fix: increase KEDA CloudWatch wait to 300s, make karpenter suite hook retry pod readiness

Author: svennam92

website/docs/fastpaths/developer/keda/test-keda.md

@@ -11,7 +11,7 @@ The command below will run the load generator with:
 - Sending 5 queries per second each
 - Running for a maximum of 10 minutes
 
-```bash hook=keda-pod-scaleout hookTimeout=660 wait=120
+```bash hook=keda-pod-scaleout hookTimeout=660 wait=300
 $ export ALB_HOSTNAME=$(kubectl get ingress ui-auto -n ui -o yaml | yq .status.loadBalancer.ingress[0].hostname)
 $ kubectl run load-generator \
   --image=williamyeh/hey:latest \

website/docs/fastpaths/operator/karpenter/tests/hook-suite.sh

@@ -9,8 +9,18 @@ after() {
   kubectl delete deployment inflate -n other --ignore-not-found
 
   # Wait for Karpenter consolidation to settle and pods to stabilize
+  # Retry the wait since pods may be evicted during consolidation
   sleep 120
-  kubectl wait --for=condition=Ready --timeout=300s pods -l app.kubernetes.io/created-by=eks-workshop -A
+  for i in $(seq 1 3); do
+    if kubectl wait --for=condition=Ready --timeout=120s pods -l app.kubernetes.io/created-by=eks-workshop -A 2>/dev/null; then
+      echo "All pods ready"
+      return 0
+    fi
+    echo "Attempt $i: some pods not ready, waiting for consolidation..."
+    sleep 30
+  done
+  # Final attempt without error suppression
+  kubectl wait --for=condition=Ready --timeout=120s pods -l app.kubernetes.io/created-by=eks-workshop -A
 }
 
 "$@"