diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
new file mode 100644
index 0000000..cf0a90d
--- /dev/null
+++ b/.github/workflows/test.yml
@@ -0,0 +1,60 @@
+name: End-to-End Test
+
+on:
+  push:
+    branches: [ main ]
+  pull_request:
+    branches: [ main ]
+  workflow_dispatch:
+
+jobs:
+  end-to-end-tests:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45  # Increased to accommodate recording rules wait time
+    
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      
+      - name: Move Docker data to larger disk
+        run: |
+          echo "=== Disk space ==="
+          df -h / /mnt | grep -E '(Filesystem|/dev)'
+          echo ""
+          echo "=== Stopping Docker ==="
+          sudo systemctl stop docker.socket docker.service containerd.service
+          echo ""
+          echo "=== Preparing /mnt for Docker data ==="
+          sudo mkdir -p /mnt/docker-data/docker /mnt/docker-data/containerd
+          echo ""
+          echo "=== Bind mounting to /mnt ==="
+          sudo mount --bind /mnt/docker-data/docker /var/lib/docker
+          sudo mount --bind /mnt/docker-data/containerd /var/lib/containerd
+          echo ""
+          echo "=== Starting Docker ==="
+          sudo systemctl start containerd.service docker.service
+          echo ""
+          echo "=== Verifying Docker works ==="
+          docker ps
+      
+      - name: Install Nix
+        uses: cachix/install-nix-action@v27
+        with:
+          nix_path: nixpkgs=channel:nixos-unstable
+      
+      - name: Show environment info
+        run: |
+          echo "=== System Info ==="
+          uname -a
+          echo "=== Docker Info ==="
+          docker version
+          echo "=== Available space ==="
+          df -h | head -5
+          echo "=== Nix version ==="
+          nix --version
+      
+      - name: Run test-1-setup.sh (15-20 min)
+        run: nix develop --command ./test-1-setup.sh
+      
+      - name: Run test-2-teardown.sh (10-15 min)
+        run: nix develop --command ./test-2-teardown.sh
diff --git a/monitoring/README.md b/monitoring/README.md
index 5ff75a2..f854ebc 100644
--- a/monitoring/README.md
+++ b/monitoring/README.md
@@ -66,9 +66,23 @@ You can find the dashboard under `Home > Dashboards > grafana > CloudNativePG`.
 > streaming is unavailable, but all other Grafana features work normally
 > when accessed via port-forward.
 
+## CloudNativePG Grafana Dashboard
+
+[CloudNativePG provides a default dashboard](https://cloudnative-pg.io/docs/devel/quickstart#grafana-dashboard) for Grafana in the dedicated [`grafana-dashboards` repository](https://github.com/cloudnative-pg/grafana-dashboards). The CNPG Playground monitoring `setup.sh` automatically installs the CNPG dashboard into grafana. You can also download the file [grafana-dashboard.json](https://github.com/cloudnative-pg/grafana-dashboards/blob/main/charts/cluster/grafana-dashboard.json) and manually import it via the GUI (menu: Dashboards > New > Import). 
+
+### Dependencies
+
+The CNPG Playground monitoring `setup.sh` also installs and configures the dependencies of this dashboard:
+
+1. `node-exporter`: Node-level metrics (CPU, memory, disk, network at the host level)
+2. `kube-state-metrics`: Kubernetes object metrics (pods, deployments, resource requests/limits)
+3. Kubelet/cAdvisor Metrics (via `/metrics/cadvisor`): Container-level metrics (CPU, memory, network, disk I/O)
+4. Canonical **Kubernetes recording rules from [kube-prometheus](https://github.com/prometheus-operator/kube-prometheus)**, which pre-compute common aggregations used by the CloudNativePG dashboard such as `node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate` and `node_namespace_pod_container:container_memory_working_set_bytes` and `namespace_cpu:kube_pod_container_resource_requests:sum`
+
 ## PodMonitor
 
 To enable Prometheus to scrape metrics from your PostgreSQL pods, you must
 create a `PodMonitor` resource as described in the
 [documentation](https://cloudnative-pg.io/documentation/current/monitoring/#creating-a-podmonitor).
