AKS Version Update + Others (#357)

* Update kured, AKS version, resource params, and add eraser * disable the new preview * enable node restrictions * define more defaults * can't even enclude it as disabled without registering it. * tenantId -> tenantID to remove warning Add firewall allowance * remove legacy warning for unsupported kubectl version * add suffix * Update to five days
mspnp · Nov 16, 2022 · 8ca6ef5 · 8ca6ef5
1 parent b8a67fd
commit 8ca6ef5
Show file tree

Hide file tree

Showing 8 changed files with 105 additions and 32 deletions.
diff --git a/01-prerequisites.md b/01-prerequisites.md
@@ -34,12 +34,15 @@ This is the starting point for the instructions on deploying the [AKS Baseline r
 
    1. [Register the Workload Identity preview feature = `EnableWorkloadIdentityPreview`](https://learn.microsoft.com/azure/aks/workload-identity-deploy-cluster#register-the-enableworkloadidentitypreview-feature-flag)
 
+   1. Register the ImageCleaner (Earser) preview feature = `EnableImageCleanerPreview`](https://learn.microsoft.com/azure/aks/image-cleaner#prerequisites)
+
    ```bash
    az feature register --namespace "Microsoft.ContainerService" -n "AKS-AzureDefender"
    az feature register --namespace "Microsoft.ContainerService" -n "EnableWorkloadIdentityPreview"
+   az feature register --namespace "Microsoft.ContainerService" -n "EnableImageCleanerPreview"
 
    # Keep running until all say "Registered." (This may take up to 20 minutes.)
-   az feature list -o table --query "[?name=='Microsoft.ContainerService/AKS-AzureDefender' || name=='Microsoft.ContainerService/EnableWorkloadIdentityPreview'].{Name:name,State:properties.state}"
+   az feature list -o table --query "[?name=='Microsoft.ContainerService/AKS-AzureDefender' || name=='Microsoft.ContainerService/EnableWorkloadIdentityPreview' || name=='Microsoft.ContainerService/EnableImageCleanerPreview'].{Name:name,State:properties.state}"
 
    # When all say "Registered" then re-register the AKS resource provider
    az provider register --namespace Microsoft.ContainerService

diff --git a/05-bootstrap-prep.md b/05-bootstrap-prep.md
@@ -55,7 +55,7 @@ We'll be bootstrapping this cluster with the Flux GitOps agent as installed as a
    echo ACR_NAME_AKS_BASELINE: $ACR_NAME_AKS_BASELINE
 
    # Import core image(s) hosted in public container registries to be used during bootstrapping
-   az acr import --source docker.io/weaveworks/kured:1.10.1 -n $ACR_NAME_AKS_BASELINE
+   az acr import --source ghcr.io/kubereboot/kured:1.11.0 -n $ACR_NAME_AKS_BASELINE
    ```
 
    > In this walkthrough, there is only one image that is included in the bootstrapping process. It's included as an reference for this process. Your choice to use Kubernetes Reboot Daemon (Kured) or any other images, including helm charts, as part of your bootstrapping is yours to make.

diff --git a/09-secret-management-and-ingress-controller.md b/09-secret-management-and-ingress-controller.md
@@ -48,7 +48,7 @@ Previously you have configured [workload prerequisites](./08-workload-prerequisi
              objectName: traefik-ingress-internal-aks-ingress-tls
              objectAlias: tls.key
              objectType: secret
-       tenantId: $TENANTID_AZURERBAC_AKS_BASELINE
+       tenantID: $TENANTID_AZURERBAC_AKS_BASELINE
    EOF
    ```
 

diff --git a/10-workload.md b/10-workload.md
@@ -54,8 +54,6 @@ The cluster now has an [Traefik configured with a TLS certificate](./08-secret-m
    exit
    ```
 
-   > :beetle: If you are running a version of kubectl less than 1.23 you cannot perform this step.  The Azure Policy for Kubernetes assignments that are deployed with this implementation requires both setting limits and also a read-only root filesystem. This isn't possible with kubectl 1.22 and lower. It's okay to skip this step if in that situation.
-
    > From this container shell, you could also try to directly access the workload via `curl -I http://<aspnetapp-service-cluster-ip>`. Instead of getting back a `200 OK`, you'll receive a network timeout because of the [`allow-only-ingress-to-workload` network policy](./cluster-manifests/a0008/ingress-network-policy.yaml) that is in place.
 
 ### Next step

diff --git a/README.md b/README.md
@@ -24,7 +24,7 @@ Finally, this implementation uses the [ASP.NET Core Docker sample web app](https
 
 #### Azure platform
 
-- AKS v1.24
+- AKS v1.25
   - System and User [node pool separation](https://learn.microsoft.com/azure/aks/use-system-pools)
   - [AKS-managed Azure AD](https://learn.microsoft.com/azure/aks/managed-aad)
   - Azure AD-backed Kubernetes RBAC (_local user accounts disabled_)
@@ -38,11 +38,13 @@ Finally, this implementation uses the [ASP.NET Core Docker sample web app](https
 
 #### In-cluster OSS components
 
-- [Flux GitOps Operator](https://fluxcd.io) _[AKS-managed extension]_
-- [Traefik Ingress Controller](https://doc.traefik.io/traefik/v2.5/routing/providers/kubernetes-ingress/)
 - [Azure Workload Identity](https://learn.microsoft.com/azure/aks/workload-identity-overview) _[AKS-managed add-on]_
+- [Flux GitOps Operator](https://fluxcd.io) _[AKS-managed extension]_
+- [ImageCleaner (Eraser)](https://learn.microsoft.com/azure/aks/image-cleaner) _[AKS-managed add-on]_
+- [Kubernetes Reboot Daemon](https://learn.microsoft.com/azure/aks/node-updates-kured)
 - [Secrets Store CSI Driver for Kubernetes](https://learn.microsoft.com/azure/aks/csi-secrets-store-driver) _[AKS-managed add-on]_
-- [Kured](https://learn.microsoft.com/azure/aks/node-updates-kured)
+- [Traefik Ingress Controller](https://doc.traefik.io/traefik/v2.5/routing/providers/kubernetes-ingress/)
+
 
 ![Network diagram depicting a hub-spoke network with two peered VNets and main Azure resources used in the architecture.](https://learn.microsoft.com/azure/architecture/reference-architectures/containers/aks/images/secure-baseline-architecture.svg)
 

diff --git a/cluster-manifests/cluster-baseline-settings/kured.yaml b/cluster-manifests/cluster-baseline-settings/kured.yaml
@@ -1,4 +1,4 @@
-# Source: https://github.com/weaveworks/kured/releases/download/1.10.1/kured-1.10.1-dockerhub.yaml
+# Source: https://github.com/kubereboot/charts/tree/kured-4.1.0/charts/kured (1.11.0)
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRole
 metadata:
@@ -16,6 +16,9 @@ rules:
 - apiGroups: [""]
   resources: ["pods"]
   verbs:     ["list","delete","get"]
+- apiGroups: ["extensions"]
+  resources: ["daemonsets"]
+  verbs:     ["get"]
 - apiGroups: ["apps"]
   resources: ["daemonsets"]
   verbs:     ["get"]
@@ -42,11 +45,15 @@ metadata:
   namespace: cluster-baseline-settings
   name: kured
 rules:
-# Allow kured to lock/unlock itself
-- apiGroups:     ["apps"]
-  resources:     ["daemonsets"]
-  resourceNames: ["kured"]
-  verbs:         ["update"]
+  # Allow kured to lock/unlock itself
+  - apiGroups:     ["extensions"]
+    resources:     ["daemonsets"]
+    resourceNames: ["n-kured"]
+    verbs:         ["update", "patch"]
+  - apiGroups:     ["apps"]
+    resources:     ["daemonsets"]
+    resourceNames: ["kured"]
+    verbs:         ["update", "patch"]
 ---
 apiVersion: rbac.authorization.k8s.io/v1
 kind: RoleBinding
@@ -76,26 +83,31 @@ metadata:
 spec:
   selector:
     matchLabels:
-      name: kured
+      app.kubernetes.io/name: kured
   updateStrategy:
     type: RollingUpdate
+    rollingUpdate:
+      maxUnavailable: 1
   template:
     metadata:
       labels:
-        name: kured
+        app.kubernetes.io/name: kured
       annotations:
         prometheus.io/scrape: "true"
         prometheus.io/path: "/metrics"
         prometheus.io/port: "8080"
     spec:
       serviceAccountName: kured
       tolerations:
+        - key: node-role.kubernetes.io/control-plane
+          effect: NoSchedule
         - key: node-role.kubernetes.io/master
           effect: NoSchedule
-        - effect: NoSchedule
-          key: CriticalAddonsOnly
+        - key: CriticalAddonsOnly
+          effect: NoSchedule
           operator: Equal
           value: "true"
+      hostNetwork: true
       hostPID: true # Facilitate entering the host mount namespace via init
       restartPolicy: Always
       nodeSelector:
@@ -106,10 +118,10 @@ spec:
