#Kubernetes
141 posts
Certified Kubernetes Security Specialist (CKS) #18: Container immutability, forensics
The eighteenth post in the Certified Kubernetes Security Specialist (CKS) series. We cover the final pieces of runtime security — container immutability and incident response. We work through YAML examples for the pattern of hardening the filesystem to read-only with readOnlyRootFilesystem and opening only the paths that need writes via emptyDir, the immutable operating model that forbids in-place changes and only swaps via redeploy, and the forensics procedure of isolating a compromised Pod with a NetworkPolicy and a node cordon, preserving evidence, and then investigating with kubectl debug.
Kubernetes and Cloud Native Associate (KCNA) #9: Full-Length Practice Exam — 50 Questions with Explanations
The final post of the KCNA series. Fifty questions sized to match the real exam domain weights (Kubernetes Fundamentals 46%, Container Orchestration 22%, Cloud Native Architecture 16%, Observability 8%, Application Delivery 8%), each followed by its answer and an explanation. Score 38 out of 50 (75%) or better and you are in passing territory — go book the exam.
Certified Kubernetes Administrator (CKA) #24 Troubleshooting 3: Control Plane (apiserver/etcd/scheduler Down), etcd Recovery
The twenty-fourth post in the Certified Kubernetes Administrator (CKA) series. Taking the fact that control plane components run as static Pods as the starting point of diagnosis, we organize how to narrow down causes by symptom — from an apiserver down that leaves kubectl unresponsive, to an etcd down, to a scheduler/controller-manager down. We get hands-on with inspecting containers directly via crictl and journalctl, and with fixing manifests so kubelet restarts them.
Certified Kubernetes Application Developer (CKAD) #19 Ingress and NetworkPolicy
The nineteenth post in the Certified Kubernetes Application Developer (CKAD) series. It covers Ingress, which routes external traffic at L7, and NetworkPolicy, which controls Pod-to-Pod communication with a whitelist, from a hands-on exam perspective. We will build everything from host/path routing and pathType, IngressClass, and TLS through to the podSelector-based default deny pattern, with YAML examples.
Certified Kubernetes Security Specialist (CKS) #17 Falco behavioral analysis, audit logs (Runtime)
The 17th post in the Certified Kubernetes Security Specialist (CKS) series. As the core of the final domain — Monitoring, Logging, and Runtime Security — we cover the rule structure of Falco, the syscall-based runtime threat-detection tool, along with writing custom rules and reading its output, then move on to the Kubernetes API audit log: policy levels and stages, apiserver flag configuration, and log analysis — all framed around the tasks that show up on the exam again and again.
Kubernetes and Cloud Native Associate (KCNA) #8: Exam Tips and Common Mistakes
A condensed recap to read one more time right before you walk into the KCNA exam. We cover time management for 60 questions in 90 minutes, the question formats people most often trip over (multiple response, double negatives), pairs of easily confused concepts (Deployment vs StatefulSet, CRI vs CNI vs CSI, HPA vs VPA, and more), techniques for narrowing down the options, a compact per-domain checklist, and a final pre-exam check for online-proctored sessions.
Certified Kubernetes Administrator (CKA) #23 Troubleshooting 2: Nodes and kubelet (NotReady, disk/memory pressure)
The twenty-third post in the Certified Kubernetes Administrator (CKA) series. We follow the diagnostic flow for a node that has dropped to NotReady from start to finish. We read conditions with k describe node, SSH into the node and narrow down the cause with systemctl status kubelet and journalctl -u kubelet, and fix a stopped kubelet, a stopped runtime, a full disk, and memory pressure symptom by symptom. We also cover how to isolate a problem node with cordon and drain.
Certified Kubernetes Application Developer (CKAD) #18 Services: ClusterIP, NodePort, LoadBalancer, ExternalName
The eighteenth post in the Certified Kubernetes Application Developer (CKAD) series. It covers the Service, the object that puts a stable entry point in front of a shifting set of Pods. We learn how selectors and labels pick Pods and how Endpoints are managed automatically, the differences between the four types ClusterIP, NodePort, LoadBalancer, and ExternalName, the distinction between port, targetPort, and nodePort, headless Services and cluster DNS, and finally debugging empty endpoints — all through YAML and kubectl.
Certified Kubernetes Security Specialist (CKS) #16 Admission control: OPA/Gatekeeper, Kyverno
The 16th post in the Certified Kubernetes Security Specialist (CKS) series. We start from how an admission controller intercepts API requests to validate and mutate them, then learn the difference between validating and mutating webhooks, OPA/Gatekeeper's ConstraintTemplate and Constraint with Rego policies, and Kyverno's YAML-based validate/mutate/generate policies through examples. We implement "no latest tag" and "trusted-registry only" in both tools, and wrap up the exam-favorite pattern of rejecting a violating manifest.
Kubernetes and Cloud Native Associate (KCNA) #7: Cloud Native Application Delivery (8%) — GitOps, CI/CD
The seventh post in the KCNA series. We cover the whole of Domain 5: the split between CI and CD and the pipeline stages, the four principles of GitOps (ArgoCD, Flux) and pull-based delivery, deployment strategies (rolling, blue-green, canary), manifest management with Helm and Kustomize, and the basics of supply chain security.
Certified Kubernetes Administrator (CKA) #22 Troubleshooting 1: Pods and Apps (Pending, CrashLoop, ImagePull, OOM)
The largest domain on the CKA exam is Troubleshooting (30%). In this post we tackle the four most common Pod-level failures within that domain. We organize Pending, CrashLoopBackOff, ImagePullBackOff/ErrImagePull, and OOMKilled each from symptom to diagnosis to fix in a single flow, and we drill the order in which to read describe, events, and logs so you reach the root cause fast.
Certified Kubernetes Application Developer (CKAD) #17 Volumes: emptyDir, PVC, projected, ephemeral
The seventeenth post in the Certified Kubernetes Application Developer (CKAD) series. Starting from the volatility of a container filesystem, it works through emptyDir and hostPath, the dynamic provisioning of PersistentVolumeClaim and StorageClass, the projected volume that bundles secret, configMap, and downwardAPI into one directory, and the generic ephemeral volume — all with hands-on YAML examples.