-
+ 
+ If a monitoring stack is running, then `demo/setup.sh` will automatically create PodMonitors.
diff --git a/monitoring/kube-state-metrics/deployment.yaml b/monitoring/kube-state-metrics/deployment.yaml
new file mode 100644
index 0000000..98480d5
--- /dev/null
+++ b/monitoring/kube-state-metrics/deployment.yaml
@@ -0,0 +1,93 @@
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: kube-state-metrics
+  namespace: kube-system
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: kube-state-metrics
+rules:
+- apiGroups: ["*"]
+  resources: ["*"]
+  verbs: ["list", "watch"]
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: kube-state-metrics
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: kube-state-metrics
+subjects:
+- kind: ServiceAccount
+  name: kube-state-metrics
+  namespace: kube-system
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: kube-state-metrics
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kube-state-metrics
+spec:
+  clusterIP: None
+  ports:
+  - name: http-metrics
+    port: 8080
+    targetPort: http-metrics
+  - name: telemetry
+    port: 8081
+    targetPort: telemetry
+  selector:
+    app.kubernetes.io/name: kube-state-metrics
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: kube-state-metrics
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kube-state-metrics
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: kube-state-metrics
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: kube-state-metrics
+    spec:
+      serviceAccountName: kube-state-metrics
+      nodeSelector:
+        node-role.kubernetes.io/infra: ""
+      containers:
+      - name: kube-state-metrics
+        image: registry.k8s.io/kube-state-metrics/kube-state-metrics:v2.13.0
+        ports:
+        - name: http-metrics
+          containerPort: 8080
+        - name: telemetry
+          containerPort: 8081
+---
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: kube-state-metrics
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kube-state-metrics
+spec:
+  endpoints:
+  - port: http-metrics
+    interval: 30s
+    honorLabels: true
+  - port: telemetry
+    interval: 30s
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: kube-state-metrics
diff --git a/monitoring/kube-state-metrics/kustomization.yaml b/monitoring/kube-state-metrics/kustomization.yaml
new file mode 100644
index 0000000..3535738
--- /dev/null
+++ b/monitoring/kube-state-metrics/kustomization.yaml
@@ -0,0 +1,6 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+namespace: kube-system
+resources:
+  - deployment.yaml
+
diff --git a/monitoring/node-exporter/kustomization.yaml b/monitoring/node-exporter/kustomization.yaml
new file mode 100644
index 0000000..bc27ffa
--- /dev/null
+++ b/monitoring/node-exporter/kustomization.yaml
@@ -0,0 +1,7 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+resources:
+  - node-exporter.yaml
+  - servicemonitor.yaml
+
+
diff --git a/monitoring/node-exporter/node-exporter.yaml b/monitoring/node-exporter/node-exporter.yaml
new file mode 100644
index 0000000..9cd0d42
--- /dev/null
+++ b/monitoring/node-exporter/node-exporter.yaml
@@ -0,0 +1,73 @@
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: node-exporter
+  namespace: kube-system
+---
+apiVersion: apps/v1
+kind: DaemonSet
+metadata:
+  name: node-exporter
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: node-exporter
+spec:
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: node-exporter
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: node-exporter
+    spec:
+      serviceAccountName: node-exporter
+      hostNetwork: true
+      hostPID: true
+      containers:
+      - name: node-exporter
+        image: quay.io/prometheus/node-exporter:v1.8.2
+        args:
+        - --path.procfs=/host/proc
+        - --path.sysfs=/host/sys
+        - --path.rootfs=/host/root
+        ports:
+        - containerPort: 9100
+          name: metrics
+        volumeMounts:
+        - name: proc
+          mountPath: /host/proc
+          readOnly: true
+        - name: sys
+          mountPath: /host/sys
+          readOnly: true
+        - name: root
+          mountPath: /host/root
+          readOnly: true
+      tolerations:
+      - operator: Exists
+      volumes:
+      - name: proc
+        hostPath:
+          path: /proc
+      - name: sys
+        hostPath:
+          path: /sys
+      - name: root
+        hostPath:
+          path: /
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: node-exporter
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: node-exporter
+spec:
+  clusterIP: None
+  ports:
+  - name: metrics
+    port: 9100
+    targetPort: metrics
+  selector:
+    app.kubernetes.io/name: node-exporter
diff --git a/monitoring/node-exporter/servicemonitor.yaml b/monitoring/node-exporter/servicemonitor.yaml
new file mode 100644
index 0000000..f8eb324
--- /dev/null
+++ b/monitoring/node-exporter/servicemonitor.yaml
@@ -0,0 +1,14 @@
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: node-exporter
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: node-exporter
+spec:
+  endpoints:
+  - port: metrics
+    interval: 30s
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: node-exporter
diff --git a/monitoring/prometheus-instance/deploy_prometheus.yaml b/monitoring/prometheus-instance/deploy_prometheus.yaml
index d948cb4..34e5d01 100644
--- a/monitoring/prometheus-instance/deploy_prometheus.yaml
+++ b/monitoring/prometheus-instance/deploy_prometheus.yaml
@@ -12,6 +12,7 @@ rules:
   resources:
   - nodes
   - nodes/metrics
+  - nodes/proxy
   - services
   - endpoints
   - pods
@@ -44,7 +45,7 @@ roleRef:
 subjects:
 - kind: ServiceAccount
   name: prometheus
-  namespace: default
+  namespace: prometheus-operator
 ---
 
 apiVersion: monitoring.coreos.com/v1
@@ -55,5 +56,9 @@ spec:
   serviceAccountName: prometheus
   podMonitorSelector: {}
   podMonitorNamespaceSelector: {}
+  serviceMonitorSelector: {}
+  serviceMonitorNamespaceSelector: {}
+  ruleSelector: {}
+  ruleNamespaceSelector: {}
   nodeSelector:
     node-role.kubernetes.io/infra: ""
diff --git a/monitoring/prometheus-instance/kustomization.yaml b/monitoring/prometheus-instance/kustomization.yaml
index 831e040..8c30ad8 100644
--- a/monitoring/prometheus-instance/kustomization.yaml
+++ b/monitoring/prometheus-instance/kustomization.yaml
@@ -2,4 +2,6 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 resources:
  - deploy_prometheus.yaml
+ # Fetch upstream recording rules from kube-prometheus
+ - https://raw.githubusercontent.com/prometheus-operator/kube-prometheus/v0.16.0/manifests/kubernetesControlPlane-prometheusRule.yaml
 namespace: prometheus-operator
diff --git a/monitoring/prometheus-instance/servicemonitor-kubelet.yaml b/monitoring/prometheus-instance/servicemonitor-kubelet.yaml
new file mode 100644
index 0000000..452d62b
--- /dev/null
+++ b/monitoring/prometheus-instance/servicemonitor-kubelet.yaml
@@ -0,0 +1,55 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: kubelet
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kubelet
+spec:
+  clusterIP: None
+  ports:
+  - name: https-metrics
+    port: 10250
+---
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: kubelet
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kubelet
+spec:
+  endpoints:
+  - port: https-metrics
+    interval: 30s
+    scheme: https
+    bearerTokenFile: /var/run/secrets/kubernetes.io/serviceaccount/token
+    tlsConfig:
+      insecureSkipVerify: true
+    relabelings:
+    - targetLabel: metrics_path
+      replacement: /metrics
+  - port: https-metrics
+    interval: 30s
+    path: /metrics/cadvisor
+    scheme: https
+    bearerTokenFile: /var/run/secrets/kubernetes.io/serviceaccount/token
+    tlsConfig:
+      insecureSkipVerify: true
+    honorLabels: true
+    relabelings:
+    - targetLabel: metrics_path
+      replacement: /metrics/cadvisor
+  - port: https-metrics
+    interval: 30s
+    path: /metrics/probes
+    scheme: https
+    bearerTokenFile: /var/run/secrets/kubernetes.io/serviceaccount/token
+    tlsConfig:
+      insecureSkipVerify: true
+    relabelings:
+    - targetLabel: metrics_path
+      replacement: /metrics/probes
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: kubelet
diff --git a/monitoring/prometheus-operator/kustomization.yaml b/monitoring/prometheus-operator/kustomization.yaml
index 6660bcd..8688e62 100644
--- a/monitoring/prometheus-operator/kustomization.yaml
+++ b/monitoring/prometheus-operator/kustomization.yaml
@@ -1,5 +1,5 @@
 apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 resources:
- - https://github.com/prometheus-operator/prometheus-operator/
+ - https://github.com/prometheus-operator/prometheus-operator/releases/download/v0.87.1/bundle.yaml
 namespace: prometheus-operator
diff --git a/monitoring/setup.sh b/monitoring/setup.sh
index 120961c..e9ed038 100755
--- a/monitoring/setup.sh
+++ b/monitoring/setup.sh
@@ -41,12 +41,24 @@ for region in "${REGIONS[@]}"; do
     K8S_CLUSTER_NAME=$(get_cluster_name "${region}")
     CONTEXT_NAME=$(get_cluster_context "${region}")
 
-# Deploy the Prometheus operator in the playground Kubernetes clusters
+    echo "🔥 Deploying Prometheus operator..."
     kubectl --context ${CONTEXT_NAME} create ns prometheus-operator || true
     kubectl kustomize ${GIT_REPO_ROOT}/monitoring/prometheus-operator | \
       kubectl --context ${CONTEXT_NAME} apply --force-conflicts --server-side -f -
+    
+    echo "⏳ Waiting for Prometheus CRDs to be ready..."
+    kubectl --context ${CONTEXT_NAME} wait --for condition=established --timeout=60s \
+      crd/servicemonitors.monitoring.coreos.com crd/prometheusrules.monitoring.coreos.com || true
 
-# We make sure that monitoring workloads are deployed in the infrastructure node.
+    echo "📊 Deploying kube-state-metrics..."
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/kube-state-metrics | \
+      kubectl --context ${CONTEXT_NAME} apply --force-conflicts --server-side -f -
+
+    echo "🖥️  Deploying node-exporter..."
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/node-exporter | \
+      kubectl --context ${CONTEXT_NAME} apply --force-conflicts --server-side -f -
+
+    echo "🔧 Deploying Prometheus instance and recording rules via kustomize..."
     kubectl kustomize ${GIT_REPO_ROOT}/monitoring/prometheus-instance | \
         kubectl --context=${CONTEXT_NAME} apply --force-conflicts --server-side -f -
     kubectl --context=${CONTEXT_NAME} -n prometheus-operator \
@@ -54,13 +66,34 @@ for region in "${REGIONS[@]}"; do
       --type='merge' \
       --patch='{"spec":{"template":{"spec":{"tolerations":[{"key":"node-role.kubernetes.io/infra","operator":"Exists","effect":"NoSchedule"}],"nodeSelector":{"node-role.kubernetes.io/infra":""}}}}}'
 
+    echo "🔧 Setting up kubelet metrics scraping..."
+    kubectl --context=${CONTEXT_NAME} apply -f ${GIT_REPO_ROOT}/monitoring/prometheus-instance/servicemonitor-kubelet.yaml
+    NODE_IPS=$(kubectl --context ${CONTEXT_NAME} get nodes -o jsonpath='{range .items[*]}{.status.addresses[?(@.type=="InternalIP")].address}{"\n"}{end}')
+    cat <<EOF | kubectl --context=${CONTEXT_NAME} apply -f -
+apiVersion: v1
+kind: Endpoints
+metadata:
+  name: kubelet
+  namespace: kube-system
+  labels:
+    app.kubernetes.io/name: kubelet
+subsets:
+- addresses:
+$(for ip in ${NODE_IPS}; do echo "  - ip: ${ip}"; done)
+  ports:
+  - name: https-metrics
+    port: 10250
+    protocol: TCP
+EOF
+
     echo "-------------------------------------------------------------"
     echo " 📈 Provisioning Grafana resources for region: ${region}"
     echo "-------------------------------------------------------------"
 
 # Deploying Grafana operator
+    # Pinned to v5.21.3 due to CRD validation bug in v5.21.4+
     kubectl --context ${CONTEXT_NAME} apply --force-conflicts --server-side \
-      -f https://github.com/grafana/grafana-operator/releases/latest/download/kustomize-cluster_scoped.yaml
+      -f https://github.com/grafana/grafana-operator/releases/download/v5.21.3/kustomize-cluster_scoped.yaml
     kubectl --context ${CONTEXT_NAME} -n grafana \
       patch deployment grafana-operator-controller-manager \
       --type='merge' \
diff --git a/monitoring/teardown.sh b/monitoring/teardown.sh
new file mode 100755
index 0000000..d309dc6
--- /dev/null
+++ b/monitoring/teardown.sh
@@ -0,0 +1,89 @@
+#!/usr/bin/env bash
+#
+# This script tears down the Prometheus and Grafana operators from the
+# CloudNativePG playground environment.