           # PRODUCTION READINESS CHANGE REQUIRED
           # This image should be sourced from a non-public container registry, such as the
           # one deployed along side of this reference implementation.
-          # az acr import --source docker.io/weaveworks/kured:1.10.1 -n <your-acr-instance-name>
+          # az acr import --source ghcr.io/kubereboot/kured:1.11.0 -n <your-acr-instance-name>
           # and then set this to
-          # image: <your-acr-instance-name>.azurecr.io/weaveworks/kured:1.10.1
-          image: docker.io/weaveworks/kured:1.10.1
+          # image: <your-acr-instance-name>.azurecr.io/kubereboot/kured:1.10.1
+          image: ghcr.io/kubereboot/kured:1.11.0
           imagePullPolicy: IfNotPresent
           securityContext:
             privileged: true # Give permission to nsenter /proc/1/ns/mnt
@@ -120,14 +132,22 @@ spec:
             requests:
               cpu: 200m
               memory: 16Mi
+          ports:
+            - containerPort: 8080
+              name: metrics
           env:
+            # Pass in the name of the node on which this pod is scheduled
+            # for use with drain/uncordon operations and lock acquisition
             - name: KURED_NODE_ID
               valueFrom:
                 fieldRef:
                   fieldPath: spec.nodeName
           command:
             - /usr/bin/kured
+          args:
             - --ds-namespace=cluster-baseline-settings
+#            - --ds-name=kured
+#            - --reboot-command=/bin/systemctl reboot
 #            - --force-reboot=false
 #            - --drain-grace-period=-1
 #            - --skip-wait-for-delete-timeout=0

diff --git a/cluster-stamp.bicep b/cluster-stamp.bicep
@@ -43,7 +43,7 @@ param clusterAuthorizedIPRanges array = []
   'southeastasia'
 ])
 param location string = 'eastus2'
-param kubernetesVersion string = '1.24.6'
+param kubernetesVersion string = '1.25.2'
 
 @description('Domain name to use for App Gateway and AKS ingress.')
 param domainName string = 'contoso.com'
@@ -1121,8 +1121,8 @@ resource paEnforceImageSource 'Microsoft.Authorization/policyAssignments@2021-06
     policyDefinitionId: pdEnforceImageSourceId
     parameters: {
       allowedContainerImagesRegex: {
-        // If all images are pull into your ARC instance as described in these instructions you can remove the docker.io entries.
-        value: '${acr.name}\\.azurecr\\.io/.+$|mcr\\.microsoft\\.com/.+$|docker\\.io/weaveworks/kured.+$|docker\\.io/library/.+$'
+        // If all images are pull into your ARC instance as described in these instructions you can remove the docker.io & ghcr.io entries.
+        value: '${acr.name}\\.azurecr\\.io/.+$|mcr\\.microsoft\\.com/.+$|ghcr\\.io/kubereboot/kured.+$|docker\\.io/library/.+$'
       }
       excludedNamespaces: {
         value: [
@@ -1651,7 +1651,7 @@ resource pdzAksIngress 'Microsoft.Network/privateDnsZones@2020-06-01' = {
   }
 }
 
-resource mc 'Microsoft.ContainerService/managedClusters@2022-03-02-preview' = {
+resource mc 'Microsoft.ContainerService/managedClusters@2022-09-02-preview' = {
   name: clusterName
   location: location
   tags: {
@@ -1669,10 +1669,14 @@ resource mc 'Microsoft.ContainerService/managedClusters@2022-03-02-preview' = {
         osDiskSizeGB: 80
         osDiskType: 'Ephemeral'
         osType: 'Linux'
+        osSKU: 'Ubuntu'
         minCount: 3
         maxCount: 4
         vnetSubnetID: targetVirtualNetwork::snetClusterNodes.id
         enableAutoScaling: true
+        enableCustomCATrust: false
+        enableFIPS: false
+        enableEncryptionAtHost: false
         type: 'VirtualMachineScaleSets'
         mode: 'System'
         scaleSetPriority: 'Regular'
@@ -1699,10 +1703,14 @@ resource mc 'Microsoft.ContainerService/managedClusters@2022-03-02-preview' = {
         osDiskSizeGB: 120
         osDiskType: 'Ephemeral'
         osType: 'Linux'
+        osSKU: 'Ubuntu'
         minCount: 2
         maxCount: 5
         vnetSubnetID: targetVirtualNetwork::snetClusterNodes.id
         enableAutoScaling: true
+        enableCustomCATrust: false
+        enableFIPS: false
+        enableEncryptionAtHost: false
         type: 'VirtualMachineScaleSets'
         mode: 'User'
         scaleSetPriority: 'Regular'
@@ -1794,14 +1802,54 @@ resource mc 'Microsoft.ContainerService/managedClusters@2022-03-02-preview' = {
     podIdentityProfile: {
       enabled: false // Using federated workload identity for Azure AD Pod identities, not the deprecated AAD Pod Identity
     }
+    autoUpgradeProfile: {
+      upgradeChannel: 'stable'
+    }
+    azureMonitorProfile: {
+      metrics: {
+        enabled: false // This is for the AKS-PrometheusAddonPreview, which is not enabled in this cluster as Container Insights is already collecting.
+      }
+    }
+    storageProfile: { // By default, do not support native state storage, enable as needed to support workloads that require state
+      blobCSIDriver: {
+        enabled: false // Azure Blobs
+      }
+      diskCSIDriver: {
+        enabled: false  // Azure Disk
+      }
+      fileCSIDriver: {
+        enabled: false  // Azure Files
+      }
+      snapshotController: {
+        enabled: false // CSI Snapshotter: https://github.com/kubernetes-csi/external-snapshotter
+      }
+    }
+    workloadAutoScalerProfile: {
+      keda: {
+        enabled: false  // Enable if using KEDA to scale workloads
+      }
+    }
     disableLocalAccounts: true
     securityProfile: {
       workloadIdentity: {
         enabled: true
       }
-      azureDefender: {
+      imageCleaner: {
         enabled: true
+        intervalHours: 120 // 5 days
+      }
+      azureKeyVaultKms: {
+        enabled: false // Not enabled in the this deployment, as it is not used. Enable as needed.
+      }
+      nodeRestriction: {
+        enabled: true // https://kubernetes.io/docs/reference/access-authn-authz/admission-controllers/#noderestriction
+      }
+      customCATrustCertificates: [] // Empty
+      defender: {
         logAnalyticsWorkspaceResourceId: la.id
+        securityMonitoring: {
+          enabled: true
+        }
       }
     }
     oidcIssuerProfile: {
@@ -1871,10 +1919,10 @@ resource acrKubeletAcrPullRole_roleAssignment 'Microsoft.Authorization/roleAssig
   }
 }
 
-// Grant the OMS Agent's Managed Identity the metrics publisher role to push alerts
-resource mcOmsAgentMonitoringMetricsPublisherRole_roleAssignment 'Microsoft.Authorization/roleAssignments@2020-10-01-preview' = {
+// Grant the Azure Monitor (fka as OMS) Agent's Managed Identity the metrics publisher role to push alerts
+resource mcAmaAgentMonitoringMetricsPublisherRole_roleAssignment 'Microsoft.Authorization/roleAssignments@2020-10-01-preview' = {
   scope: mc
-  name: guid(mc.id, 'omsagent', monitoringMetricsPublisherRole.id)
+  name: guid(mc.id, 'amagent', monitoringMetricsPublisherRole.id)
   properties: {
     roleDefinitionId: monitoringMetricsPublisherRole.id
     principalId: mc.properties.addonProfiles.omsagent.identity.objectId

diff --git a/networking/hub-regionA.bicep b/networking/hub-regionA.bicep
@@ -651,8 +651,10 @@ resource fwPolicy 'Microsoft.Network/firewallPolicies@2021-05-01' = {
                 '${split(environment().authentication.loginEndpoint, '/')[2]}' // Prevent the linter from getting upset at login.microsoftonline.com
                 '*.blob.${environment().suffixes.storage}' // required for the extension installer to download the helm chart install flux. This storage account is not predictable, but does look like eusreplstore196 for example.
                 'azurearcfork8s.azurecr.io' // required for a few of the images installed by the extension.
-                '*.docker.io' // Only required if you use the default bootstrapping manifests included in this repo. Kured is sourced from here by default.
-                '*.docker.com' // Only required if you use the default bootstrapping manifests included in this repo. Kured is sourced from here by default.
+                '*.docker.io' // Only required if you use the default bootstrapping manifests included in this repo.
+                '*.docker.com' // Only required if you use the default bootstrapping manifests included in this repo.
+                'ghcr.io' // Only required if you use the default bootstrapping manifests included in this repo. Kured is sourced from here by default.
+                'pkg-containers.githubusercontent.com' // Only required if you use the default bootstrapping manifests included in this repo. Kured is sourced from here by default.
               ]
               targetUrls: []
               destinationAddresses: []