Monitoring has always been vitally important to ensure the health and service continuity of computerized information systems. The onslaught of cloud-native computing paradigms such as containerization and microservices has further emphasized its importance, Moreover - modern scale and fluidity of IT systems has made traditional monitoring approaches and systems irrelevant.
Prometheus is an open source monitoring system for this new brave world. It aims to provide an end-to-end solution for instrumenting your cloud-native applications and infrastructure, gathering the data and translating it into actionable operational metrics.
In the workshop we’ll look into modern approaches to monitoring and instrumentation and get our hands dirty running and configuring a Prometheus instance, instrumenting an application and writing basic PromQL queries for data collection and analysis.
Software development and operation professionals interested in modern monitoring and instrumentation approaches.
Basic knowledge of Linux command-line and shell scripting. Basic understanding of web application development, preferably in Python. Experience with Docker is helpful but not required.
Each participant will need a laptop with a browser and an internet connection.
○ The Short History of Monitoring
○ Modern Monitoring Challenges and Approaches
○ Pull-based vs. push-based systems
○ Defining code instrumentation.
○ How Prometheus is different.
○ Prometheus architecture
○ Prometheus client libraries
■ Metric types
○ Prometheus Query Language - the basics
○ Installing and running Prometheus
○ Defining targets and instances
○ Monitoring system metrics with node exporter
○ Instrumenting a Flask app with python client library
○ Running basic queries
○ Integrating with Grafana
○ Setting up alerts