+# When run without arguments, it automatically detects all cnpg-playground
+# Kind clusters in your environment and removes the monitoring stack from each.
+# To remove monitoring for specific regions only, pass the region names as arguments.
+#
+#
+# Copyright The CloudNativePG Contributors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+# Source the common setup script
+source $(git rev-parse --show-toplevel)/scripts/common.sh
+
+# --- Main Logic ---
+# Determine regions from arguments, or auto-detect if none are provided
+detect_running_regions "$@"
+
+if [ ${#REGIONS[@]} -eq 0 ]; then
+    echo "🤷 No regions found to tear down monitoring from. Exiting."
+    exit 0
+fi
+
+echo "🔥 Tearing down monitoring from regions: ${REGIONS[*]}"
+
+for region in "${REGIONS[@]}"; do
+    echo "-------------------------------------------------------------"
+    echo " 🔥 Removing monitoring resources from region: ${region}"
+    echo "-------------------------------------------------------------"
+
+    K8S_CLUSTER_NAME=$(get_cluster_name "${region}")
+    CONTEXT_NAME=$(get_cluster_context "${region}")
+
+    # Check if cluster exists
+    if ! kind get clusters | grep -q "^${K8S_CLUSTER_NAME}$"; then
+        echo "⚠️  Cluster '${K8S_CLUSTER_NAME}' not found, skipping region '${region}'"
+        continue
+    fi
+
+    echo "📈 Removing Grafana resources..."
+    # Delete Grafana instance and dashboards
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/grafana/ 2>/dev/null | \
+      kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true -f - 2>/dev/null || true
+
+    # Delete Grafana operator
+    echo "📈 Removing Grafana operator..."
+    kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true \
+      -f https://github.com/grafana/grafana-operator/releases/latest/download/kustomize-cluster_scoped.yaml || true
+
+    echo "🔧 Removing kubelet monitoring resources..."
+    kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true \
+      -f ${GIT_REPO_ROOT}/monitoring/prometheus-instance/servicemonitor-kubelet.yaml 2>/dev/null || true
+    
+    echo "🔥 Removing Prometheus resources..."
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/prometheus-instance 2>/dev/null | \
+      kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true -f - 2>/dev/null || true
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/prometheus-operator 2>/dev/null | \
+      kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true -f - 2>/dev/null || true
+    
+    echo "🖥️  Removing node-exporter..."
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/node-exporter 2>/dev/null | \
+      kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true -f - 2>/dev/null || true
+    
+    echo "📊 Removing kube-state-metrics..."
+    kubectl kustomize ${GIT_REPO_ROOT}/monitoring/kube-state-metrics 2>/dev/null | \
+      kubectl --context ${CONTEXT_NAME} delete --ignore-not-found=true -f - 2>/dev/null || true
+
+    echo "🗑️  Removing prometheus-operator namespace..."
+    kubectl --context ${CONTEXT_NAME} delete namespace prometheus-operator --ignore-not-found=true || true
+
+    echo "✅ Monitoring teardown complete for region: ${region}"
+    echo
+done
+
+echo "✅ Monitoring cleanup complete!"
+
diff --git a/test-1-setup.sh b/test-1-setup.sh
new file mode 100755
index 0000000..41446f7
--- /dev/null
+++ b/test-1-setup.sh
@@ -0,0 +1,155 @@
+#!/usr/bin/env bash
+# End-to-end smoke test for cnpg-playground - Part 1: Setup & Validation
+# Usage: ./test-1-setup.sh  OR  nix develop -c ./test-1-setup.sh
+set -euo pipefail
+
+ROOT="$(git rev-parse --show-toplevel)"
+source "${ROOT}/scripts/common.sh"
+export KUBECONFIG="${KUBE_CONFIG_PATH}"
+PASSED=0 FAILED=0
+
+pass() { echo "✅ $*"; PASSED=$((PASSED+1)); }
+fail() { echo "❌ $*"; FAILED=$((FAILED+1)); return 1; }
+log() { echo "🧪 $*"; }
+
+for cmd in kind kubectl curl jq docker; do command -v "$cmd" >/dev/null || { fail "Missing: $cmd"; exit 1; }; done
+
+cleanup() { pkill -f "kubectl port-forward" || true; }
+trap cleanup EXIT
+
+# Setup: infrastructure → monitoring → PostgreSQL  
+log "Setting up infrastructure (eu, us)...  start at $(date +%H:%M:%S)"
+if "${ROOT}/scripts/setup.sh"; then pass "Infrastructure setup finish at $(date +%H:%M:%S)"; else fail "Infrastructure setup"; exit 1; fi
+
+# Test infrastructure health (MinIO API + Kubernetes API)
+log "Testing infrastructure health..."
