From c570a476b59f789c1964805da2543c784a38ef45 Mon Sep 17 00:00:00 2001
From: MichaelFisher1997 <contact@michaelfisher.tech>
Date: Thu, 23 Apr 2026 02:15:45 +0000
Subject: [PATCH] fix: make helm-based addon kustomizations health-check
 HelmReleases only

These addon Kustomizations were using wait=true, which made Flux treat transient
HelmRepository fetch timeouts as addon failures even when the HelmRelease and
runtime workloads were healthy. Switch the affected Kustomizations to explicit
HelmRelease healthChecks so readiness reflects the actual deployed platform
state instead of repository fetch flakiness.
---
 .../addons/kustomization-nfs-storage.yaml         |  7 ++++++-
 .../addons/kustomization-observability.yaml       | 15 ++++++++++++++-
 .../addons/kustomization-rancher-backup.yaml      |  6 ++++++
 infrastructure/addons/kustomization-rancher.yaml  |  6 ++++++
 4 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/infrastructure/addons/kustomization-nfs-storage.yaml b/infrastructure/addons/kustomization-nfs-storage.yaml
index 9f450bb..67e7a1e 100644
--- a/infrastructure/addons/kustomization-nfs-storage.yaml
+++ b/infrastructure/addons/kustomization-nfs-storage.yaml
@@ -10,6 +10,11 @@ spec:
     kind: GitRepository
     name: platform
   path: ./infrastructure/addons/nfs-storage
-  wait: true
+  wait: false
+  healthChecks:
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: nfs-subdir-external-provisioner
+      namespace: flux-system
   timeout: 10m
   suspend: false
diff --git a/infrastructure/addons/kustomization-observability.yaml b/infrastructure/addons/kustomization-observability.yaml
index 2a2e269..2f9b2f3 100644
--- a/infrastructure/addons/kustomization-observability.yaml
+++ b/infrastructure/addons/kustomization-observability.yaml
@@ -14,6 +14,19 @@ spec:
     - name: addon-external-secrets
     - name: addon-tailscale-operator
     - name: addon-tailscale-proxyclass
-  wait: true
+  wait: false
+  healthChecks:
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: kube-prometheus-stack
+      namespace: flux-system
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: loki
+      namespace: flux-system
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: promtail
+      namespace: flux-system
   timeout: 5m
   suspend: false
diff --git a/infrastructure/addons/kustomization-rancher-backup.yaml b/infrastructure/addons/kustomization-rancher-backup.yaml
index 7dc4a5c..7925370 100644
--- a/infrastructure/addons/kustomization-rancher-backup.yaml
+++ b/infrastructure/addons/kustomization-rancher-backup.yaml
@@ -16,3 +16,9 @@ spec:
   dependsOn:
     - name: addon-external-secrets
     - name: addon-rancher
+  wait: false
+  healthChecks:
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: rancher-backup
+      namespace: flux-system
diff --git a/infrastructure/addons/kustomization-rancher.yaml b/infrastructure/addons/kustomization-rancher.yaml
index 77e8911..329f1f8 100644
--- a/infrastructure/addons/kustomization-rancher.yaml
+++ b/infrastructure/addons/kustomization-rancher.yaml
@@ -18,3 +18,9 @@ spec:
     - name: addon-tailscale-proxyclass
     - name: addon-external-secrets
     - name: addon-cert-manager
+  wait: false
+  healthChecks:
+    - apiVersion: helm.toolkit.fluxcd.io/v2
+      kind: HelmRelease
+      name: rancher
+      namespace: flux-system