Book Image

Linux Administration Best Practices

By : Scott Alan Miller
3.3 (3)
Book Image

Linux Administration Best Practices

3.3 (3)
By: Scott Alan Miller

Overview of this book

Linux is a well-known, open source Unix-family operating system that is the most widely used OS today. Linux looks set for a bright future for decades to come, but system administration is rarely studied beyond learning rote tasks or following vendor guidelines. To truly excel at Linux administration, you need to understand how these systems work and learn to make strategic decisions regarding them. Linux Administration Best Practices helps you to explore best practices for efficiently administering Linux systems and servers. This Linux book covers a wide variety of topics from installation and deployment through to managing permissions, with each topic beginning with an overview of the key concepts followed by practical examples of best practices and solutions. You'll find out how to approach system administration, Linux, and IT in general, put technology into proper business context, and rethink your approach to technical decision making. Finally, the book concludes by helping you to understand best practices for troubleshooting Linux systems and servers that'll enable you to grow in your career as well as in any aspect of IT and business. By the end of this Linux administration book, you'll have gained the knowledge needed to take your Linux administration skills to the next level.
Table of Contents (16 chapters)
1
Section 1: Understanding the Role of Linux System Administrator
4
Section 2: Best Practices for Linux Technologies
9
Section 3: Approaches to Effective System Administration

Tooling and impact

One of the fundamental natures of physics, as well as a rule that you learn straight away in industrial engineering, is that you cannot observe or measure events without in some way impacting them. In computing, we face the same problem. If anything, we face it far more than in most other places.

The more that we measure, log, or put metrics on our systems the more of the system resources needed for our workloads is taken up by the measurement processes. As computers have gotten faster over the years the ability to measure without completely crippling our workloads has become more common and now, we often even track checkpoints inside of applications in addition to operating system metrics. But we always have to maintain an awareness of what this impact is.

At some point there is more value to just letting the systems that we have run as fast as they can rather than trying to measure them to see how fast they are going. A sprinter running flat out is faster...