Minimize the admission of privileged containers

How it is checked

Audit method: Shell command

out=$(kubectl get pods -A -o go-template='{{range .items}}{{$ns:=.metadata.namespace}}{{$p:=.metadata.name}}{{range .spec.containers}}{{$ns}} {{$p}} {{.name}} {{with .securityContext}}{{.privileged}}{{end}}{{"\n"}}{{end}}{{range .spec.initContainers}}{{$ns}} {{$p}} {{.name}} {{with .securityContext}}{{.privileged}}{{end}}{{"\n"}}{{end}}{{range .spec.ephemeralContainers}}{{$ns}} {{$p}} {{.name}} {{with .securityContext}}{{.privileged}}{{end}}{{"\n"}}{{end}}{{end}}') || exit 1
bad=$(printf '%s\n' "$out" | awk -v ex='^(kube-system|kube-public|kube-node-lease)$' 'NF && $1 !~ ex && ($4=="true")')
if [ -z "$bad" ]; then echo compliant; else printf '%s\n' "$bad" | head -n 25; fi

Expected output

compliant

How it is fixed

Fix method: Guidance or infrastructure patch

Configure Pod Security Admission on workload namespaces so privileged containers are denied when new pods are created. In EKS, Pod Security Standards are applied through namespace labels, and current AWS guidance is to enforce at least baseline cluster-wide and restricted for workload namespaces. For this control, both baseline and restricted disallow privileged containers, but restricted is typically the stronger choice for user workloads. Only enforce blocks noncompliant pods. warn and audit are useful during rollout, but they do not remediate the issue by themselves. • Label each workload namespace to enforce restricted. kubectl label ns NAMESPACE pod-security.kubernetes.io/enforce=restricted --overwrite • Add warn and audit labels during transition so violations are surfaced before or alongside enforcement.

Check this host

Run this on a machine you are allowed to audit. The scan is read-only.

tmaegis scan --target eks