Hidden Signals in K8s Clusters
summary
Cloud-agnostic cluster dashboard Commodore UI shows change tracking, event timelines and a new automated RCA agent (Plaudia/Claudia) that claims to mimic SRE reasoning and surface correlated insights; demo walks through investigations and a cost-rightsizing estimate (advisory only). Practical tactics include 15-minute baseline buckets with ~900s granularity, velocity-aware alerts (example: disk filling 10%/hour → six hours to full), drilling into iostat for EBS microburst IOPS, particulate sensors that auto-trigger purifiers, and 30-day log retention.
AI-written summary · about
chapters (8)
- 0:00Introduction and Speaker Background
- 1:21Personal Anecdotes and Home Lab Insights
- 4:32Hidden Environmental Factors and Monitoring
- 6:52Advanced Metrics and Analysis Techniques
- 8:32Career Journey and Lessons Learned
- 13:49Deep Dive into Storage and Instrumentation
- 17:32Commodore Demo and Features
- 20:16Q&A Session
related talks
36mWhat's going on in my cluster?
31mBest Practices for Effective Multi-Cluster Observability: Lessons Learned
37mProactive Kubernetes Security: Anomaly Detection and Runtime Alertin...
35mAcing Distributed Container Management
30mEffective Detection in Kubernetes Clusters
32mSurviving From Endless Issues Coming From 7K+ Kubernetes Clusters
44mEssential-8 Security in Practice with Kubernetes
29mKubernetes-Defined Monitoring [I]