination thing. It's probably not going to help if we sit down and write a 400 page Raiders Of The Printing Network fantasy. But remember this: if logic alone could solve all problems, the computer itself could help you do it. Like those Microsoft Troubleshooting Wizards that ask questions and lead you to the "solution". Except they never do solve anything, do they?

Sometimes, that's where our problem solving leads us. Most of us have better track records than those Microsoft Wizards, but we can still come up dry. Absolute dead ends, no place to turn, no clue as to what to do next. Obviously there has to be something we've missed, or misunderstood, misinterpreted or never knew about at all: computers don't work by magical spells. Every computer problem has a logical cause, and a logical solution. In most cases, there is at least a slightly better solution than tossing it on the floor and throwing heavy objects at it. Or jumping up and down on it, which is what I watched a brother-in-law once do to a misbehaving printer. There just had to be a better way to convince it to print italics, wouldn't you agree?

My brother-in-law later claimed that he needed a new printer anyway, and that he was under a deadline, and very stressed, and the physical release of stomping that printer to death was therapeutic and enabled him to return to his work with a clear mind. Sure, Bob, whatever you say. But there is a lesson there: sometimes you need to walk away from a problem entirely, if for no other reason than to clear your head. Take a walk, take a break, talk about something else for while. Take a nap, even. The folks who need this problem fixed NOW will understand, right? OK, so they won't. Sneak out the back door.

So what do you do when you come back from that empowering sabbatical and you still have no clue? No hunches, no insights, just a big empty space where the ideas are supposed to be?

Do something dumb.

cartoon

I'm serious. Look, logic has failed you. It hasn't really, because we both know that you have missed something, misunderstood something, and so on. You've tried listing everything out on 3 by 5 cards and shuffling them to see relationships you might have missed. You've tried explaining the whole thing to your cat (who wouldn't tell YOU how to fix it even if she did know), but you are still dead in the water. Something in your logic is wrong, is not what it seems. You've made a mistake, boopy, and we both know it, because if you hadn't, this problem would be solved by now, wouldn't it? Are you ready to admit that yet? OK, good, so go ahead and do something dumb, something that you already know will not work. "What's the point?", you ask, making that face you always make when you think I'm the one with no clue.

Here's the point: we've already established that your logic is faulty somewhere. If that's the case, and it surely is, why do you still suffer under the delusion that you know what actions are dumb and pointless? Since we know that you plainly don't really understand the nature of the problem, how can you be so darn certain that doing whatever to the thing-or-morackety won't clear everything up? It probably won't, but you might get lucky, or at least you might learn something that will help you see why your thinking isn't getting to the solution. Besides, you need to do something. Sitting there with your jaw hanging open isn't adding to your "clever guy" credits. Still no ideas? Sheesh, do I have to do all your thinking for you?

OK, how about this: is there any way to simplify or isolate the problem? In a way, that might have been why my subconscious wanted that Mac server shut off. If I did realize that the server was a new addition, I also knew that the network used to work without it. So maybe it still could. So shut off the new thing. For the same reason, I usually carry a little five port network switch in my tool bag. If I run into something on a network that just makes no sense whatsoever, I sometimes isolate a few machines on their own little lan just to see what happens when everything else isn't hooked up to them. Or I might run down to Radio Shack and buy a little router and put a few machines on a different subnet. None of this may help at all, but if I'm getting nowhere anyway, why not?

Hardware people understand the value of isolation and simplification. If a machine doesn't boot, or crashes, disconnect it from the network, and start pulling cards. Pull everything that isn't absolutely needed to boot. If the problem goes away, start putting things back in, one at a time, until the problem comes back. The same thing applies to software: if you have a program that crashes, try to write the smallest possible program that still crashes. Strip out everything until you are left with the handful of lines that exhibit the problem.

cartoon

But still.. sometimes none of this helps. Maybe we just don't have enough information. We can't know everything, can we? Moreover, not all problems have solutions. A dead goldfish remains dead no matter how badly we want to change that fact for the benefit of a sobbing three year old. As Scotty said to Kirk, "I'm an engineer, not a miracle worker". Ignomious failure is part of trouble shooting, and you just have to get used to it. You don't know everything, and you never will. Mix that with the small errors and mistakes that we all make, and the wollapalooza bonehead moves that we are bound to do at least now and then, and there it is: we're only human. We will sometimes fail.

I don't like failing. Who does? It ticks me off, and especially so when a later review reveals that I was a bit lazy, or careless, or had forgotten something important that I should have remembered. I also get annoyed with myself when the failure was due to lack of knowledge or skill that I should have possessed. "You never really did understand quantum mechanics, and NOW look what that's cost us!". Well, maybe I wouldn't beat myself up for that, but you get the idea. There's only so much time in our lives, but it's easy to look back and think you could have done better. There was plenty of time to study source code instead of eating breakfast every day.

Yeah. And while we're on the subject, let's get it all out. No point beating yourself up halfway. You ARE lazy, incompetent, wasteful, not worth anything at all, plainly unintelligent, incapable of contributing anything useful, not qualified, and very close to stupid. And let's not forget clumsy, forgetful, malicious, petty, cruel - oh, my, you are a most wretched sort, aren't you? No wonder the printer still isn't printing - it is your moral failings that prevent it from being fixed!

Wait, I have a better idea. It's not your fault, it's the stupid people who designed the operating system. The idiots have made a proper mess of it, no wonder you can't fix it. It is they who are the fools, they who have caused this misery, this printing hell for the fine employees of Amalmagated Widgets, Inc. They should be sued, but we all know that they have bribed all the judges. There is no Justice!

OK, do we all feel better now? True, the printer still is stubbornly producing no pages, but wasn't all that venting helpful? A breath of fresh air, and I'm sure the people who say you look a little scary right now are just kidding. Deep breaths, deep breaths. Oh look, what's that? Is that cable unplugged? Ohh..



Got something to add? Send me email.





(OLDER)    <- More Stuff -> (NEWER)    (NEWEST)   

Printer Friendly Version

-> -> Troubleshooting: Try something different



Increase ad revenue 50-250% with Ezoic


More Articles by

Find me on Google+

© Tony Lawrence



Kerio Samepage


Have you tried Searching this site?

Support Rates

This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.

Contact us





The successful construction of all machinery depends on the perfection of the tools employed; and whoever is a master in the arts of tool-making possesses the key to the construction of all machines... The contrivance and construction of tools must therefore ever stand at the head of the industrial arts. (Charles Babbage)





This post tagged: