Enhancing Observability and Security in Microservices and Kubernetes:
summary
Explains using Calico (open source/Enterprise/Cloud) with Prometheus + Grafana to monitor Kubernetes networking by tracking Typha and Felix metrics, provides concrete queries and downloadable dashboards. Key facts: Typha handles ~100 clients to reduce API server connections, breadcrumb_size = 100 default and blocking vs nonblocking breadcrumbs, watch breadcrumb_size, nonblocking_fraction, client_fall_behind, client_ping_latency, graph_update_time and data_plane_apply_time, and active_calico_nodes. Live lab restarts Typha and rapidly scales pods to show metric spikes and confirm alerts work.
AI-written summary · about
chapters (8)
- 0:00Introduction and setup
- 1:03Observability vs monitoring
- 4:14Key Calico components
- 5:17Metrics and lab setup
- 10:33TAA metrics dashboard
- 12:42Felix metrics explained
- 15:33Stressing the system demo
- 17:48Summary and resources
related talks
28mImprove Monitoring and Observability for Kubernetes with OSS Tools
24mPractical Kubernetes Monitoring with Prometheus
11mKubernetes Observability
32mUnderstanding Observability in Kubernetes
37mMonitoring Kubernetes with Prometheus
19mKeynote: Monitoring Kubernetes Clusters with Prometheus
32mTesting Kubernetes apps' observability end-to-end
42mDistributed Monitoring: How to understand the Chaos