[gpu-metrics-exporter] add support for clusterIDSecretRef

charts/gpu-metrics-exporter/README.md

@@ -0,0 +1,43 @@
+# gpu-metrics-exporter
+
+![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 0.1.16](https://img.shields.io/badge/AppVersion-0.1.16-informational?style=flat-square)
+
+A Helm chart for Kubernetes
+
+## Values
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| castai.apiKey | string | `""` |  |
+| castai.clusterId | string | `""` |  |
+| castai.clusterIdSecretKeyRef.key | string | `"CLUSTER_ID"` |  |
+| castai.clusterIdSecretKeyRef.name | string | `""` |  |
+| dcgmExporter.enabled | bool | `true` |  |
+| dcgmExporter.image.pullPolicy | string | `"IfNotPresent"` |  |
+| dcgmExporter.image.repository | string | `"nvcr.io/nvidia/k8s/dcgm-exporter"` |  |
+| dcgmExporter.image.tag | string | `"3.3.7-3.5.0-ubuntu22.04"` |  |
+| dcgmExporter.useExternalHostEngine | bool | `false` |  |
+| gke.affinity.nodeAffinity.requiredDuringSchedulingIgnoredDuringExecution.nodeSelectorTerms[0].matchExpressions[0].key | string | `"cloud.google.com/gke-accelerator"` |  |
+| gke.affinity.nodeAffinity.requiredDuringSchedulingIgnoredDuringExecution.nodeSelectorTerms[0].matchExpressions[0].operator | string | `"Exists"` |  |
+| gpuMetricsExporter.config | object | `{}` |  |
+| gpuMetricsExporter.image.pullPolicy | string | `"IfNotPresent"` |  |
+| gpuMetricsExporter.image.repository | string | `"ghcr.io/castai/gpu-metrics-exporter/gpu-metrics-exporter"` |  |
+| gpuMetricsExporter.image.tag | string | `""` |  |
+| gpuMetricsExporter.port | int | `6061` |  |
+| gpuMetricsExporter.resources.limits.cpu | string | `"100m"` |  |
+| gpuMetricsExporter.resources.limits.memory | string | `"128Mi"` |  |
+| gpuMetricsExporter.resources.requests.cpu | string | `"100m"` |  |
+| gpuMetricsExporter.resources.requests.memory | string | `"128Mi"` |  |
+| gpuMetricsExporter.securityContext.readOnlyRootFilesystem | bool | `true` |  |
+| gpuMetricsExporter.securityContext.runAsNonRoot | bool | `true` |  |
+| imagePullSecrets | list | `[]` |  |
+| migrate.image.pullPolicy | string | `"IfNotPresent"` |  |
+| migrate.image.repository | string | `"alpine/k8s"` |  |
+| migrate.image.tag | string | `"1.31.0"` |  |
+| provider | string | `""` |  |
+| serviceAccount.annotations | object | `{}` |  |
+| serviceAccount.automount | bool | `true` |  |
+| serviceAccount.create | bool | `true` |  |
+
+----------------------------------------------
+Autogenerated from chart metadata using [helm-docs v1.11.0](https://github.com/norwoodj/helm-docs/releases/v1.11.0)

charts/gpu-metrics-exporter/ci/eks-values.yaml

@@ -0,0 +1 @@
+provider: eks

charts/gpu-metrics-exporter/templates/daemonset.yaml

@@ -88,6 +88,16 @@ spec:
                 secretKeyRef:
                   name: {{ include "gpu-metrics-exporter.fullname" . }}
                   key: API_KEY
+          {{- if .Values.castai.clusterIdSecretKeyRef.name }}
+            {{- if ne .Values.castai.clusterId "" }}
+            {{- fail "clusterId and clusterIdSecretKeyRef are mutually exclusive" }}
+            {{- end }}
+            - name: "CLUSTER_ID"
+              valueFrom:
+                secretKeyRef:
+                  name: {{ .Values.castai.clusterIdSecretKeyRef.name }}
+                  key: {{ .Values.castai.clusterIdSecretKeyRef.key }}
+          {{- end }}
           {{- if .Values.dcgmExporter.enabled }}
             - name: "DCGM_HOST"
               value: "localhost"

charts/gpu-metrics-exporter/values.yaml

@@ -1,4 +1,5 @@
-provider: "" # gke | eks | aks
+# Can be one of: `gke`, `eks`, `aks`.
+provider: ""
 imagePullSecrets: []
 
 serviceAccount:
@@ -8,7 +9,11 @@ serviceAccount:
 
 castai:
   apiKey: ""
+  # clusterId and clusterIdSecretRef are mutually exclusive
   clusterId: ""
+  clusterIdSecretKeyRef:
+    name: ""
+    key: "CLUSTER_ID"
 
 gpuMetricsExporter:
   image:
@@ -53,4 +58,4 @@ migrate:
   image:
     repository: alpine/k8s
     pullPolicy: IfNotPresent
-    tag: 1.31.0
+    tag: 1.31.0