Ensure the kubelet read-only port is disabled

How it is checked

Audit method: Shell command

nodes=$(kubectl get nodes -o jsonpath='{range .items[*]}{.metadata.name}{"\n"}{end}') || exit 1
[ -n "$nodes" ] || exit 1
ok=1
for n in $nodes; do
  c=$(kubectl get --raw "/api/v1/nodes/$n/proxy/configz") || exit 1
  if ! { ! printf '%s' "$c" | grep -Eq '"readOnlyPort":[1-9]'; }; then echo "noncompliant: $n"; ok=0; fi
done
if [ "$ok" -eq 1 ]; then echo compliant; fi

Expected output

compliant

How it is fixed

Fix method: Guidance or infrastructure patch

If the audit returns a nonzero readOnlyPort, the worker node is exposing the kubelet read-only service and should be treated as noncompliant. If the field is missing from configz, first confirm whether it is simply omitted because the effective value is 0, since the kubelet configuration schema defines readOnlyPort as the unauthenticated read-only service and states that 0 disables it. Once a noncompliant state is confirmed, remediation in EKS should be performed through the supported node provisioning or node update workflow, not by directly editing the kubelet config file on the host. AWS documents that managed node customization is applied through launch templates and bootstrap logic, and that updating a managed node group to a new launch template version replaces the nodes to match that configuration.

Check this host

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

tmaegis scan --target eks