Kubernetes: Monitoring with Prometheus (2018)

Course details

In order to prevent outages, it's essential that you leverage a monitoring and alerting tool in your Kubernetes environment. Prometheus—an open-source systems monitoring and alerting toolkit—pairs particularly well with Kubernetes. In this course, learn how this toolkit integrates with Kubernetes and works to monitor distributed systems. Instructor Robert Starmer steps through how to enable Prometheus monitoring and shares how Prometheus monitors Kubernetes systems. He also discusses monitoring application-specific data, adding sidecar containers for app data, filtering and combining metrics, and displaying metrics in the web console. To wrap up, he discusses how to create a simple alert in Prometheus and generate application-driven alerts.

Learning objectives
Logging vs. monitoring
Enabling Prometheus monitoring
Capturing Kubernetes infrastructure data
Capturing container data with cAdvisor
Monitoring application-specific data
Filtering and combining metrics
Displaying metrics in the web console
Using Grafana as a dashboard
Using metrics data

Concepts

0. Introduction

01 - Prometheus monitoring to prevent outages
02 - What you should know

1. Enable Prometheus Monitoring

03 - Logging vs. monitoring
04 - Monitoring Kubernetes
05 - Enabling Prometheus monitoring, part 1
06 - Enabling Prometheus monitoring, part 2

2. Monitoring Kubernetes Systems

07 - Capturing Kubernetes infrastructure data
08 - Capturing container data with cAdvisor

3. Prometheus Basics

09 - Monitoring application-specific data
10 - Adding sidecar containers for app data
11 - Discovering collections

4. Prometheus Math

12 - Filtering and combining metrics
13 - Finding top talkers

5. Graphing

14 - Displaying metrics in the web console
15 - Using Grafana as a dashboard

6. Using Metrics Data

16 - Creating alerts in Prometheus
17 - Generating application-driven alerts

Conclusion

18 - Next steps