[nvidia] chore: chart improvements (#357)

* [nvidia] chore: chart improvements * fixed requirements file * fixed missing field in chart file * values file fixed
mesosphere · Jan 16, 2020 · 5dbdd66 · 5dbdd66
1 parent 1efeb5c
commit 5dbdd66
Show file tree

Hide file tree

Showing 20 changed files with 132 additions and 222 deletions.
diff --git a/staging/nvidia/Chart.yaml b/staging/nvidia/Chart.yaml
@@ -1,8 +1,13 @@
 apiVersion: v1
 name: nvidia
 description: Nvidia GPU driver and device plugin for running Nvidia GPU
-home: https://docs.nvidia.com/datacenter/kubernetes/kubernetes-upstream/
-version: 0.2.0
-maintainers:
-- name: gilbert88
+keywords:
+  - gpu
+  - nvidia
+version: 0.3.0
 appVersion: "0.2.0"
+home: http://github.com/mesosphere/charts/staging
+sources:
+  - http://github.com/mesosphere/charts/staging
+maintainers:
+  - name: gilbert88
diff --git a/staging/nvidia/charts/grafana/Chart.yaml b/staging/nvidia/charts/grafana/Chart.yaml
@@ -1,5 +1,5 @@
 apiVersion: v1
 description: Nvidia GPU Grafana Dashboard
 name: grafana
-version: 0.1.0
+version: 0.1.1
 appVersion: "0.1.0"
diff --git a/staging/nvidia/charts/grafana/templates/configmap-dashboards.yaml b/staging/nvidia/charts/grafana/templates/configmap-dashboards.yaml
@@ -1,12 +1,8 @@
-# Generated from 'nodes' from https://raw.githubusercontent.com/coreos/kube-prometheus/master/manifests/grafana-dashboardDefinitions.yaml
-# Do not change in-place! In order to change this file first read following link:
-# https://github.com/helm/charts/tree/master/stable/prometheus-operator/hack
-{{- if .Values.enabled }}
 apiVersion: v1
 kind: ConfigMap
 metadata:
   name: {{ template "grafana.fullname" . }}
-  namespace: {{ $.Values.namespace }}
+  namespace: {{ .Release.Namespace }}
   labels:
     {{- if $.Values.dashboards.label }}
     {{ $.Values.dashboards.label }}: "1"
@@ -93,7 +89,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU SM Clocks",
+                        "title": "SM Clock Frequency",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -169,7 +165,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU Memory Clocks",
+                        "title": "Memory Clock Frequency",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -256,7 +252,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU Utilization",
+                        "title": "Utilization",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -334,7 +330,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU Mem Cpy Utilization",
+                        "title": "Memory Copy Utilization",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -422,7 +418,7 @@ data:
                                 "refId": "A"
                             }
                         ],
-                        "title": "GPU Power Usage",
+                        "title": "Power Usage",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -509,7 +505,7 @@ data:
                             }
                         ],
                         "thresholds": "1800, 2200",
-                        "title": "GPU Power Total",
+                        "title": "Power Total",
                         "transparent": false,
                         "type": "singlestat",
                         "valueFontSize": "80%",
@@ -581,7 +577,7 @@ data:
                                 "refId": "A"
                             }
                         ],
-                        "title": "GPU Temperature",
+                        "title": "Temperature",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -669,7 +665,7 @@ data:
                             }
                         ],
                         "thresholds": "83, 87",
-                        "title": "GPU Avg. Temp",
+                        "title": "Average Temperature",
                         "transparent": false,
                         "type": "singlestat",
                         "valueFontSize": "80%",
@@ -741,7 +737,7 @@ data:
                                 "refId": "A"
                             }
                         ],
-                        "title": "GPU Memory Temperature",
+                        "title": "Memory Temperature",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -829,7 +825,7 @@ data:
                             }
                         ],
                         "thresholds": "83, 87",
-                        "title": "GPU Avg. Memory Temp",
+                        "title": "Average Memory Temperature",
                         "transparent": false,
                         "type": "singlestat",
                         "valueFontSize": "80%",
@@ -902,7 +898,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU Framebuffer Mem Used",
+                        "title": "Framebuffer Memory Used",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -980,7 +976,7 @@ data:
                                 "step": 50
                             }
                         ],
-                        "title": "GPU Framebuffer Mem Free",
+                        "title": "Framebuffer Memory Free",
                         "tooltip": {
                             "msResolution": false,
                             "shared": true,
@@ -1016,7 +1012,10 @@ data:
         "schemaVersion": 14,
         "sharedCrosshair": false,
         "style": "dark",
-        "tags": [],
+        "tags": [
+          "Nvidia",
+          "GPU"
+        ],
         "templating": {
             "list": [
                 {
@@ -1119,7 +1118,6 @@ data:
             ]
         },
         "timezone": "browser",
-        "title": "GPU",
+        "title": "GPUs / Nvidia",
         "version": 2
-    }
-{{- end }}
+    }
diff --git a/staging/nvidia/charts/grafana/values.yaml b/staging/nvidia/charts/grafana/values.yaml
@@ -1,7 +1,2 @@
-#
-# grafana dashboards
-#
-enabled: true
-namespace: kubeaddons
 dashboards:
   label: grafana_dashboard
diff --git a/staging/nvidia/charts/nvidia-dcgm-exporter/Chart.yaml b/staging/nvidia/charts/nvidia-dcgm-exporter/Chart.yaml
@@ -2,4 +2,6 @@ apiVersion: v1
 appVersion: "0.1.0"
 description: Nvidia GPU Metrics Exporter
 name: nvidia-dcgm-exporter
-version: 0.1.0
+version: 0.1.1
+maintainers:
+  - name: gilbert88
diff --git a/staging/nvidia/charts/nvidia-dcgm-exporter/templates/daemonset.yaml b/staging/nvidia/charts/nvidia-dcgm-exporter/templates/daemonset.yaml
@@ -1,9 +1,8 @@
-{{- if .Values.enabled }}
 apiVersion: apps/v1
 kind: DaemonSet
 metadata:
   name: {{ template "nvidia-dcgm-exporter.fullname" . }}
-  namespace: {{ $.Values.namespace }}
+  namespace: {{ .Release.Namespace }}
 spec:
   selector:
     matchLabels:
@@ -19,49 +18,54 @@ spec:
 {{- end }}
 {{- if .Values.tolerations }}
       tolerations:
-{{ toYaml .Values.tolerations | indent 8 }}
+{{- toYaml .Values.tolerations | indent 8 }}
 {{- end }}
+{{- if .Values.nodeSelector }}
+      nodeSelector:
+{{- toYaml .Values.nodeSelector | indent 8 }}
+{{- end }}
+{{- if .Values.priorityClassName }}
       priorityClassName: {{ .Values.priorityClassName }}
+{{- end }}
       containers:
-      - image: "{{ .Values.podNvidiaGpuMetricsExporter.image.repository }}:{{ .Values.podNvidiaGpuMetricsExporter.image.tag }}"
-        name: pod-nvidia-gpu-metrics-exporter
+      - name: pod-nvidia-gpu-metrics-exporter
+        image: "{{ .Values.podNvidiaGpuMetricsExporter.image.repository }}:{{ .Values.podNvidiaGpuMetricsExporter.image.tag }}"
+        imagePullPolicy: {{ .Values.podNvidiaGpuMetricsExporter.image.pullPolicy }}
