A Kubernetes Node Can Be Empty And Full
Kubelet cleans image bytes before crisis, but small-file-heavy containers can exhaust ext4 inodes while the disk still appears to have room.
// Topic
Kubernetes
Kubernetes Failover Needs A Service Plane
A CNCF Linkerd multicluster demo shows why spare Kubernetes clusters need service identity, routing, and failure behavior to matter.
Self-Hosted LLMs Become Platform Workloads
A CNCF walkthrough of vLLM on Kubernetes shows what private inference really needs: persistent model weights, service discovery, secrets, restart behavior, and an API boundary apps already understand.
The Mesh Knows What Your App Does Not
A CNCF reference stack for OpenTelemetry and service-mesh metrics is a practical reminder that application telemetry and proxy telemetry answer different questions. Production observability needs both layers.
Security Profiles Need an API Contract
Security Profiles Operator v1 is a reminder that container hardening does not become operational until profiles can be recorded, reviewed, shipped, and bound like normal Kubernetes objects.
End of list