Debugging: The 9 Indispensable Rules for Finding Even the Most Elusive Software and Hardware Problemsby David J. Agans
When the pressure is on to root out an elusive software or hardware glitch, what's needed is a cool head courtesy of a set of rules guaranteed to work on any system, in any circumstance. Written in a frank but engaging style, Debugging provides simple, foolproof principles guaranteed to help find any bug quickly. This book makes those shelves of/i>… See more details below
When the pressure is on to root out an elusive software or hardware glitch, what's needed is a cool head courtesy of a set of rules guaranteed to work on any system, in any circumstance. Written in a frank but engaging style, Debugging provides simple, foolproof principles guaranteed to help find any bug quickly. This book makes those shelves of application-specific debugging books (on C++, Perl, Java, etc.) obsolete. It changes the way readers think about debugging, making those pesky problems suddenly much easier to find and fix.
Illustrating the rules with real-life bug-detection war stories, the book shows readers how to:
* Understand the system: how perceiving the "roadmap" can hasten your journey
* Quit thinking and look: when hands-on investigation can't be avoided
* Isolate critical factors: why changing one element at a time can be an essential tool
* Keep an audit trail: how keeping a record of the debugging process can win the day
Author Biography: David J. Agans (Milford, NH) is a recognized expert called in to help with tough debugging problems. He currently runs PointSource, a computer systems consultancy. He has worked with industrial control and monitoring systems, integrated circuit design, handheld PCs, videoconferencing, and countless other systems.
San Francisco, CA February 2003
Debugging explains the fundamentals of finding and fixing bugs (once a bug has been detected), rather than any particular technology. It's best for developers who are novices or who are only moderately experienced, but even old pros will find helpful reminders of things they know they should do but forget in the rush of the moment. This book will help you fix those inevitable bugs, particularly if you're not a pro at debugging. It's hard to bottle experience; this book does a good job. This is a book I expect to find useful many, many, years from now.
- AMACOM Books
- Publication date:
- Sales rank:
- Product dimensions:
- 6.69(w) x 9.61(h) x 0.50(d)
Read an Excerpt
The Nine Indispensable Rules for Finding Even the Most Elusive Software and Hardware Problems
By David J. Agans
AMACOMCopyright © 2002 David J. Agans
All rights reserved.
"At present I am, as you know, fairly busy, but I propose to devote my declining years to the composition of a textbook which shall focus the whole art of detection into one volume."
—SHERLOCK HOLMES, THE ADVENTURE OF THE ABBEY GRANGE
This book tells you how to find out what's wrong with stuff, quick. It's short and fun because it has to be—if you're an engineer, you're too busy debugging to read anything more than the daily comics. Even if you're not an engineer, you often come across something that's broken, and you have to figure out how to fix it.
Now, maybe some of you never need to debug. Maybe you sold your dot.com IPO stock before the company went belly-up and you simply have your people look into the problem. Maybe you always luck out and your design just works—or, even less likely, the bug is always easy to find. But the odds are that you and all your competitors have a few hard-to-find bugs in your designs, and whoever fixes them quickest has an advantage. When you can find bugs fast, not only do you get quality products to customers quicker, you get yourself home earlier for quality time with your loved ones.
So put this book on your nightstand or in the bathroom, and in two weeks you'll be a debugging star.
How Can That Work?
How can something that's so short and easy to read be so useful? Well, in my twenty-six years of experience designing and debugging systems, I've discovered two things (more than two, if you count stuff like "the first cup of coffee into the pot contains all the caffeine"):
1. When it took us a long time to find a bug, it was because we had neglected some essential, fundamental rule; once we applied the rule, we quickly found the problem.
2. People who excelled at quick debugging inherently understood and applied these rules. Those who struggled to understand or use these rules struggled to find bugs.
I compiled a list of these essential rules; I've taught them to other engineers and watched their debugging skill and speed increase. They really, really work.
Isn't It Obvious?
As you read these rules, you may say to yourself, "But this is all so obvious." Don't be too hasty; these things are obvious (fundamentals usually are), but how they apply to a particular problem isn't always so obvious. And don't confuse obvious with easy—these rules aren't always easy to follow, and thus they're often neglected in the heat of battle.
