How Kubernetes Audit Logging Causes etcd Performance Degradation
from DevOps Daily with Fexingo: CI/CD, Kubernetes, and Modern Software Operations Hosted by Fexingo
In this episode of DevOps Daily with Fexingo, Lucas and Luna dive into a subtle but costly Kubernetes performance issue: how verbose audit logging can silently degrade etcd performance, leading to increased latency, timeouts, and even cluster instability. They walk through a real-world case from a mid-sized fintech that saw API response times jump from 5ms to over 200ms after enabling audit logs without tuning. You'll learn about the audit log pipeline, how etcd handles writes, and three practical mitigations: adjusting audit policy verbosity, using webhook batching, and offloading logs via sidecar. No theory—just actionable advice for anyone running Kubernetes in production. #Kubernetes #AuditLogging #etcd #Performance #CloudNative #SRE #DevOps #ProductionIncident #Fintech #K8s #Logging #Latency #ClusterStability #Monitoring #Tuning #Technology #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
Embed this episode
No transcript yet.