Observability in OpenStack Learnings and building blocks from the SIG Monitoring at SCS
summary
SCS teams standardize on Prometheus + Grafana for metrics, reuse the Awesome Alert Rules repo, and run behavior checks that measure the time to create workloads via public OpenStack APIs; an existing 4,000-line shell script (OpenStack health monitor) is being replaced by OTC’s more modular monitoring project after evaluating Rally and Tempest. OVN has a Paul Greenberg exporter but it duplicates many metrics, logs require selectable external shipping for legal needs, and metering is entering a funded tender.
AI-written summary · about
chapters (14)
- 0:00Introduction and speakers
- 1:54What is SCS project
- 5:28Definition of observability
- 6:33Challenges in OpenStack monitoring
- 7:38OpenMetrics and Prometheus choice
- 9:24Behavior-based monitoring in SCS
- 11:13Current observability tools and dashboards
- 11:56Future tasks and capacity management
- 13:18Observing OVN and exporters
- 13:39Behavior monitoring frameworks
- 15:02Logs and log analysis
- 16:27Metering and billing integration
- 17:29Learnings from SIG Monitoring
- 18:34How to get involved
related talks
25mOpen source observability for private cloud: mission impossible or not?
29mParanoid Observability –Success Stories and Misfortunes observing an observability stack
28mMonitoring Stack was yesterday – Observability Platform at scale!
28mImprove Monitoring and Observability for Kubernetes with OSS Tools
27mMonitoring as Code: Managing your Observability Assets the Cloud-Native Way
27mObservability with Prometheus and beyond
24mPractical Kubernetes Monitoring with Prometheus
45mDecentralizing SaaS Monitoring Using Open-Source Tools and Standards