fix: retry health checks after rancher webhook startup

.gitea/workflows/deploy.yml

@@ -1137,21 +1137,67 @@ jobs:
           cat >"${health_script}" <<'EOF'
           #!/usr/bin/env bash
           set -euo pipefail
+
+          wait_for_rancher_webhook_if_present() {
+            local elapsed=0
+            local endpoint_ip
+
+            if ! kubectl -n cattle-system get svc/rancher-webhook >/dev/null 2>&1; then
+              return 0
+            fi
+
+            kubectl -n cattle-system rollout status deployment/rancher-webhook --timeout=300s
+            while [ "${elapsed}" -lt 300 ]; do
+              endpoint_ip="$(kubectl -n cattle-system get endpoints/rancher-webhook -o jsonpath='{.subsets[0].addresses[0].ip}' 2>/dev/null || true)"
+              if [ -n "${endpoint_ip}" ]; then
+                return 0
+              fi
+              sleep 10
+              elapsed=$((elapsed + 10))
+            done
+
+            kubectl -n cattle-system describe svc/rancher-webhook || true
+            kubectl -n cattle-system get endpoints/rancher-webhook -o yaml || true
+            return 1
+          }
+
+          wait_for_kustomization_ready() {
+            local name="$1"
+            local timeout="$2"
+            local message
+
+            if kubectl -n flux-system wait --for=condition=Ready "kustomization/${name}" --timeout="${timeout}"; then
+              return 0
+            fi
+
+            message="$(kubectl -n flux-system get "kustomization/${name}" -o jsonpath='{.status.conditions[?(@.type=="Ready")].message}' 2>/dev/null || true)"
+            if printf '%s' "${message}" | grep -q 'rancher-webhook'; then
+              echo "Kustomization ${name} is blocked by Rancher webhook admission; waiting for webhook endpoints and retrying"
+              wait_for_rancher_webhook_if_present
+              kubectl -n flux-system annotate "kustomization/${name}" reconcile.fluxcd.io/requestedAt="$(date +%s)" --overwrite >/dev/null
+              kubectl -n flux-system wait --for=condition=Ready "kustomization/${name}" --timeout="${timeout}"
+              return 0
+            fi
+
+            kubectl -n flux-system describe "kustomization/${name}" || true
+            return 1
+          }
+
           kubectl get nodes -o wide
           kubectl -n flux-system get gitrepositories,kustomizations,helmreleases,ocirepositories
-          kubectl -n flux-system wait --for=condition=Ready kustomization/infrastructure --timeout=300s
+          wait_for_kustomization_ready infrastructure 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-cert-manager --timeout=300s
+          wait_for_kustomization_ready addon-cert-manager 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-external-secrets --timeout=300s
+          wait_for_kustomization_ready addon-external-secrets 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-external-secrets-store --timeout=300s
+          wait_for_kustomization_ready addon-external-secrets-store 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-nfs-storage --timeout=300s
+          wait_for_kustomization_ready addon-nfs-storage 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-tailscale-operator --timeout=300s
+          wait_for_kustomization_ready addon-tailscale-operator 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-tailscale-proxyclass --timeout=300s
+          wait_for_kustomization_ready addon-tailscale-proxyclass 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-rancher-secrets --timeout=300s
+          wait_for_kustomization_ready addon-rancher-secrets 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-rancher --timeout=900s
+          wait_for_kustomization_ready addon-rancher 900s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-rancher-config --timeout=300s
+          wait_for_kustomization_ready addon-rancher-config 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-observability-secrets --timeout=300s
+          wait_for_kustomization_ready addon-observability-secrets 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-observability --timeout=300s
+          wait_for_kustomization_ready addon-observability 300s
-          kubectl -n flux-system wait --for=condition=Ready kustomization/addon-observability-content --timeout=300s
+          wait_for_kustomization_ready addon-observability-content 300s
           kubectl -n flux-system wait --for=condition=Ready helmrelease --all --timeout=600s
           kubectl annotate storageclass local-path storageclass.kubernetes.io/is-default-class=false --overwrite
           kubectl annotate storageclass flash-nfs storageclass.kubernetes.io/is-default-class=true --overwrite