Enable vm boot diagnostics

charts/internal/machineclass/templates/machineclass.yaml

@@ -51,6 +51,13 @@ providerSpec:
     networkProfile:
       acceleratedNetworking: {{ $machineClass.network.acceleratedNetworking }}
     {{- end }}
+    {{- if hasKey $machineClass "diagnosticsProfile" }}
+    diagnosticsProfile:
+      enabled: {{ $machineClass.diagnosticsProfile.enabled }}
+      {{- if hasKey $machineClass.diagnosticsProfile "storageURI" }}
+      storageURI: {{ $machineClass.diagnosticsProfile.storageURI }}
+      {{- end }}
+    {{- end }}
     hardwareProfile:
       vmSize: {{ $machineClass.machineType }}
     osProfile:

charts/internal/machineclass/values.yaml

@@ -11,6 +11,9 @@ machineClasses:
     subnet: my-subnet-in-my-vnet
     # vnetResourceGroup: my-vnet-resource-group
     # acceleratedNetworking: true
+  diagnosticsProfile:
+    enabled: false
+    # storageURI: my-custom-azure-storage
   tags:
     Name: shoot-crazy-botany
     kubernetes.io-cluster-shoot-crazy-botany: "1"

docs/usage/usage.md

@@ -353,13 +353,20 @@ nodeTemplate: # (to be specified only if the node capacity would be different fr
     cpu: 2
     gpu: 1
     memory: 50Gi
+diagnosticsProfile:
+  enabled: true
+  # storageURI: <string>
 ```
 
 The `.nodeTemplate` is used to specify resource information of the machine during runtime. This then helps in Scale-from-Zero. 
 Some points to note for this field:
-    - Currently only cpu, gpu and memory are configurable.
-    - a change in the value lead to a rolling update of the machine in the workerpool
-    - all the resources needs to be specified
+- Currently only cpu, gpu and memory are configurable.
+- a change in the value lead to a rolling update of the machine in the worker pool
+- all the resources needs to be specified
+
+The `.diagnosticsProfile` is used to enable [machine boot diagnostics](https://learn.microsoft.com/en-us/azure/virtual-machines/boot-diagnostics) (disabled per default).
+A storage account is used for storing vm's boot console output and screenshots.
+If `.diagnosticsProfile.StorageURI` is not specified azure managed storage will be used (recommended way).
 
 ## Example `Shoot` manifest (non-zoned)
 

example/30-worker.yaml

@@ -113,3 +113,8 @@ spec:
   # - name: kubelet-dir
   #   type: standard
   #   size: 36Gi
+  # providerConfig:
+  #   apiVersion: azure.provider.extensions.gardener.cloud/v1alpha1
+  #   kind: WorkerConfig
+  #   diagnosticsProfile:
+  #     enabled: false

hack/api-reference/api.md

@@ -403,6 +403,19 @@ github.com/gardener/gardener/pkg/apis/extensions/v1alpha1.NodeTemplate
 <p>NodeTemplate contains resource information of the machine which is used by Cluster Autoscaler to generate nodeTemplate during scaling a nodeGroup from zero.</p>
 </td>
 </tr>
+<tr>
+<td>
+<code>diagnosticsProfile</code></br>
+<em>
+<a href="#azure.provider.extensions.gardener.cloud/v1alpha1.DiagnosticsProfile">
+DiagnosticsProfile
+</a>
+</em>
+</td>
+<td>
+<p>DiagnosticsProfile specifies boot diagnostic options</p>
+</td>
+</tr>
 </tbody>
 </table>
 <h3 id="azure.provider.extensions.gardener.cloud/v1alpha1.WorkerStatus">WorkerStatus
@@ -652,6 +665,48 @@ map[string]bool
 </tr>
 </tbody>
 </table>
+<h3 id="azure.provider.extensions.gardener.cloud/v1alpha1.DiagnosticsProfile">DiagnosticsProfile
+</h3>
+<p>
+(<em>Appears on:</em>
+<a href="#azure.provider.extensions.gardener.cloud/v1alpha1.WorkerConfig">WorkerConfig</a>)
+</p>
+<p>
+<p>DiagnosticsProfile specifies boot diagnostic options</p>
+</p>
+<table>
+<thead>
+<tr>
+<th>Field</th>
+<th>Description</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>
+<code>enabled</code></br>
+<em>
+bool
+</em>
+</td>
+<td>
+<p>Enabled configures boot diagnostics to be stored or not</p>
+</td>
+</tr>
+<tr>
+<td>
+<code>storageURI</code></br>
+<em>
+string
+</em>
+</td>
+<td>
+<p>StorageURI is the URI of the storage account to use for storing console output and screenshot.
+If not specified azure managed storage will be used.</p>
+</td>
+</tr>
+</tbody>
+</table>
 <h3 id="azure.provider.extensions.gardener.cloud/v1alpha1.DomainCount">DomainCount
 </h3>
 <p>

pkg/apis/azure/types_worker.go

@@ -16,6 +16,8 @@ type WorkerConfig struct {
 	metav1.TypeMeta
 	// NodeTemplate contains resource information of the machine which is used by Cluster Autoscaler to generate nodeTemplate during scaling a nodeGroup from zero.
 	NodeTemplate *extensionsv1alpha1.NodeTemplate
+	// DiagnosticsProfile specifies boot diagnostic options
+	DiagnosticsProfile *DiagnosticsProfile
 }
 
 // +genclient
@@ -65,3 +67,12 @@ type VmoDependency struct {
 	// Name is the name of the VMO resource on Azure.
 	Name string
 }
+
+// DiagnosticsProfile specifies boot diagnostic options
+type DiagnosticsProfile struct {
+	// Enabled configures boot diagnostics to be stored or not
+	Enabled bool
+	// StorageURI is the URI of the storage account to use for storing console output and screenshot.
+	// If not specified azure managed storage will be used.
+	StorageURI *string
+}

pkg/apis/azure/v1alpha1/types_worker.go

@@ -19,6 +19,8 @@ type WorkerConfig struct {
 	// NodeTemplate contains resource information of the machine which is used by Cluster Autoscaler to generate nodeTemplate during scaling a nodeGroup from zero.
 	// +optional
 	NodeTemplate *extensionsv1alpha1.NodeTemplate `json:"nodeTemplate,omitempty"`
+	// DiagnosticsProfile specifies boot diagnostic options
+	DiagnosticsProfile *DiagnosticsProfile `json:"diagnosticsProfile,omitempty"`
 }
 
 // +genclient
@@ -76,3 +78,12 @@ type VmoDependency struct {
 	// Name is the name of the VMO resource on Azure.
 	Name string `json:"name"`
 }
+
+// DiagnosticsProfile specifies boot diagnostic options
+type DiagnosticsProfile struct {
+	// Enabled configures boot diagnostics to be stored or not
+	Enabled bool `json:"enabled,omitempty"`
+	// StorageURI is the URI of the storage account to use for storing console output and screenshot.
+	// If not specified azure managed storage will be used.
+	StorageURI *string `json:"storageURI,omitempty"`
+}

pkg/controller/worker/machines.go

@@ -204,6 +204,16 @@ func (w *workerDelegate) generateMachineConfig(ctx context.Context) error {
 				machineClassSpec["zone"] = zone.name
 			}
 
+			if workerConfig.DiagnosticsProfile != nil {
+				diagnosticProfile := map[string]interface{}{
+					"enabled": workerConfig.DiagnosticsProfile.Enabled,
+				}
+				if workerConfig.DiagnosticsProfile.StorageURI != nil {
+					diagnosticProfile["storageURI"] = workerConfig.DiagnosticsProfile.StorageURI
+				}
+				machineClassSpec["diagnosticsProfile"] = diagnosticProfile
+			}
+
 			if pool.NodeTemplate != nil {
 				//	Currently Zone field is mandatory, and passing it an
 				//	empty string turns it to `null` string during marshalling which fails CRD validation

pkg/controller/worker/machines_test.go

@@ -6,6 +6,7 @@ package worker_test
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
 	"path/filepath"
 	"strings"
@@ -141,6 +142,10 @@ var _ = Describe("Machines", func() {
 				nodeTemplateZone3 machinev1alpha1.NodeTemplate
 				nodeTemplateZone4 machinev1alpha1.NodeTemplate
 
+				diagnosticProfile apiv1alpha1.DiagnosticsProfile
+				providerConfig    *runtime.RawExtension
+				workerConfig      apiv1alpha1.WorkerConfig
+
 				shootVersionMajorMinor string
 				shootVersion           string
 
@@ -226,6 +231,25 @@ var _ = Describe("Machines", func() {
 					Zone:         "no-zone",
 				}
 
+				diagnosticProfile = apiv1alpha1.DiagnosticsProfile{
+					Enabled:    true,
+					StorageURI: ptr.To("azure-storage-uri"),
+				}
+
+				workerConfig = apiv1alpha1.WorkerConfig{
+					TypeMeta: metav1.TypeMeta{
+						APIVersion: apiv1alpha1.SchemeGroupVersion.String(),
+						Kind:       "WorkerConfig",
+					},
+					DiagnosticsProfile: &diagnosticProfile,
+				}
+
+				marshalledWorkerConfig, err := json.Marshal(workerConfig)
+				Expect(err).To(BeNil())
+				providerConfig = &runtime.RawExtension{
+					Raw: marshalledWorkerConfig,
+				}
+
 				namePool2 = "pool-zones"
 				minPool2 = 30
 				maxPool2 = 45
@@ -310,7 +334,8 @@ var _ = Describe("Machines", func() {
 							Type: &dataVolume2Type,
 						},
 					},
-					Labels: labels,
+					Labels:         labels,
+					ProviderConfig: providerConfig,
 				}
 
 				pool2 = extensionsv1alpha1.WorkerPool{
@@ -502,6 +527,11 @@ var _ = Describe("Machines", func() {
 					machineClassPool3["nodeTemplate"] = nodeTemplateZone3
 					machineClassPool4["nodeTemplate"] = nodeTemplateZone4
 
+					machineClassPool1["diagnosticsProfile"] = map[string]interface{}{
+						"enabled":    diagnosticProfile.Enabled,
+						"storageURI": diagnosticProfile.StorageURI,
+					}
+
 					machineClassPool1["dataDisks"] = []map[string]interface{}{
 						{
 							"name":               dataVolume2Name,