Observability is key when running reliable production systems and it's no difference when applications are deployed on top of Kubernetes. The Cloud Native ecosystem provides numerous tools to help "observe" the system on Kubernetes. Some of these are Prometheus, Grafana, Thanos, Fluentd, Loki, and Jaeger. Making them properly work together and scale on K8s it's a difficult task alone. Moreover, when a problem occurs you need to quickly access logs, metrics, and traces and use all the tools above. This talk focuses on deep-diving into the K8s observability tools in complex multi-cluster environments to make you confident at the wheel.