Seeing the Forest for the Trees: An Operator’s Guide to Observability of CF with...
summary
Metric Store ingests everything from the Loggregator firehose via a nozzle, writes envelopes to disk (sharded daily) and gives a persistent six weeks / 42 days retention by default. It embeds the Prometheus query engine so queries use PromQL (instant and range), supports topk and arithmetic (example: top 5 apps by CPU, compute unused memory across Diego cells), exposes Prometheus API for Grafana visualization and uses UAA/Cloud Controller for auth; Alertmanager integration is in progress.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:16Overview
- 0:55What is Observability
- 2:26The Four Golden Signals
- 3:48What is a Good SLO
- 5:51What Makes a Good SLO
- 7:47Metricstor
- 11:01Prometheus
- 13:30Ravana
- 15:10Metrics Store
- 15:58Prometheus Query Language
- 18:19TopK
- 20:26Memory Quota
- 22:04API Endpoints
- 24:44Recap
related talks
29mCloud Foundry Logging & Metrics
36mObservability with Prometheus and Beyond: Leveraging Cloud-native Technologies Outside of Cloud
27mObservability with Prometheus and beyond
30mMonitoring Cloud Foundry: Learning about the Firehose
48mMetrics All the Way: Data Driven Operations
41mMetrics Applied
37mCloud-native Metric Monitoring with Prometheus
41mResilient Observability at the Retail Edge: A Lightweight, Scalable, and...