This article explains how to use Gatekeeper to enforce in-cluster admission policies, such as rejecting :latest images, mandating labels, and disallowing privileged workloads.

More: https://ku.bz/1Zskfkkvg

New on LearnKube: Server-side apply: what happens when you run kubectl apply

Server-side apply changes how Kubernetes handles field ownership.

Kubernetes objects are shared state. Manifests, controllers, release tools, autoscalers, webhooks, and operators can all shape the same object.

With client-side apply, stale intent can overwrite live changes.

With server-side apply, ownership moves into the API server. Kubernetes tracks which manager owns each field and surfaces conflicts when ownership is contested.

Chiara put serious work into this guide, and you can read it in full here: https://learnkube.com/server-side-apply-kubernetes

John Howard, Senior Software Engineer at Solo.io, explains that the biggest challenge in Kubernetes security isn't features but adoption.

He shares how Ambient Mesh originated from users repeatedly requesting "just mTLS" without the overhead of full-service mesh implementation. The team designed Ambient Mesh with compatibility and low footprint as core principles - it doesn't modify traffic patterns, works with any application, and can be deployed cluster-wide without breaking existing workloads. John emphasizes that establishing baseline security everywhere must precede implementing advanced features like sophisticated authorization policies or AI-based anomaly detection.

Watch the full episode: https://kube.fmhttps://ku.bz/sk-ZF1PG9

This week on Learn Kubernetes Weekly 190:

🌪️ Taming the Storm: Building Groww's Internal Chaos Engineering Platform
🧪 Taracode Testing a Go-Based CLI AI Agent in My Homelab
🧠 Building self-evolving AI systems: exploring the architecture
🔄 Migrating from slurm to Kubernetes
🏠 Lessons Learnt Self-hosting an AI Assistant

Read it now: https://kube.today/issues/190

⭐️ This newsletter is brought to you by LearnKube — master Kubernetes with hands-on training designed for engineers who want to learn the smart way https://ku.bz/hypSbyc-V

This tutorial shows how to connect on-prem Kubernetes workloads to Google Cloud without service account keys using Workload Identity Federation, OIDC, Terraform, Kyverno, and IAM attribute conditions.

More: https://ku.bz/1YVD6c3FP

Kaniop is a Kubernetes operator written in Rust for managing Kanidm identity management clusters, providing declarative identity management through GitOps workflows.

More: https://ku.bz/pFkd88jPW

Copy Fail Blocker deploys a privileged DaemonSet that blocks AF_ALG and AF_RXRPC socket creation cluster-wide to mitigate Copy Fail and similar Linux kernel privilege-escalation paths.

More: https://ku.bz/hMfdC6WGc

This week on Learn Kubernetes Weekly 191:

🔥 What Does 4.4% GPU Utilization Actually Mean?
🛠️ GKE IP Exhaustion Fixed: The Class E Migration Guide
🧹 Evicting MCP Tool Calls from Your Kubernetes Cluster
🔄 The Feedback Loops Behind Kubernetes
🧠 You Don't Have a GIL Problem — You Have a CPU Problem
Read it now: https://kube.today/issues/191

⭐️ This newsletter is brought to you by LearnKube — master Kubernetes with hands-on training designed for engineers who want to learn the smart way https://ku.bz/hypSbyc-V

This article explains Kubernetes v1.36 fine-grained kubelet authorization and how teams can replace broad nodes/proxy access with safer permissions for metrics, stats, logs, pods, and health checks.

More: https://ku.bz/M6WZq580X

This article explains the Kubernetes v1.36 SELinux volume labeling change and why clusters using SELinux should audit workloads before SELinuxMount becomes the default.

More: https://ku.bz/KGR_FN-3w

k8scout maps realistic Kubernetes escalation paths from a compromised pod to cluster-admin, node access, secret theft, or cloud IAM takeover, with graph output and reviewer mode.

More: https://ku.bz/Jt-LJm0f2

Copy Fail Destroyer runs on Kubernetes nodes to detect and remediate Copy Fail and Dirty Frag by probing vulnerable kernel modules, unloading them, exposing metrics, and supporting Helm or ArgoCD deployment.

More: https://ku.bz/xvFl18wxv

How do you give developers access to the Kubernetes API without letting them break things?

Peter Kelly describes staged policies in Project Calico: developers can dry-run network policies to see which flows would be affected before enforcing them. Combined with namespace-scoped policy tiers, teams get real autonomy without risking production traffic.

Instead of blocking developers, give them a safe way to test.

Full interview: https://ku.bz/xgqZJhdyn



Watch the full interview: https://ku.bz/xgqZJhdyn

This interview is a reaction to Mac Chaffee's episode https://ku.bz/9nFPmG85f

This case study explains how a Kubernetes secrets audit exposed weak secret handling and forced a move toward safer secret management.

It covers encoded secrets, RBAC, encryption, external secret stores, and audit-ready controls.

More: https://ku.bz/z0ylnRsvd

This week on Learn Kubernetes Weekly 192:

🔧 Our Kubernetes Operator Didn't Scale, So We Rebuilt It
🔀 ClickHouse Shard Rebalancing on Kubernetes: From Talk to Operator
💥 Invisible OOMkill: Java Pods Crashing in Kubernetes
🗂️ Using Kubernetes ConfigMaps as a Real-Time State Store
🚨 From Container Escape to Cloud Takeover: A Real-World Cloud Security Assessment

Read it now: https://kube.today/issues/192

⭐️ This newsletter is brought to you by Buoyant, the creators of Linkerd https://ku.bz/BB-RtVFWs

This article explains how a local 7B model was fine-tuned to answer cloud security, Kubernetes, Terraform, and compliance questions from a rule-based dataset.

More: https://ku.bz/NNjhbSslC

This tutorial shows how to modernize Kyverno policies with CEL using practical Kubernetes security examples like namespace rules, image checks, service account tokens, and safer policy testing.

More: https://ku.bz/PcpzWX_N6

Nicholaos Mouzourakis, Staff Product Security Engineer at Gusto, shares practical advice for Kubernetes administrators implementing Open Policy Agent (OPA). He recommends starting with Rego fundamentals using resources like Styra Academy and the Rego Playground for testing policies with real-time feedback.

He emphasizes:

- Getting team buy-in by explaining security benefits, especially for those who will read or write policies
- Using observability tools to determine scale and latency requirements
- Choosing appropriate deployment patterns (sidecar, daemon set, or standalone deployment)
- Deciding between baking configurations into Docker images, using config maps, or pulling policies from services like S3 or Styra DAS

Watch the full episode: https://kube.fmhttps://ku.bz/S-2vQ_j-4

This article explains four Kubernetes isolation patterns for AI agents: no exec, sidecar exec, separate exec pod, and ephemeral job dispatchers, with OpenShift-validated threat modeling.

More: https://ku.bz/KC6H2m-VF

This tutorial explains why standard GKE Ingress breaks under Istio STRICT mTLS and shows how to replace it with an Istio Ingress Gateway, Gateway resource, and VirtualService.

More: https://ku.bz/lNmNzN4HW

This article explains how Kubernetes zero-trust egress policy can contain the Axios npm supply-chain attack by blocking C2 traffic, data exfiltration, and lateral movement from compromised pods.

More: https://ku.bz/kz47HBml6