From New Cluster to Insight. Deploying Monitoring and Logging to Kubernetes
summary
Deploys a full observability stack to a fresh cluster with Helm: Prometheus operator for scraping, node-exporter plus exporters for node and pod metrics, and Grafana wired to a Loki and promtail logging pipeline. Demonstrates bootstrapping with persistent volumes (5 GB in demo), runs an emoji-voter load generator, and shows Grafana Explore correlating Prometheus queries with log streams. DigitalOcean reports 4.5M metric samples/sec (spiking to 9M during upgrades) and ~30k log lines/sec; Loki favors lightweight ingestion over heavy full‑text indexing.
AI-written summary · about
chapters (34)
- 0:00Introduction
- 0:33Show of Hands
- 1:24Take a Step Back
- 1:57Create a Cluster
- 2:52The Second Day
- 3:23The Map
- 4:09Two Sections
- 6:10Tools
- 6:48Prometheus
- 7:29Prometheus Operator
- 8:16Digital Ocean
- 8:58Lightstep
- 9:18Logging
- 10:29Shippers
- 11:29Datadog
- 11:51Efk Stack
- 12:48Kubernetes Standard
- 13:56syslog
- 14:23Helm
- 15:35Installing Helm
- 18:38Loki Install
- 19:39Insight Install
- 20:56Emoji Voter
- 22:02Dashboards
- 22:33Microservices
- 23:27Logs
- 23:46Emoji Voting
- 24:03Split Job Container
- 24:34Query View
- 25:03Log Labels
- 25:34Prometheus Alert Manager
- 26:16Highlevel overview
- 26:35Github repo
- 27:26Question
related talks
24mPractical Kubernetes Monitoring with Prometheus
27mFrom Zero Visibility to Full Observability with Grafana
30mImplementing Prometheus and Grafana in a Kubernetes Cluster on EKS
28mImprove Monitoring and Observability for Kubernetes with OSS Tools
40mBuilding a Scalable Observability Platform for Edge Compute Nodes
19mKeynote: Monitoring Kubernetes Clusters with Prometheus
30mCrash course in Kubernetes monitoring
32mDemocratizing Application Monitoring with Prometheus