Minimize the admission of containers wishing to share the host network namespace

Kako se proverava

Metod provere: Komanda ljuske

out=$(kubectl get pods -A -o go-template='{{range .items}}{{.metadata.namespace}} {{.metadata.name}} {{.spec.hostNetwork}}{{"\n"}}{{end}}') || exit 1
bad=$(printf '%s\n' "$out" | awk -v ex='^(kube-system|kube-public|kube-node-lease)$' 'NF && $1 !~ ex && ($3=="true")')
if [ -z "$bad" ]; then echo compliant; else printf '%s\n' "$bad" | head -n 25; fi

Očekivani izlaz

compliant

Kako se ispravlja

Metod ispravke: Smernica ili zakrpa infrastrukture

Configure Pod Security Admission on workload namespaces so pods that request spec.hostNetwork: true are denied at admission time. Use enforce on workload namespaces with restricted, or at minimum baseline, and reserve any exceptions for narrowly approved platform workloads. After updating namespace enforcement, remove hostNetwork: true from affected manifests and redeploy those workloads so new pods are evaluated under the enforced policy. AWS also notes that pods with hostNetwork: true always have IMDS access on EKS EC2 nodes, which is another reason to minimize this setting. • Label each workload namespace to enforce restricted. kubectl label ns NAMESPACE pod-security.kubernetes.io/enforce=restricted --overwrite • Add warn and audit during rollout if you want visibility into violations, but do not rely on them for remediation because they do not block pod creation.

Proverite ovaj računar

Pokrenite ovo na računaru za koji imate dozvolu za proveru. Skeniranje radi u režimu samo za čitanje.

tmaegis scan --target eks