Fix: expose image config for fluentd-forwarder

Author: lunkan93

changelog/0.44.md

@@ -29,6 +29,7 @@ Released 2025-02-12
 - [#2348](https://github.com/elastisys/compliantkubernetes-apps/pull/2348) - apps sc: Added possibility to add additional receivers for alertmanager [@Xartos](https://github.com/Xartos)
 - [#2405](https://github.com/elastisys/compliantkubernetes-apps/pull/2405) - apps: Safeguard to prevent accidental cluster deletion [@Eliastisys](https://github.com/Eliastisys)
 - [#2425](https://github.com/elastisys/compliantkubernetes-apps/pull/2425) - Add GPU operator chart [@Ajarmar](https://github.com/Ajarmar)
+- [aab754c](https://github.com/elastisys/compliantkubernetes-apps/commit/aab754c3b68bfd9e3849f491c649bf8e0090047b) - Expose fluentd-forwarder image config [@lunkan93](https://github.com/lunkan93)
 
 ### Improvement(s)
 

config/common-config.yaml

@@ -255,6 +255,9 @@ fluentd:
     affinity: {}
 
   forwarder:
+    image:
+      repository: ghcr.io/elastisys/fluentd-forwarder
+      tag: v4.7.5-ck8s1
     buffer:
       totalLimitSize: 20GB
 

config/schemas/config.yaml

@@ -5301,6 +5301,17 @@ properties:
         type: object
         additionalProperties: false
         properties:
+          image:
+            title: Fluentd Forwarder Image Config
+            description: Configure Fluentd forwarder image repository and tag
+            type: object
+            properties:
+              repository:
+                default: ghcr.io/elastisys/fluentd-forwarder
+                type: string
+              tag:
+                default: v4.7.5-ck8s1
+                type: string
           buffer:
             $ref: '#/$defs/fluentdBuffer'
           livenessThresholdSeconds:

helmfile.d/values/fluentd/forwarder-common.yaml.gotmpl

@@ -4,8 +4,8 @@ image:
   # This is our own fork-image (https://github.com/elastisys/fluentd-elasticsearch).
   # This can later be replaced with the original image (https://github.com/monotek/fluentd-elasticsearch) when our changes is merged.
   # This image is included in falco k8s_containers and user_known_package_manager_in_container macros
-  repository: ghcr.io/elastisys/fluentd-forwarder
-  tag: v4.7.5-ck8s1
+  repository: {{ .Values.fluentd.forwarder.image.repository }}
+  tag: {{ .Values.fluentd.forwarder.image.tag }}
 
 serviceMonitor:
   enabled: true

migration/v0.44/README.md

@@ -82,6 +82,16 @@ As with all scripts in this repository `CK8S_CONFIG_PATH` is expected to be set.
     ./bin/ck8s upgrade wc v0.44 apply
     ```
 
+1. If the environment is still on Ubuntu 20.04, you may want to override the fluentd-forwarder image to the previous one, as the new image has some known issues on Ubuntu 20.04. This can be done in the `common-config.yaml`:
+
+    ```yaml
+    fluentd:
+      forwarder:
+        image:
+          repository: ghcr.io/elastisys/fluentd
+          tag: v4.3.9-ck8s1
+    ```
+
 1. Apply upgrade - _disruptive_
 
     > _Done during maintenance window._
@@ -109,6 +119,16 @@ As with all scripts in this repository `CK8S_CONFIG_PATH` is expected to be set.
     export CK8S_CLUSTER=<wc|sc|both>
     ```
 
+1. If the environment is still on Ubuntu 20.04, you may want to override the fluentd-forwarder image to the previous one, as the new image has some known issues on Ubuntu 20.04. This can be done in the `common-config.yaml`:
+
+    ```yaml
+    fluentd:
+      forwarder:
+        image:
+          repository: ghcr.io/elastisys/fluentd
+          tag: v4.3.9-ck8s1
+    ```
+
 1. Because `harbor.core.replicas` was previously used by the jobservice component any nondefault value for it will be copied to `harbor.jobservice.replicas`:
 
     ```bash