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  
+  # Test Kubernetes API
+  kubectl --context "${CTX}" get nodes && \
+    pass "${region}: Kubernetes API responsive" || fail "${region}: Kubernetes API failed"
+  
+  # Test MinIO API
+  MINIO_PORT=$((MINIO_BASE_PORT + $([ "$region" = "us" ] && echo 1 || echo 0)))
+  MINIO_STATUS=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:${MINIO_PORT}/minio/health/live || echo "000")
+  [ "${MINIO_STATUS}" = "200" ] && \
+    pass "${region}: MinIO API responsive" || fail "${region}: MinIO API failed (HTTP ${MINIO_STATUS})"
+done
+
+log "Setting up monitoring stack...  start at $(date +%H:%M:%S)"
+if "${ROOT}/monitoring/setup.sh"; then pass "Monitoring setup finish at $(date +%H:%M:%S)"; else fail "Monitoring setup"; exit 1; fi
+
+# Allow time for pod creation and initial metrics scraping
+log "Waiting 60 seconds for pod creation and initial metrics scrape...  start wait at $(date +%H:%M:%S)"
+sleep 60
+
+# Test monitoring health (Prometheus + Grafana + node/system metrics)
+log "Testing monitoring stack health..."
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  
+  # Wait for Prometheus pod
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l app.kubernetes.io/name=prometheus \
+    -n prometheus-operator --timeout=90s && \
+    pass "${region}: Prometheus pod ready" || fail "${region}: Prometheus pod not ready"
+  
+  # Wait for Grafana pod
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l app=grafana \
+    -n grafana --timeout=90s && \
+    pass "${region}: Grafana pod ready" || fail "${region}: Grafana pod not ready"
+done
+
+for region in eu us; do
+  # Test Prometheus HTTP and metrics
+  CTX=$(get_cluster_context "${region}")
+  PORT=9090; [ "$region" = "us" ] && PORT=9091
+  kubectl port-forward -n prometheus-operator prometheus-prometheus-0 ${PORT}:9090 --context "${CTX}" &
+  sleep 3
+
+  [ "$(curl -s http://localhost:${PORT}/-/ready)" = "Prometheus Server is Ready." ] && \
+    pass "${region}: Prometheus HTTP API" || fail "${region}: Prometheus HTTP failed"
+  
+  # Test node/system metrics (from kubelet/node-exporter)
+  NODE_METRICS=$(curl -s "http://localhost:${PORT}/api/v1/query?query=up%7Bjob%3D%22kubelet%22%7D" | jq -r '.data.result | length' || echo "0")
+  [ "${NODE_METRICS}" -ge 1 ] && \
+    pass "${region}: Kubelet metrics available (${NODE_METRICS})" || fail "${region}: Kubelet metrics missing"
+  
+  CONTAINER_METRICS=$(curl -s "http://localhost:${PORT}/api/v1/query?query=container_cpu_usage_seconds_total" | jq -r '.data.result | length' || echo "0")
+  [ "${CONTAINER_METRICS}" -ge 1 ] && \
+    pass "${region}: Container metrics available (${CONTAINER_METRICS})" || fail "${region}: Container metrics missing"
+  
+  # Test Grafana dashboard
+  GPORT=3000; [ "$region" = "us" ] && GPORT=3001
+  kubectl port-forward -n grafana service/grafana-service ${GPORT}:3000 --context "${CTX}" &
+  sleep 3
+  
+  DASHBOARD_COUNT=$(curl -s -u admin:admin "http://localhost:${GPORT}/api/search?query=cloudnativepg" | jq 'length' 2>/dev/null || echo "0")
+  [ "${DASHBOARD_COUNT}" -ge 1 ] && \
+    pass "${region}: CloudNativePG dashboard exists" || fail "${region}: CloudNativePG dashboard missing"
+done
+
+cleanup
+
+log "Deploying PostgreSQL clusters...  start at $(date +%H:%M:%S)"
+if LEGACY=true "${ROOT}/demo/setup.sh"; then pass "PostgreSQL setup finish at $(date +%H:%M:%S)"; else fail "PostgreSQL setup"; exit 1; fi
+
+# Test PostgreSQL health and metrics
+log "Testing PostgreSQL clusters..."
