As a geek I’m often asked, “How do you know this stuff?”
Well, there’s no magic bullet, we (geeks) do not have inherent knowledge of technology. We didn’t simply wake up one morning and build a network, or repair a server. A good system engineer will need to be an excellent troubleshooter, this skill, like art, can be cultivated and groomed, but is very difficult (if not impossible) to learn. Like art, you’ve either got it, or you don’t.
The journey to becoming a good system administrator / engineer (i.e. geek) will require all forms of learning. Traditional methods include; reading, classroom instruction, seminars, and most important hands-on experience. Non traditional methods are (CIO’s, Directors, and the like cover your ears) system failures, crisis and emergency repairs.
Let’s start with the easy stuff, traditional learning. Reading is essential to learning technology. Yes, it’s arduous, and the three inch thick IT books (replaced by gigantic PDF’s) are about as exciting to read, as your mortgage papers, and usually written as well. Fear not, it’s rare that it will be necessary to read one of these monsters from cover to cover. Read the first chapter or two, skim through the rest and add it to your library to use as a reference as needed (and you will need it). The point here is that unless you’re a true genius, you can’t retain all of that information anyhow, understand the concepts, and refer to it as needed for the detailed information and procedures. Classroom instruction will have higher learning retention than reading (at least it does for me). The reason for the higher retention is the hands on experience most classrooms provide. The triple play of lecture, reading and doing, drives the lesson home. Seminars are good resources to discover what’s available, and to keep up with the new technologies, products and trends. The seminar will rarely teach you how to use or implement the new technology, but only serve as an introduction.
There are two ways to retain 100% of your knowledge: repetition and crisis. I certainly don’t recommend causing a crisis as a learning tool, but if you’re in IT long enough, crisis will find you, and you will never forget the solution. A crisis forces you out of your comfort zone, requires you to dig deep, and implement a solution in minutes or hours, not weeks and months. This rapid problem solving and the associated stress will set the knowledge firmly into your grey matter.
Did you know that Microsoft clusters rely on the disk signature in the MBR (master boot record) to mount the drives? I do. How did I learn this, well it wasn’t in any book I read, or covered in during my clustering class, I learned this one the hard way. After an otherwise successful SAN migration none of the clustered servers would come online. The event log said that the disk could not be found, but the disks were mounted and accessible to the nodes, this didn’t make sense. An hour or so of combing the internet reveiled the solution. Using DISKPART it’s possible to edit the disk signature. Fortunately the signatures it was looking for were listed in the error messages, and after all of the signatures were updated; the clusters came online to the great relief of my boss and myself. I will forever remember the relationship of disk signatures and cluster volumes, I’m now what some would call, an expert.
These catastrophic events can also make or break a career in IT, did you attack the problem head on, and deliver a solution, or did you pee your pants. The best IT professionals, have done their homework, learned their lessons, and most importantly (but hopefully not often) can work under the enormous stress of a crisis and apply their knowledge.
That’s how I know this stuff.
Photo courtesy of Seth Sawyers.