[{"id":"kubernetesbeginner-001","courseId":"kubernetes-beginner","domain":"workloads","topic":"Pods as the smallest deployable unit","prompt":"You want to run one nginx container on your cluster. What is the smallest unit that Kubernetes schedules and manages for it?","hint":"Think about the wrapper Kubernetes puts around one or more containers.","explain":"Kubernetes does not schedule containers one by one. It places them inside a wrapper that shares network and storage, and that wrapper is what lands on a node. Think about which object is that wrapper.","source":"https://kubernetes.io/docs/concepts/workloads/pods/","interaction":{"type":"single"},"options":[{"id":"a","text":"The nginx container on its own","whyNot":"Kubernetes manages Pods rather than managing containers directly, so a container is not the unit it schedules."},{"id":"b","text":"The node that runs nginx","whyNot":"A node is a machine that hosts Pods. It is where Pods run, not the deployable unit."},{"id":"c","text":"A Pod that wraps the nginx container","why":"A Pod is a group of one or more containers with shared network and storage. It is the smallest deployable unit, so Kubernetes manages Pods."},{"id":"d","text":"A Deployment that names nginx","whyNot":"A Deployment manages Pods on your behalf. It is a higher-level object, so it is not the smallest unit."}],"answer":"c","explanation":"A Pod is a group of one or more containers with shared storage and network resources; it is the smallest deployable unit Kubernetes creates and manages.","concept":"A Pod is a group of one or more containers with shared storage and network resources; it is the smallest deployable unit Kubernetes creates and manages.","rev":1,"updatedAt":"2026-10-23","kind":"scenario","takeaways":["Kubernetes schedules Pods, not individual containers. A Pod may hold one container or several tightly coupled ones."]},{"id":"kubernetesbeginner-002","courseId":"kubernetes-beginner","domain":"workloads","topic":"Deployment declares desired state","prompt":"You want three identical web Pods kept running, and you want to roll out a new image without downtime. Which object should you create?","hint":"Pick the object built to keep a set number of replicas running and to manage updates over time.","explain":"Workload objects serve different goals: some run tasks to completion, some run one Pod per node, and some keep a chosen number of replicas running. Think about which fits a long-running app you update.","source":"https://kubernetes.io/docs/concepts/workloads/controllers/deployment/","interaction":{"type":"single"},"options":[{"id":"a","text":"A Deployment","why":"A Deployment declares the desired state, manages ReplicaSets that keep the Pods running, and provides declarative updates to Pods."},{"id":"b","text":"Three bare Pods, each created by hand","whyNot":"Bare Pods have no controller to keep a count of three or to replace them. Usually you create Pods through workload resources."},{"id":"c","text":"A Job","whyNot":"A Job runs Pods to completion, which suits tasks that finish, not a web server that stays up."},{"id":"d","text":"A DaemonSet","whyNot":"A DaemonSet runs one Pod per node, so its count follows the number of nodes, not three."}],"answer":"a","explanation":"A Deployment is the usual way to run replicated Pods; it manages ReplicaSets and provides declarative updates, so you rarely manage ReplicaSets or Pods directly.","concept":"A Deployment is the usual way to run replicated Pods; it manages ReplicaSets and provides declarative updates, so you rarely manage ReplicaSets or Pods directly.","rev":1,"updatedAt":"2026-10-23","kind":"scenario","takeaways":["Define a Deployment for a long-running app and let it manage the ReplicaSets and Pods under it."]},{"id":"kubernetesbeginner-003","courseId":"kubernetes-beginner","domain":"workloads","topic":"ReplicaSet keeps a set of replica pods","prompt":"A teammate creates a ReplicaSet named frontend from a manifest and then inspects it. Put these steps in the order the ReplicaSet docs show.","hint":"Each command needs the thing it works on to exist first.","explain":"A ReplicaSet is created from a manifest that you submit to the cluster. Kubectl offers commands that list it and show its details. Think about what each step needs before it can work.","source":"https://kubernetes.io/docs/concepts/workloads/controllers/replicaset/","interaction":{"type":"sort","items":[{"text":"Save the manifest as frontend.yaml","why":"The file must exist before anything can be submitted from it."},{"text":"kubectl apply -f frontend.yaml","why":"Submitting the manifest creates the ReplicaSet and the Pods it manages."},{"text":"kubectl get rs","why":"Listing shows the ReplicaSet with its desired, current and ready counts, once it exists."