Book Image

Datadog Cloud Monitoring Quick Start Guide

By : Thomas Kurian Theakanath
Book Image

Datadog Cloud Monitoring Quick Start Guide

By: Thomas Kurian Theakanath

Overview of this book

Datadog is an essential cloud monitoring and operational analytics tool which enables the monitoring of servers, virtual machines, containers, databases, third-party tools, and application services. IT and DevOps teams can easily leverage Datadog to monitor infrastructure and cloud services, and this book will show you how. The book starts by describing basic monitoring concepts and types of monitoring that are rolled out in a large-scale IT production engineering environment. Moving on, the book covers how standard monitoring features are implemented on the Datadog platform and how they can be rolled out in a real-world production environment. As you advance, you'll discover how Datadog is integrated with popular software components that are used to build cloud platforms. The book also provides details on how to use monitoring standards such as Java Management Extensions (JMX) and StatsD to extend the Datadog platform. Finally, you'll get to grips with monitoring fundamentals, learn how monitoring can be rolled out using Datadog proactively, and find out how to extend and customize the Datadog platform. By the end of this Datadog book, you will have gained the skills needed to monitor your cloud infrastructure and the software applications running on it using Datadog.
Table of Contents (19 chapters)
Section 1: Getting Started with Datadog
Section 2: Extending Datadog
Section 3: Advanced Monitoring

Best practices

Now, let's review the best practices related to monitors and alerts:

  • Define a comprehensive list of monitors that will cover all aspects of the working of the software system that Datadog is monitoring.
  • It's possible that you might only be using Datadog to address a certain part of the monitoring requirement, and, in that case, make sure that the monitors defined cover your area of focus.
  • To avoid alert fatigue, make sure that all alert notifications are actionable, and the remediation steps are documented in the notification itself or in a runbook. Continue fine-tuning the thresholds and data window size until you start getting credible alert notifications. If monitors are too sensitive, they can start generating noise.
  • If any amount of fine-tuning is not enough to tame a monitor from sending out too many alerts, consider deleting it, and plan to monitor the related scenario in some other reasonable way.
  • Integrate monitors with your...