Inclusive monitoring of a microservice architecture with Rancher and the Prometheus ecosystem – (Martin Baillie @ IAG)
In this talk and demo, we will look at how you can use the Prometheus ecosystem to monitor, alert, graph and forecast on the internal state of a microservice architecture. We’ll focus on how you can scale this solution out as your platform grows using techniques like zero-conf discoverability and telemetry, and also how you can better encourage an inclusive monitoring culture within your engineering teams.
- Traefik with the new Rancher metadata service provider and Let's Encrypt
Official upstream images with confd sidekicks to provide custom configuration discovered from the Rancher metadata service. Grafana dashboards, Prometheus alerts stored as code beside the images.
- Prometheus
- Grafana
- Alertmanager
- Node Exporter (Host metrics)
- CAdvisor (Docker metrics)
- Rancher Exporter (Rancher metrics)
- Container Monitor (IPSec metrics)
Rob Pike's talk on Go Proverbs