Why I Wrote This Book

The monitoring landscape of today is vastly different than it was only a few years ago, and even more so than it was 10 years ago. With the widespread popularity of cloud infrastructure came new problems for monitoring, as well as creating new ways to solve old problems.

The rise in popularity of microservices has especially stretched how we think about monitoring. Since there is no longer a monolithic app server, how do we monitor interactions between the dozens, or even hundreds, of small app servers that communicate constantly? A common pattern in microservice architecture is that a server may exist for only hours or even minutes, which has wreaked havoc on the age-old tactics and monitoring tools we once relied on.

Of course, while some things change, some things stay the same. We still worry about web server performance. We still have concerns over root volumes unexpectedly running out of space. Database server performance still keeps many of us awake at night. While some of the problems we have today are similar to (or the same as) the problems we had 10 years ago, the tools and methodologies available to us are much improved. It is my goal to teach you the advances we’ve made and how to leverage them for your purposes.

I believe it is helpful to bear in mind throughout this book what the purpose and goals of monitoring are. To that end, allow me to pass along a definition of monitoring. The best definition I’ve heard comes from Greg Poirier, proposed at the Monitorama 2016 conference: