Ensure Integrity Monitoring For Shielded Gke Nodes Is Enabled

Manual Steps

Identify node pools and check Integrity Monitoring status

Run on: any machine with gcloud configured.

PROJECT_ID="<your-project-id>"
CLUSTER_NAME="<your-cluster-name>"
LOCATION="<cluster-location>"   # e.g. us-central1, us-central1-a

gcloud container node-pools list \
  --cluster "$CLUSTER_NAME" \
  --location "$LOCATION" \
  --project "$PROJECT_ID"

For each node pool name from the output:

POOL_NAME="<node-pool-name>"

gcloud container node-pools describe "$POOL_NAME" \
  --cluster "$CLUSTER_NAME" \
  --location "$LOCATION" \
  --project "$PROJECT_ID" \
  --format json | jq .config.shieldedInstanceConfig

Review: enableIntegrityMonitoring should be true for compliant pools.

Decide which node pools must be protected
- Based on your security policy, decide whether all node pools, or at least those running sensitive workloads (e.g., production, regulated, or Internet-exposed apps), must have Shielded Integrity Monitoring enabled.
- Document any justified exceptions (e.g., specialized images that are incompatible).
Plan migration from non-compliant node pools
- For each non-compliant node pool you intend to remediate, list workloads scheduled there:
# On any machine with kubectl access kubectl get pods -A -o wide | grep "<non-compliant-node-pool-name-fragment>" || true
- Decide a maintenance window and verify PodDisruptionBudgets, autoscaling, and capacity to ensure workloads can move to a new pool without unacceptable downtime.
Create new compliant node pools with Integrity Monitoring enabled
- For each target pool you are replacing, create a new one with Shielded Integrity Monitoring turned on:
NEW_POOL_NAME="<new-node-pool-name>" gcloud container node-pools create "$NEW_POOL_NAME" \ --cluster "$CLUSTER_NAME" \ --location "$LOCATION" \ --project "$PROJECT_ID" \ --shielded-integrity-monitoring
- Optionally add flags (e.g. --machine-type, --num-nodes, --node-locations, --enable-autoscaling) to match your existing pool’s configuration.

Migrate workloads and delete old node pools

Cordon and drain old nodes so workloads move to the new, compliant pool:

# Identify nodes belonging to the old pool
kubectl get nodes --show-labels | grep "<old-node-pool-name>" || true

# For each such node:
NODE_NAME="<node-name>"

kubectl cordon "$NODE_NAME"
kubectl drain "$NODE_NAME" --ignore-daemonsets --delete-emptydir-data

After confirming workloads are healthy on the new pool, delete the old non-compliant node pool:

gcloud container node-pools delete "$POOL_NAME" \
  --cluster "$CLUSTER_NAME" \
  --location "$LOCATION" \
  --project "$PROJECT_ID"

Verify remediation (re-run the audit)

Re-check each remaining node pool:

for POOL_NAME in $(gcloud container node-pools list \
    --cluster "$CLUSTER_NAME" \
    --location "$LOCATION" \
    --project "$PROJECT_ID" \
    --format="value(name)"); do
  echo "=== $POOL_NAME ==="
  gcloud container node-pools describe "$POOL_NAME" \
    --cluster "$CLUSTER_NAME" \
    --location "$LOCATION" \
    --project "$PROJECT_ID" \
    --format json | jq .config.shieldedInstanceConfig
done

Confirm that all node pools you decided must be protected show "enableIntegrityMonitoring": true.

Using kubectl

kubectl cannot enable Integrity Monitoring for Shielded GKE nodes because this setting is configured on the managed control plane via GCP (gcloud, console, or IaC), not via Kubernetes objects. Please follow the guidance in the Manual Steps section to recreate/migrate node pools with --shielded-integrity-monitoring enabled.

Automation

#!/usr/bin/env bash
#
# Report Shielded GKE node pools and Integrity Monitoring status for all clusters
# in a given project. Requires:
#   - gcloud configured and authenticated
#   - jq installed
#
# Usage:
#   PROJECT_ID="my-project" LOCATION="us-central1" ./check_shielded_integrity_monitoring.sh
#

set -euo pipefail

PROJECT_ID="${PROJECT_ID:-}"
LOCATION="${LOCATION:-}"   # Optional: if empty, checks all regions/zones

if [[ -z "$PROJECT_ID" ]]; then
  echo "ERROR: PROJECT_ID environment variable must be set" >&2
  exit 1
fi

echo "Project: $PROJECT_ID"
if [[ -n "$LOCATION" ]]; then
  echo "Location filter: $LOCATION"
fi
echo

# List all GKE clusters in the project (and optionally a given location)
if [[ -n "$LOCATION" ]]; then
  clusters_json="$(gcloud container clusters list \
    --project "$PROJECT_ID" \
    --location "$LOCATION" \
    --format=json)"
else
  clusters_json="$(gcloud container clusters list \
    --project "$PROJECT_ID" \
    --format=json)"
fi

if [[ "$(echo "$clusters_json" | jq 'length')" -eq 0 ]]; then
  echo "No GKE clusters found."
  exit 0
fi

echo "Cluster,Location,NodePool,ShieldedVMEnabled,IntegrityMonitoringEnabled,STATUS"
echo "-------,--------,--------,-----------------,---------------------------,------"

echo "$clusters_json" | jq -r '.[] | [.name, .location] | @tsv' | while IFS=$'\t' read -r CLUSTER_NAME CLUSTER_LOCATION; do
  # List node pools for this cluster
  node_pools_json="$(gcloud container node-pools list \
    --cluster "$CLUSTER_NAME" \
    --location "$CLUSTER_LOCATION" \
    --project "$PROJECT_ID" \
    --format=json)"

  if [[ "$(echo "$node_pools_json" | jq 'length')" -eq 0 ]]; then
    echo "${CLUSTER_NAME},${CLUSTER_LOCATION},<none>,N/A,N/A,NO_NODE_POOLS"
    continue
  fi

  echo "$node_pools_json" | jq -r '.[].name' | while read -r POOL_NAME; do
    np_json="$(gcloud container node-pools describe "$POOL_NAME" \
      --cluster "$CLUSTER_NAME" \
      --location "$CLUSTER_LOCATION" \
      --project "$PROJECT_ID" \
      --format=json)"

    shielded_config="$(echo "$np_json" | jq -r '.config.shieldedInstanceConfig // empty')"

    if [[ -z "$shielded_config" || "$shielded_config" == "null" ]]; then
      shielded_vm="false"
      integrity_monitoring="false"
    else
      shielded_vm="$(echo "$shielded_config" | jq -r '.enableSecureBoot // false')"
      integrity_monitoring="$(echo "$shielded_config" | jq -r '.enableIntegrityMonitoring // false')"
    fi

    status="OK"
    # Problem conditions:
    # - Shielded VM not enabled
    # - Integrity Monitoring not enabled
    if [[ "$shielded_vm" != "true" || "$integrity_monitoring" != "true" ]]; then
      status="NON_COMPLIANT"
    fi

    echo "${CLUSTER_NAME},${CLUSTER_LOCATION},${POOL_NAME},${shielded_vm},${integrity_monitoring},${status}"
  done
done

cat <<'EOF'

Interpretation:

- NON_COMPLIANT rows indicate a problem for this control:
  * ShieldedVMEnabled != true  -> node pool is not using Shielded GKE nodes.
  * IntegrityMonitoringEnabled != true -> Integrity Monitoring is not enabled.

- To meet CIS GKE 5.5.6:
  * Every node pool that should run sensitive workloads MUST have
    IntegrityMonitoringEnabled=true (and be ShieldedVMEnabled=true).
  * For any NON_COMPLIANT node pool, plan to:
    - Create a new node pool with --shielded-integrity-monitoring
    - Migrate workloads
    - Delete the non-conforming node pool

This script only reports state; remediation must be planned and applied manually.
EOF

GCP Introduction

GCP Pricing

GCP Threats

GCP Misconfigurations

Resources

GCP Threats

Ensure Integrity Monitoring For Shielded Gke Nodes Is Enabled

More Info:

Risk Level

Address

Compliance Standards

Triage and Remediation

Remediation

Additional Reading:

​More Info:

​Risk Level

​Address

​Compliance Standards

​Triage and Remediation

​Remediation

​Additional Reading: