Detailed white-box monitoring is maybe the single most important thing to get right when you’re moving to microservices.
Weaveworks has been running a cloud service with Kubernetes and Prometheus for over a year, and we’ve learned a lot about what works and what doesn’t.
This talk will cover epic fails we experienced so that you don’t have to, using the RED method to monitor what matters, and production outages we solved with detailed telemetry. You’ll learn ways you’re under-instrumenting your services, what you can do to fix it, and how to make sure your graphs aren’t lies.
You can view Paul’s slides below: