Exact question counts, passing scores, and allowed reference materials vary by testing window and state/agency. Always confirm current rules with the official testing provider.
Sample questions
A cluster administrator enforces Kubernetes Pod Security Admission (PSA) on namespace pci-workloads using label pod-security.kubernetes.io/enforce: restricted. A deployment fails admission even though the container sets runAsNonRoot: true and readOnlyRootFilesystem: true. Which additional securityContext settings are required to satisfy the PSA restricted profile?
Set privileged: false and drop NET_RAW only, as restricted allows other default capabilities.
Set runAsUser: 0 with seLinuxOptions enabled and remove readOnlyRootFilesystem, because PSA restricted blocks read-only root filesystems.
Mount an AppArmor annotation only; PSA restricted ignores seccomp and privilege escalation.
Set allowPrivilegeEscalation: false, capabilities: { drop: ["ALL"] }, and seccompProfile.type: RuntimeDefault (or Localhost).
Show the answer
D. Set allowPrivilegeEscalation: false, capabilities: { drop: ["ALL"] }, and seccompProfile.type: RuntimeDefault (or Localhost).
The Kubernetes Pod Security Standards `restricted` profile enforces all baseline rules plus strict container hardening: `runAsNonRoot: true` (and non-zero runAsUser if specified), `allowPrivilegeEscalation: false`, `capabilities.drop: ["ALL"]` (only NET_BIND_SERVICE may optionally be re-added), and `seccompProfile.type` set to `RuntimeDefault` or `Localhost`. While readOnlyRootFilesystem: true (paired with an emptyDir for writable scratch paths like /tmp) is a strong CKS defense-in-depth practice, omitting any of the four mandatory restricted fields causes PSA to reject the Pod.
A security team deploys Falco as a DaemonSet on Kubernetes worker nodes to detect runtime container intrusions such as an attacker executing kubectl exec to spawn /bin/sh or reading /etc/shadow. How does Falco observe these events, and what is the correct way to customize alert behavior for a legitimate debugging rule?
Falco inspects kernel syscalls via an eBPF probe or kernel module; place overrides in /etc/falco/falco_rules.local.yaml.
Falco intercepts HTTP requests inside kube-apiserver via a MutatingAdmissionWebhook; edit /etc/falco/falco_rules.yaml directly to persist overrides.
Falco polls docker inspect and /var/log/pods every 60 seconds; disable default rules by removing the containerd socket mount from the DaemonSet.
Falco runs as a sidecar inside each application pod to block syscalls inline using seccomp; customize it by adding annotations to each Deployment.
Show the answer
A. Falco inspects kernel syscalls via an eBPF probe or kernel module; place overrides in /etc/falco/falco_rules.local.yaml.
Falco runs at the node level (typically as a DaemonSet) consuming Linux kernel system calls in real time via a modern eBPF probe (or kernel module) enriched with container runtime and Kubernetes metadata. It detects behavioral anomalies (like spawned_process with container.id != host and proc.name in (bash, sh), or open_read on /etc/shadow) asynchronously without breaking running workloads. Custom rules and override blocks belong in /etc/falco/falco_rules.local.yaml (or rules.d/) so upstream rule updates do not clobber them.