From 0fb7ac95c02cabfa566a6063939b69078bafdbc9 Mon Sep 17 00:00:00 2001
From: Stefan Prodan <stefan.prodan@gmail.com>
Date: Wed, 10 Apr 2024 19:53:50 +0300
Subject: [PATCH] Enable Helm drift detection

Signed-off-by: Stefan Prodan <stefan.prodan@gmail.com>
---
 README.md                | 6 ++++++
 deploy/apps/podinfo.yaml | 6 ++++++
 2 files changed, 12 insertions(+)

diff --git a/README.md b/README.md
index c8135a7..801b220 100644
--- a/README.md
+++ b/README.md
@@ -141,6 +141,12 @@ ingress-nginx   4.10.0          False           True
 podinfo         6.6.2           False           True
 ```
 
+> [!NOTE]
+> Note that Flux detects changes made directly in-cluster on the objects managed
+> by a HelmRelease and automatically [corrects the drift](https://fluxcd.io/flux/components/helm/helmreleases/#drift-detection).
+> During an incident or for debugging purposes, you can manually suspend the reconciliation
+> of a HelmRelease with `flux suspend hr <name> -n <namespace>`.
+
 ## Customize the workloads
 
 Assuming you want to ship workloads to the production cluster with a different configuration,
diff --git a/deploy/apps/podinfo.yaml b/deploy/apps/podinfo.yaml
index 2a7587a..02e1674 100644
--- a/deploy/apps/podinfo.yaml
+++ b/deploy/apps/podinfo.yaml
@@ -23,6 +23,12 @@ spec:
         kind: HelmRepository
         name: podinfo
   interval: 50m
+  driftDetection:
+    mode: enabled
+    ignore:
+      - paths: ["/spec/replicas"]
+        target:
+          kind: Deployment
   install:
     remediation:
       retries: 3