+{{- if .Values.podNvidiaGpuMetricsExporter.resources }}
+        resources:
+{{ toYaml .Values.podNvidiaGpuMetricsExporter.resources | indent 12 }}
+{{- end }}
         ports:
         - name: gpu-metrics
           containerPort: 9400
           hostPort: 9400
         securityContext:
           runAsNonRoot: false
           runAsUser: 0
-        imagePullPolicy: {{ .Values.podNvidiaGpuMetricsExporter.image.pullPolicy }}
-        resources:
-{{ toYaml .Values.podNvidiaGpuMetricsExporter.resources | indent 12 }}
         volumeMounts:
         - name: pod-gpu-resources
           readOnly: true
           mountPath: /var/lib/kubelet/pod-resources
         - name: device-metrics
           readOnly: true
           mountPath: /run/prometheus
-      - image: "{{ .Values.nvidiaDcgmExporter.image.repository }}:{{ .Values.nvidiaDcgmExporter.image.tag }}"
-        name: nvidia-dcgm-exporter
+      - name: nvidia-dcgm-exporter
+        image: "{{ .Values.nvidiaDcgmExporter.image.repository }}:{{ .Values.nvidiaDcgmExporter.image.tag }}"
+        imagePullPolicy: {{ .Values.nvidiaDcgmExporter.image.pullPolicy }}
+{{- if .Values.nvidiaDcgmExporter.resources }}
+        resources:
+{{- toYaml .Values.nvidiaDcgmExporter.resources | indent 12 }}
+{{- end }}
         securityContext:
           runAsNonRoot: false
           runAsUser: 0
-        imagePullPolicy: {{ .Values.nvidiaDcgmExporter.image.pullPolicy }}
-        resources:
-{{ toYaml .Values.nvidiaDcgmExporter.resources | indent 12 }}
         volumeMounts:
         - name: device-metrics
           mountPath: /run/prometheus
-{{- if .Values.nodeSelector }}
-      nodeSelector:
-{{ toYaml .Values.nodeSelector | indent 8 }}
-{{- end }}
       volumes:
       - name: pod-gpu-resources
         hostPath:
           path: /var/lib/kubelet/pod-resources
       - name: device-metrics
         emptyDir:
-          medium: Memory
-{{- end }}
+          medium: Memory
diff --git a/staging/nvidia/charts/nvidia-dcgm-exporter/templates/tests/test-connection.yaml b/staging/nvidia/charts/nvidia-dcgm-exporter/templates/tests/test-connection.yaml
@@ -11,5 +11,5 @@ spec:
     - name: wget
       image: busybox
       command: ['wget']
-      args:  ['{{ include "nvidia-dcgm-exporter.fullname" . }}:{{ .Values.service.port }}']
+      args:  ['{{ include "nvidia-dcgm-exporter.fullname" . }}:80']
   restartPolicy: Never
diff --git a/staging/nvidia/charts/nvidia-dcgm-exporter/values.yaml b/staging/nvidia/charts/nvidia-dcgm-exporter/values.yaml
@@ -1,13 +1,3 @@
-# Default values for nvidia-dcgm-exporter.
-# This is a YAML-formatted file.
-# Declare variables to be passed into your templates.
-
-enabled: true
-
-namespace: kubeaddons
-
-replicaCount: 1
-
 podNvidiaGpuMetricsExporter:
   image:
     repository: nvidia/pod-gpu-metrics-exporter
@@ -22,24 +12,6 @@ nvidiaDcgmExporter:
     pullPolicy: IfNotPresent
   resources: {}
 
-service:
-  type: ClusterIP
-  port: 80
-
-ingress:
-  enabled: false
-  annotations: {}
-    # kubernetes.io/ingress.class: nginx
-    # kubernetes.io/tls-acme: "true"
-  hosts:
-    - host: chart-example.local
-      paths: []
-
-  tls: []
-  #  - secretName: chart-example-tls
-  #    hosts:
-  #      - chart-example.local
-
 nodeSelector: {}
 
 priorityClassName: ""

diff --git a/staging/nvidia/charts/nvidia-device-plugin/Chart.yaml b/staging/nvidia/charts/nvidia-device-plugin/Chart.yaml
@@ -3,6 +3,6 @@ name: nvidia-device-plugin
 home: https://github.com/mesosphere/charts
 appVersion: "1.0"
 description: Nvidia Device Plugin for konvoy
-version: 0.1.1
+version: 0.1.2
 maintainers:
   - name: gilbert88
diff --git a/staging/nvidia/charts/nvidia-device-plugin/templates/NOTES.txt b/staging/nvidia/charts/nvidia-device-plugin/templates/NOTES.txt