},{"text":"kubectl describe rs/frontend","why":"Describing one ReplicaSet shows its detailed state, including Pod status, after it has been created."}]},"options":[],"answer":"o:abcd","explanation":"A ReplicaSet is created by submitting its manifest; kubectl get rs lists it and kubectl describe rs/NAME shows its details.","concept":"A ReplicaSet is created by submitting its manifest; kubectl get rs lists it and kubectl describe rs/NAME shows its details.","rev":1,"updatedAt":"2026-10-23","kind":"scenario","takeaways":["Use kubectl get rs for a quick count of desired, current and ready replicas, and describe for details."]},{"id":"kubernetesbeginner-004","courseId":"kubernetes-beginner","domain":"workloads","topic":"Labels and label selectors","prompt":"A Service should find only the Pods labeled tier: frontend. Which TWO statements about labels and label selectors are true?","hint":"Think about what labels are for, and what the page says they do not do.","explain":"Labels are key/value pairs attached to objects so you can group and pick out subsets of them. Selectors use those pairs to match objects. Consider what labels promise and what they leave open.","source":"https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/","interaction":{"type":"multiple","tellCount":true},"options":[{"id":"a","text":"Each key must be unique on a given object","why":"An object holds a set of key/value labels, and each key may appear only once on that object."},{"id":"b","text":"Label keys must be unique across the whole cluster","whyNot":"Labels do not provide uniqueness. Many objects are expected to carry the same labels."},{"id":"c","text":"Labels can be added or changed after an object is created","why":"Labels can be attached at creation time and modified at any time afterwards."},{"id":"d","text":"Non-identifying notes belong in labels so they stay queryable","whyNot":"Non-identifying information belongs in annotations. Labels are for identifying and selecting objects."},{"id":"e","text":"A label key with no prefix is shared by every user","whyNot":"A key without a prefix is presumed private to the user. The kubernetes.io and k8s.io prefixes are reserved for core components."},{"id":"f","text":"Label values can be up to 253 characters long","whyNot":"Label values are 63 characters or less. The 253 limit applies to the DNS subdomain used as a key prefix."}],"answer":"ac","explanation":"Labels are key/value pairs for selecting subsets of objects. Each key is unique per object, labels can change at any time, and they do not give cluster-wide uniqueness.","concept":"Labels are key/value pairs for selecting subsets of objects. Each key is unique per object, labels can change at any time, and they do not give cluster-wide uniqueness.","rev":1,"updatedAt":"2026-10-23","kind":"scenario","takeaways":["Labels select and group objects. Put non-identifying data in annotations instead."]},{"id":"kubernetesbeginner-005","courseId":"kubernetes-beginner","domain":"workloads","topic":"Namespaces divide cluster resources","prompt":"Team A and team B each create a Deployment named api, in namespace team-a and namespace team-b. Does the second create fail?","hint":"Ask where a resource name has to be unique: across the cluster, or somewhere smaller.","explain":"Namespaces are scopes within one cluster. Objects in different namespaces are kept apart, while cluster-wide objects like nodes are not placed in any namespace. Think about where a name must be unique.","source":"https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces/","interaction":{"type":"single"},"options":[{"id":"a","text":"No, names only need to be unique within one namespace","why":"Resource names must be unique within a namespace, not across namespaces, so both Deployments can exist."},{"id":"b","text":"Yes, names must be unique across the whole cluster","whyNot":"Name uniqueness is scoped to a namespace, so the same name can appear in another namespace."},{"id":"c","text":"Only if the two Deployments use the same labels","whyNot":"Labels do not affect name uniqueness. The naming rule comes from namespaces."},{"id":"d","text":"Yes, if one of them is in the default namespace","whyNot":"The default namespace is only a starting point. It does not change how names are scoped."}],"answer":"a","explanation":"Resource names must be unique within a namespace but not across namespaces; namespace scoping applies only to namespaced objects, not cluster-wide ones like Nodes.","concept":"Resource names must be unique within a namespace but not across namespaces; namespace scoping applies only to namespaced objects, not cluster-wide ones like Nodes.","rev":1,"updatedAt":"2026-10-23","kind":"scenario","takeaways":["Names repeat across namespaces but not within one. Use namespaces for teams or projects, not for minor version differences."]}]