How to Solve a Software Bug FAST

[ @todo: Insert photos of bugs and fun stuff.]

It’s not always important to solve bugs with great speed.  In fact, “slow down” is a tip I give myself regularly.  BUT, it’s true that I’ve figured out some favorite techniques that help me solve software development bugs in a timely manner, and I hope you find something helpful here!

#1 Read the bug report carefully and ask questions.

Most initial bug reports don’t have all of the important details.  Ask questions so that you don’t lose time going down the wrong road.  Well thought out questions show that you are working on the issue, which is good news to anyone who cares about it getting fixed.

#2 Reproduce the bug, ideally on your local development sandbox.

You’re going to have to set up the bug locally anyway, in order to test your fix, and my advice is to make reproducing it locally something you do very early on in the debugging process.

The process of trying to reproduce it may shed some light on the situation, which helps with knowing what questions to ask (see point #1).

I have an approximately 2000x greater chance in fixing a bug I can reproduce than in fixing a bug that I can’t reproduce.

#3 Ask for Help

If the importance is on the speed on which the bug gets solved, then the more brains, the better!  How soon you should ask for help probably depends on the seriousness of the issue, and your likelihood of solving the issue on your own.  Be honest, do you have the technical knowledge here?   Have you solved other similar bugs before?

There is no shame in not knowing something.  Often it is just a matter of experience, which you will gain in time.

#4 Watch out for the Wild Guesses

The reason not to ask for help, is that sometimes people will help by giving suggestions that could not possibly be correct.  Be careful!

I only consider answers which “could not possibly be correct” after I’ve already checked out the technically most likely causes.  Don’t go down a rabbit hole that’s a 1 in a million chance.  (On the other hand, don’t assume that you have superior knowledge, and be careful to not dismiss a suggestion just because you don’t understand it.  Ask questions.)

#5 Define the scope

Software systems are HUGE!  Let’s make the haystack smaller…

If this is a website bug, does it still happen with Javascript disabled?  That’s a good way to tell if it’s a backend or frontend issue.

If you disable a plugin/module/theme or otherwise remove some included files, does the bug go away?

Another way to define scope, is to use a debugger and put a break point before and after where the bug exists.  Examine what happens to the relevant variable values until you can be certain of the code block where the error happens.

#6 Assume it’s a recent change.

Did you do something recently?   Look at your recent commits, software updates, etc.  If this is a huge problem that did not exist before, something must have changed to cause it.  You can also check out previous git commits/other git branches and see if the bug existed in them.

#7 Stay Focused; Put the “Emotions” aside.

Has this bug caused your adrenaline to shoot up?

Or, has it caused others to stress?

Try to put all of your feelings aside, and follow a process driven by logic instead.

I find that understanding emotion is very helpful when interacting with people, but computers don’t care how I’m feeling.

YOUR emotions, as well as the emotions of OTHERS can interfere with your ability to use reason to solve an issue that’s logic based.

Do what you need to do in order to stay focused on solving the issue.  I (try to) banish my emotions to the other room.

Leave a Reply

Your email address will not be published. Required fields are marked *