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  log "Testing region: ${region}"
+  
+  # PostgreSQL readiness
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l cnpg.io/cluster=pg-${region} --timeout=30s && \
+    pass "${region}: PostgreSQL ready" || fail "${region}: PostgreSQL not ready"
+  
+  # SQL query test
+  SQL=$(kubectl --context "${CTX}" exec pg-${region}-1 -- psql -U postgres -tAc "SELECT 1" | tr -d '\r' || echo "")
+  [ "${SQL}" = "1" ] && pass "${region}: SQL query OK" || fail "${region}: SQL query failed"
+done
+
+# Test PostgreSQL metrics in Prometheus (needs time to scrape)
+log "Waiting 60 seconds for PostgreSQL metrics to be scraped...  start wait at $(date +%H:%M:%S)"
+sleep 60  # Increased from 45s to ensure both regions have time to scrape
+
+cleanup  # Clean any lingering port-forwards
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  PORT=9090; [ "$region" = "us" ] && PORT=9091
+  kubectl port-forward -n prometheus-operator prometheus-prometheus-0 ${PORT}:9090 --context "${CTX}" &
+  sleep 3
+  
+  PGMETRICS=$(curl -s "http://localhost:${PORT}/api/v1/query?query=cnpg_collector_up" | jq -r '.data.result | length' || echo "0")
+  [ "${PGMETRICS}" -ge 1 ] && \
+    pass "${region}: PostgreSQL metrics (${PGMETRICS})" || fail "${region}: PostgreSQL metrics missing"
+done
+
+cleanup
+
+# Test recording rules (needs evaluation time and historical data)
+# Recording rules need ~5 minutes of metrics data for irate() to work
+log "Waiting 3.5 minutes for recording rules to evaluate (data + buffer)...  start wait at $(date +%H:%M:%S)"
+sleep 210  # 3.5 minutes wait for sufficient time series data
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  PORT=9090; [ "$region" = "us" ] && PORT=9091
+  
+  kubectl port-forward -n prometheus-operator prometheus-prometheus-0 ${PORT}:9090 --context "${CTX}" &
+  sleep 3
+  
+  # Check if recording rule metrics exist
+  RULE_METRICS=$(curl -s "http://localhost:${PORT}/api/v1/query" --data-urlencode "query=node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate{pod=~\"pg-${region}-.*\",namespace=\"default\",node!=\"\"}" | jq -r '.data.result | length' || echo "0")
+  [ "${RULE_METRICS}" -ge 1 ] && \
+    pass "${region}: Recording rule metrics (${RULE_METRICS})" || fail "${region}: Recording rule metrics missing"
+done
+
+cleanup
+
+# Summary
+log "=========================================="
+log "Results: ${PASSED} passed, ${FAILED} failed"
+[ ${FAILED} -eq 0 ] && { log "✅ All tests PASSED"; exit 0; } || { log "❌ FAILED"; exit 1; }
+
diff --git a/test-2-teardown.sh b/test-2-teardown.sh
new file mode 100755
index 0000000..e7a9349
--- /dev/null
+++ b/test-2-teardown.sh
@@ -0,0 +1,94 @@
+#!/usr/bin/env bash
+# End-to-end smoke test for cnpg-playground - Part 2: Teardown & Recreation
+# Usage: ./test-2-teardown.sh  OR  nix develop -c ./test-2-teardown.sh
+set -euo pipefail
+
+ROOT="$(git rev-parse --show-toplevel)"
+source "${ROOT}/scripts/common.sh"
+export KUBECONFIG="${KUBE_CONFIG_PATH}"
+PASSED=0 FAILED=0
+
+pass() { echo "✅ $*"; PASSED=$((PASSED+1)); }
+fail() { echo "❌ $*"; FAILED=$((FAILED+1)); return 1; }
+log() { echo "🧪 $*"; }
+
+for cmd in kind kubectl curl jq docker; do command -v "$cmd" >/dev/null || { fail "Missing: $cmd"; exit 1; }; done
+
+cleanup() { pkill -f "kubectl port-forward" || true; }
+trap cleanup EXIT
+
+# Pre-flight check: Verify database is running before teardown tests
+log "Pre-flight check: Verifying database is running..."
