Running Microservices on Kubernetes – Best Practices
summary
Recommends using managed Kubernetes like EKS/AKS/GKE or on‑prem OpenShift/Rancher, enforcing namespaces with resource quotas and RBAC, and preferring one container per pod except for injected sidecars. Specifies health checks: liveness/readiness/startup probes (use different endpoints to avoid restart loops and add shutdown delay), set sensible CPU/memory requests and limits, and spread replicas with anti‑affinity. Advocates shifting security left with image scans (Black Duck/Checkmarx), runtime secrets (CyberArk) and private registries (JFrog Artifactory), and using HPA + Prometheus and service mesh (Linkerd) for observability and scaling.
AI-written summary · about
related talks
49mDeveloping microservices applications on the Kubernetes platform
19mKube Me This! Kubernetes Ideas and Best Practices
57mMicroservices with Kubernetes
59mA DevOps State of Mind: Managing Microservices + Databases with Kubernetes
44mTop 5 container and Kubernetes best practices
35mAdvanced Microservices Architectures
37mKnee Deep in Microservices
45mKube Me This! Kubernetes Ideas and Best Practices
