Monitoring Elasticsearch in OpenShift
summary
OCP 4.0 adds built-in monitoring for the openshift-logging Elasticsearch cluster using Prometheus and Grafana, and the talk explains concrete tradeoffs: choose a Prometheus exporter (Go) or an in-process Elasticsearch plugin (Java) — the Java plugin gives stronger integration and test frameworks; avoid an index-per-namespace daily pattern because it creates too many indices/shards and mapping explosion; require fast local storage and dedicated master nodes for stability; use the Elasticsearch operator to run multiple clusters and Grafana mixins to deploy dashboards.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:26Agenda
- 3:30Data Model
- 7:09Performance Tuning
- 8:48Cluster Management
- 9:25Prometheus
- 15:51Graph
- 17:11Elastic Search
- 19:23Questions
related talks
30mImplementing Prometheus and Grafana in a Kubernetes Cluster on EKS
32mDemocratizing Application Monitoring with Prometheus
35mFrom New Cluster to Insight. Deploying Monitoring and Logging to Kubernetes
46mObservability Ecosystem in Kubernetes
29m“If you Don’t Monitor your Infrastructure, you Don’t Own it!” Regain Control Thanks to Prometheus
01h 02mDesigning an open-source observability stack for Kubernetes
12mOpen-source observability is the way forward, and here’s why
28mImprove Monitoring and Observability for Kubernetes with OSS Tools