feat(otlp-gateway): make VPA configurable and fix deprecated updateMode

- Change updateMode from deprecated "Auto" to "InPlaceOrRecreate"
- Make VPA configurable via values (enabled by default)
- Lower minAllowed (cpu: 10m, memory: 32Mi) to allow right-sizing
- Increase maxAllowed (cpu: 1, memory: 4Gi) for headroom

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: loafoe

charts/otlp-gateway/Chart.yaml

@@ -1,5 +1,5 @@
 apiVersion: v2
 name: otlp-gateway
-version: 0.50.0
+version: 0.51.0
 description: A Helm chart for Kubernetes to deploy an OTLP gateway with authentication and authorization
 type: application

charts/otlp-gateway/README.md

@@ -1,6 +1,6 @@
 # otlp-gateway
 
-![Version: 0.50.0](https://img.shields.io/badge/Version-0.50.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square)
+![Version: 0.51.0](https://img.shields.io/badge/Version-0.51.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square)
 
 OTLP gateway is a reference implementation which creates a single otlphttp endpoint that proxies Loki, Tempo and Mimir OTLP endpoints
 It supports authentication and authorization using both static and JWT tokens and tokens through the [caddy-token](https://github.com/loafoe/caddy-token) plugin.
@@ -114,6 +114,13 @@ authn:
 | topologySpreadConstraints.maxSkew | int | `1` |  |
 | topologySpreadConstraints.whenUnsatisfiable | string | `"DoNotSchedule"` |  |
 | useCustomFqdn | bool | `true` |  |
+| vpa.enabled | bool | `true` |  |
+| vpa.maxAllowed.cpu | int | `1` |  |
+| vpa.maxAllowed.memory | string | `"4Gi"` |  |
+| vpa.minAllowed.cpu | string | `"10m"` |  |
+| vpa.minAllowed.memory | string | `"32Mi"` |  |
+| vpa.minReplicas | int | `3` |  |
+| vpa.updateMode | string | `"InPlaceOrRecreate"` |  |
 
 ----------------------------------------------
 Autogenerated from chart metadata using [helm-docs v1.14.2](https://github.com/norwoodj/helm-docs/releases/v1.14.2)

charts/otlp-gateway/templates/vpa.yaml

@@ -1,3 +1,4 @@
+{{- if .Values.vpa.enabled }}
 apiVersion: "autoscaling.k8s.io/v1"
 kind: VerticalPodAutoscaler
 metadata:
@@ -8,15 +9,16 @@ spec:
     kind: Deployment
     name: otlp-gateway
   updatePolicy:
-    minReplicas: 3
-    updateMode: Auto
+    minReplicas: {{ .Values.vpa.minReplicas }}
+    updateMode: {{ .Values.vpa.updateMode }}
   resourcePolicy:
     containerPolicies:
       - containerName: '*'
         minAllowed:
-          cpu: 100m
-          memory: 512Mi
+          cpu: {{ .Values.vpa.minAllowed.cpu }}
+          memory: {{ .Values.vpa.minAllowed.memory }}
         maxAllowed:
-          cpu: 500m
-          memory: 3Gi
+          cpu: {{ .Values.vpa.maxAllowed.cpu }}
+          memory: {{ .Values.vpa.maxAllowed.memory }}
         controlledResources: ["cpu","memory"]
+{{- end }}

charts/otlp-gateway/values.yaml

@@ -131,3 +131,14 @@ ruler:
   enabled: true
   pathPrefix: /ruler
   service: mimir-ruler.mimir-system.svc.cluster.local:8080
+
+vpa:
+  enabled: true
+  updateMode: InPlaceOrRecreate
+  minReplicas: 3
+  minAllowed:
+    cpu: 10m
+    memory: 32Mi
+  maxAllowed:
+    cpu: 1
+    memory: 4Gi