+for region in eu us; do 
+  CTX=$(get_cluster_context "${region}")
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l cnpg.io/cluster=pg-${region} --timeout=10s && \
+    pass "${region}: PostgreSQL running" || { fail "${region}: PostgreSQL not ready. Run test-1-setup.sh first."; exit 1; }
+done
+
+# Test teardown and recreation of PostgreSQL layer
+log "Tearing down PostgreSQL clusters..."
+"${ROOT}/demo/teardown.sh" && pass "PostgreSQL teardown" || fail "PostgreSQL teardown"
+sleep 10   # extra time for pods to shut down (otherwise, the following test occasionally fails)
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  PGPODS=$(kubectl --context "${CTX}" get pods -l cnpg.io/cluster=pg-${region} --no-headers | wc -l || echo "0")
+  [ "${PGPODS}" -eq 0 ] && pass "${region}: PostgreSQL removed" || fail "${region}: PostgreSQL still exists"
+done
+
+log "Recreating PostgreSQL clusters..."
+if LEGACY=true "${ROOT}/demo/setup.sh"; then pass "PostgreSQL re-setup"; else fail "PostgreSQL re-setup"; exit 1; fi
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l cnpg.io/cluster=pg-${region} --timeout=30s && \
+    pass "${region}: PostgreSQL ready" || fail "${region}: PostgreSQL not ready"
+done
+
+# Test teardown and recreation of monitoring layer
+log "Tearing down monitoring stack..."
+"${ROOT}/monitoring/teardown.sh" && pass "Monitoring teardown" || fail "Monitoring teardown"
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  for ns in prometheus-operator grafana; do
+    ! kubectl --context "${CTX}" get namespace "${ns}" && \
+      pass "${region}: ${ns} removed" || fail "${region}: ${ns} still exists"
+  done
+done
+
+log "Recreating monitoring stack..."
+if "${ROOT}/monitoring/setup.sh"; then pass "Monitoring re-setup"; else fail "Monitoring re-setup"; exit 1; fi
+
+for region in eu us; do
+  CTX=$(get_cluster_context "${region}")
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l app.kubernetes.io/name=prometheus \
+    -n prometheus-operator --timeout=90s && \
+    pass "${region}: Prometheus ready" || fail "${region}: Prometheus not ready"
+  
+  kubectl --context "${CTX}" wait --for=condition=Ready pod -l app=grafana \
+    -n grafana --timeout=90s && \
+    pass "${region}: Grafana ready" || fail "${region}: Grafana not ready"
+done
+
+# Test infrastructure teardown
+log "Tearing down infrastructure..."
+"${ROOT}/scripts/teardown.sh" && pass "Infrastructure teardown" || fail "Infrastructure teardown"
+for region in eu us; do
+  CLUSTER_NAME=$(get_cluster_name "${region}")
+  ! kind get clusters | grep -qx "${CLUSTER_NAME}" && \
+    pass "${region}: Cluster removed" || fail "${region}: Cluster still exists"
+  
+  # Verify MinIO containers are removed
+  MINIO_NAME="${MINIO_BASE_NAME}-${region}"
+  ! docker ps -a --format '{{.Names}}' | grep -qx "${MINIO_NAME}" && \
+    pass "${region}: MinIO container removed" || fail "${region}: MinIO container still exists"
+done
+
+cleanup
+
+# Summary
+log "=========================================="
+log "Results: ${PASSED} passed, ${FAILED} failed"
+[ ${FAILED} -eq 0 ] && { log "✅ All tests PASSED"; exit 0; } || { log "❌ FAILED"; exit 1; }
+