The key is to remember them and apply them. If that was obvious and easy, I wouldn't have to keep reminding engineers to use them, and I wouldn't have a few dozen war stories about what happened when we didn't. Debuggers who naturally use these rules are hard to find. I like to ask job applicants, "What rules of thumb do you use when debugging?" It's amazing how many say, "It's an art." Great—we're going to have Picasso debugging our image-processing algorithm. The easy way and the artistic way do not find problems quickly.
This book takes these "obvious" principles and helps you remember them, understand their benefits, and know how to apply them, so you can resist the temptation to take a "shortcut" into what turns out to be a rat hole. It turns the art of debugging into a science.
Even if you're a very good debugger already, these rules will help you become even better. When an early draft of this book was reviewed by skilled debuggers, they had several comments in common: Besides teaching them one or two rules that they weren't already using (but would in the future), the book helped them crystallize the rules they already unconsciously followed. The team leaders (good debuggers rise to the top, of course) said that the book gave them the right words to transmit their skills to other members of the team.
Anyone Can Use It
Throughout the book I use the term engineer to describe the reader, but the rules can be useful to a lot of you who may not consider yourselves engineers. Certainly, this includes you if you're involved in figuring out what's wrong with a design, whether your title is engineer, programmer, technician, customer support representative, or consultant.
If you're not directly involved in debugging, but you have responsibility for people who are, you can transmit the rules to your people. You don't even have to understand the details of the systems and tools your people use—the rules are fundamental, so after reading this book, even a pointy-haired manager should be able to help his far-more-intelligent teams find problems faster.
If you're a teacher, your students will enjoy the war stories, which will give them a taste of the real world. And when they burst onto that real world, they'll have a leg up on many of their more experienced (but untrained in debugging) competitors.
It'll Debug Anything
This book is general; it's not about specific problems, specific tools, specific programming languages, or specific machines. Rather, it's about universal techniques that will help you to figure out any problem on any machine in any language using whatever tools you have. It's a whole new level of approach to the problem—for example, rather than tell you how to set the trigger on a Glitch-O-Matic digital logic analyzer, I'm going to tell you why you have to use an analyzer, even though it's a lot of trouble to hook it up.
It's also applicable to fixing all kinds of problems. Your system may have been designed wrong, built wrong, used wrong, or just plain got broken; in any case, these techniques will help you get to the heart of the problem quickly.
The methods presented here aren't even limited to engineering, although they were honed in the engineering environment. They'll help you figure out what's wrong with other things, like cars, houses, stereo equipment, plumbing, and human bodies. (There are examples in the book.) Admittedly, there are systems that resist these techniques—the economy is too complex, for example. And some systems don't need these methods; e.g., everybody already knows what's wrong with the government.
But It Won't Prevent, Certify, or Triage Anything
While this book is general about methods and systems, it's very focused on finding the causes of bugs and fixing them.
It's not about quality development processes aimed at preventing bugs in the first place, such as ISO-9000, code reviews, or risk management. If you want to read about that, I recommend books like The Tempura Method of Totalitarian Quality Management Processes or The Feng Shui Guide to Vermin-Free Homes. Quality process techniques are valuable, but they're often not implemented; even when they are, they leave some bugs in the system.
Once you have bugs, you have to detect them; this takes place in your quality assurance (QA) department or, if you don't have one of those, at your customer site. This book doesn't deal with this stage either—test coverage analysis, test automation, and other QA techniques are well handled by other resources. A good book of poetry, such as How Do I Test Thee, Let Me Count the Ways, can help you while away the time as you check the 6,467,826 combinations of options in your product line.
And sooner or later, at least one of those combinations will fail, and some QA guy or customer is going to write up a bug report. Next, some managers, engineers, salespeople, and customer support people will probably get together in a triage meeting and argue passionately about how important the bug is, and therefore when and whether to fix it. This subject is deeply specific to your market, product, and resources, and this book will not touch it with a ten-foot pole. But when these people decide it has to be fixed, you'll have to look at the bug report and ask yourself, "How the heck did that happen?" That's when you use this book (see Figure 1-1).
The following chapters will teach you how to prepare to find a bug, dig up and sift through the clues to its cause, home in on the actual problem so you can fix it, and then make sure you really fixed it so you can go home triumphant.
More Than Just Troubleshooting
Though the terms are often interchanged, there's a difference between debugging and troubleshooting, and there's a difference between this debugging book and the hundreds of troubleshooting guides available today. Debugging usually means figuring out why a design doesn't work as planned. Troubleshooting usually means figuring out what's broken in a particular copy of a product when the product's design is known to be good—there's a deleted file, a broken wire, or a bad part. Software engineers debug; car mechanics troubleshoot. Car designers debug (in an ideal world). Doctors troubleshoot the human body—they never got a chance to debug it. (It took God one day to design, prototype, and release that product; talk about schedule pressure! I guess we can forgive priority-two bugs like bunions and male pattern baldness.)
The techniques in this book apply to both debugging and troubleshooting. These techniques don't care how the problem got in there; they just tell you how to find it. So they work whether the problem is a broken design or a broken part. Troubleshooting books, on the other hand, work only on a broken part. They boast dozens of tables, with symptoms, problems, and fixes for anything that might go wrong with a particular system. These are useful; they're a compendium of everything that has ever broken in that type of system, and what the symptoms and fixes were. They give a troubleshooter the experience of many others, and they help in finding known problems faster. But they don't help much with new, unknown problems. And thus they can't help with design problems, because engineers are so creative, they like to make up new bugs, not use the same old ones.
So if you're troubleshooting a standard system, don't ignore Rule 8 ("Get a Fresh View"); go ahead and consult a troubleshooting guide to see if your problem is listed. But if it isn't, or if the fix doesn't work, or if there's no troubleshooting guide out yet because you're debugging the world's first digital flavor transmission system, you won't have to worry, because the rules in this book will get you to the heart of your brand-new problem.
A Word About War Stories
I'm a male American electronics engineer, born in 1954. When I tell a "war story" about some problem that got solved somehow, it's a real story, so it comes from things that male American electronics engineers born in 1954 know about. You may not be all or any of those, so you may not understand some of the things I mention. If you're an auto mechanic, you may not know what an interrupt is. If you were born in 1985, you may not know what a record player is. No matter; the principle being demonstrated is still worth knowing, and I'll explain enough as I go along so you'll be able to get the principle.
You should also know that I've taken some license with the details to protect the innocent, and especially the guilty.
In this book I'll introduce the nine golden rules of debugging, then devote a chapter to each. I'll start each chapter with a war story where the rule proved crucial to success; then I'll describe the rule and show how it applies to the story. I'll discuss various ways of thinking about and using the rule that are easy to remember in the face of complex technological problems (or even simple ones). And I'll give you some variations showing how the rule applies to other stuff like cars and houses.
In the final few chapters, I've included a set of war stories to exercise your understanding, a section on using the rules under the trying circumstances of the help desk, and a few last hints for putting what you've learned to work in your job.
When you're done with this book, your debugging efficiency will be much higher than before. You may even find yourself wandering around, looking for engineers in distress so you can swoop in and save the day. One bit of advice, though: Leave the leotard and cape at home.CHAPTER 2
The Rules-Suitable for Framing
"The theories which I have expressed there, and which appear to you to be so chimerical, are really extremely practical—so practical that I depend upon them for my bread and cheese."
—SHERLOCK HOLMES, A STUDY IN SCARLET
Here is a list of the rules. Memorize them. Tape them to your wall. Tape them to all of your walls. (Coming soon: Debugging Rules wallpaper, for the seriously decor-challenged home office; definitely not recommended by feng shui experts.)
Understand the System
Make It Fail
Quit Thinking and Look
Divide and Conquer
Change One Thing at a Time
Keep an Audit Trail
Check the Plug
Get a Fresh View
If You Didn't Fix It, It Ain't Fixed
© 2001 David Agans WWW.DEBUGGINGRULES.COMCHAPTER 3
Understand the System
"It is not so impossible, however, that a man should possess all knowledge which is likely to be useful to him in his work, and this, I have endeavoured in my case to do."
—SHERLOCK HOLMES, THE FIVE ORANGE PIPS
War Story. When I was just out of college and trying to get experience (among other things), I took on a moonlighting job building a microprocessor-based valve controller. The device was designed to control the amount of metal powder going into a mold, and it took measurements from a scale that weighed the powder (see Figure 3-1). As many engineers do (especially still-wet-behind-the-ears engineers), I copied the basic design (from a frat brother who had used the same processor chip in his thesis project).
But my design didn't work. When the scale tried to interrupt the processor with a new measurement, the processor ignored it. Since I was moonlighting, I didn't have many tools for debugging, so it took me a long time to figure out that the chip that got the interrupt signal from the scale wasn't passing it on to the processor.
I was working with a software guy, and at one o'clock in the morning, very frustrated, he insisted that I get out the data book and read it, front to back. I did, and there on page 37, I read, "The chip will interrupt the microprocessor on the first deselected clock strobe." This is like saying the interrupt will happen the first time the phone rings, but the call isn't for the chip—which would never happen in my design, since I (like my frat brother) had saved some hardware by combining the clock with the address lines. To continue with the phone analogy, the phone rang only when the call was for the chip. My frat brother's system didn't have any interrupts, however, so his worked. Once I added the extra hardware, so did mine.
Later, I described my marathon debugging effort to my dad (who knew little about electronics, and nothing about microprocessors). He said, "That's just common sense—when all else fails, read the instructions." This gave me my first inkling that there are general rules of debugging that can apply to more than just computers and software. (I also realized that even a college education wasn't enough to impress my dad.) In this case, the rule was "Understand the System."
I couldn't begin to debug the problem until I understood how the chip worked. As soon as I did understand it, the problem was obvious. Bear in mind that I understood much of the system, and that was important; I knew how my design worked, that the scale had to interrupt the processor, and what clock strobes and address lines do. But I didn't look closely at everything, and I got burned.
You need a working knowledge of what the system is supposed to do, how it's designed, and, in some cases, why it was designed that way. If you don't understand some part of the system, that always seems to be where the problem is. (This is not just Murphy's Law; if you don't understand it when you design it, you're more likely to mess up.)
Excerpted from Debugging by David J. Agans. Copyright © 2002 David J. Agans. Excerpted by permission of AMACOM.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
What People are saying about this
Cleveland OH March 2003
IEEE Software magazine: "Unlike most books about debugging, this book isn't a technical manual. It's easy to follow and logical in approach and progression. The author elegantly presents the concepts and makes the debugging task unintimidating but, at the same time, points out that the major task is learning to apply his rules. Debugging not only makes the subject area easy to understand; it's an excellent example of how you can present the art of debugging in a way that's meaningful and exciting. This book should be required reading for all technology college students. It explains the logic you need in the real world, logic that so many students don't learn before graduation. After reading this book, I got excited about debugging and went out to look for something to test the ninerule approach on.""
Slashdot.com, David A. Wheeler: “It's not often you find a classic, but I think I've found a new classic for software and computer hardware developers. It's David J. Agan's Debugging….It's hard to bottle experience; this book does a good job. This is a book I expect to find useful many, many, years from now….I think this is a great book….Novices need to learn the fundamentals, and pros need occasional reminders of them; this book is a good way to learn or be reminded of them. Get this book.”"
NC, June 2003
and post it to your social network
Most Helpful Customer Reviews
See all customer reviews >
David Agans does a great job of explaining how to approach debugging as a science rather than an art. If you're a novice programmer, you'll find a wealth of distilled knowledge here that would take many years to acquire on your own. While experienced programmers may consider most of the rules to be obvious, that doesn't mean they're common practice. I've been debugging for more than 20 years, and still learnt some useful new tricks. Rule 3 ("Quit thinking and look") is a particularly good one to keep in mind. Peppered throughout the text are a large number of war stories from the author's own experience with embedded systems. As well as illustrating how to (and more commonly, how not to) approach a particular problem, these are all well written and often entertaining. Some of my favourites: how wearing the wrong shirt to work caused a new video compression chip to crash; a vacuum cleaner that made the house lights flash on and off; a noisy read/write line that led a junior engineer to mistakenly redesign an entire co-processor memory circuit; the well pump that wasn't broken; and the self-test feature on an old Pong video game. Although most examples are hardware related, the approach described can be applied to almost any problem; indeed, several of the examples used have nothing to do with computing. This is not a large book, but it's well laid out, easy to follow, and doesn't talk down to the reader. It's also packed with enough meat to satisfy the hungriest of programmers. Highly recommended.