fix(e2e): add cluster diagnostics to deploy step for faster triage

Add pre-deployment node/namespace/resource diagnostics and wrap
kubectl apply in explicit error handling with cluster state dump on
failure. This gives us actionable output in the GitHub Actions logs
when the Deploy E2E step fails, instead of a silent exit code.

PRI-956

Co-Authored-By: Paperclip <noreply@paperclip.ing>

.github/workflows/e2e.yaml

@@ -16,8 +16,9 @@ concurrency:
 
 jobs:
   e2e:
-    uses: privilegedescalation/.github/.github/workflows/plugin-e2e.yaml@hugh/add-pnpm-support-plugin-e2e
+    uses: privilegedescalation/.github/.github/workflows/plugin-e2e.yaml@main
     with:
-      node-version: '22'
+      node-version: "22"
       headlamp-version: v0.40.1
       e2e-namespace: headlamp-dev
+      plugin-name: rook

e2e/rook.spec.ts

@@ -24,14 +24,14 @@ test.describe('Rook plugin smoke tests', () => {
 
     await page.waitForLoadState('networkidle');
     await expect(page).toHaveURL(/rook-ceph/);
-    await expect(page.getByRole('heading', { name: /overview/i })).toBeVisible();
+    await expect(page.getByRole('heading', { name: /overview/i }).first()).toBeVisible();
   });
 
   test('overview page renders content', async ({ page }) => {
     await page.goto('/c/main/rook-ceph');
     await waitForSidebar(page);
 
-    await expect(page.getByRole('heading', { name: /overview/i })).toBeVisible({
+    await expect(page.getByRole('heading', { name: /overview/i }).first()).toBeVisible({
       timeout: 15_000,
     });
 
@@ -42,22 +42,27 @@ test.describe('Rook plugin smoke tests', () => {
 
   test('navigation to storage classes view works', async ({ page }) => {
     await page.goto('/c/main/rook-ceph');
-
     const sidebar = page.getByRole('navigation', { name: 'Navigation' });
+
+    const rookBtn = sidebar.getByRole('button', { name: /rook/i });
+    await rookBtn.click();
+    await page.waitForLoadState('networkidle');
+
     const storageClassesLink = sidebar.getByRole('link', { name: /storage classes/i });
     await expect(storageClassesLink).toBeVisible({ timeout: 10_000 });
     await storageClassesLink.click();
 
     await page.waitForLoadState('networkidle');
     await expect(page).toHaveURL(/rook-ceph\/storage-classes/);
-    await expect(page.getByRole('heading', { name: /storage class/i })).toBeVisible({ timeout: 15_000 });
+    await expect(page.getByRole('heading', { name: /storage class/i }).first()).toBeVisible({ timeout: 15_000 });
   });
 
   test('plugin settings page shows rook plugin entry', async ({ page }) => {
     await page.goto('/settings/plugins');
     await page.waitForLoadState('networkidle');
+    await page.waitForSelector('table, [class*="PluginList"], [class*="plugin"]', { timeout: 10_000 }).catch(() => {});
 
-    const pluginEntry = page.locator('text=rook').first();
+    const pluginEntry = page.locator('text=/rook/i').first();
     await expect(pluginEntry).toBeVisible({ timeout: 30_000 });
   });
 });

scripts/deploy-e2e-headlamp.sh

@@ -35,6 +35,17 @@ if ! kubectl auth can-i delete configmaps -n "$E2E_NAMESPACE" --quiet 2>/dev/nul
   exit 1
 fi
 
+echo ""
+echo "=== Pre-deployment cluster diagnostics ==="
+echo "Nodes:"
+kubectl get nodes -o wide 2>&1 || true
+echo ""
+echo "headlamp-dev namespace state:"
+kubectl get ns headlamp-dev -o yaml 2>&1 || true
+echo ""
+echo "Existing E2E resources in namespace:"
+kubectl get all -n "$E2E_NAMESPACE" -l "app.kubernetes.io/instance=$E2E_RELEASE" 2>&1 || true
+
 echo "=== E2E Headlamp Deployment ==="
 echo "  Image:     ghcr.io/headlamp-k8s/headlamp:${HEADLAMP_VERSION}"
 echo "  Namespace: $E2E_NAMESPACE"
@@ -60,7 +71,7 @@ kubectl delete serviceaccount "${E2E_RELEASE}" -n "$E2E_NAMESPACE" --ignore-not-
 echo ""
 echo "Deploying Headlamp E2E instance..."
 
-kubectl apply -f - <<EOF
+if ! kubectl apply -f - <<EOF
 apiVersion: v1
 kind: ServiceAccount
 metadata:
@@ -113,7 +124,7 @@ spec:
               port: http
             initialDelaySeconds: 5
             periodSeconds: 5
-            failureThreshold: 6
+              failureThreshold: 6
           livenessProbe:
             httpGet:
               path: /
@@ -148,6 +159,12 @@ spec:
       targetPort: http
       protocol: TCP
 EOF
+then
+  echo "ERROR: kubectl apply failed. Dumping cluster state..." >&2
+  kubectl get all -n "$E2E_NAMESPACE" 2>&1 || true
+  kubectl get events -n "$E2E_NAMESPACE" --sort-by='.lastTimestamp' 2>&1 | tail -30 || true
+  exit 1
+fi
 
 echo "Waiting for rollout..."
 kubectl rollout status "deployment/${E2E_RELEASE}" \