Single machine deployment

You can deploy AKS Edge Essentials on either a single machine or on multiple machines. In a single machine Kubernetes deployment, both the Kubernetes control node and worker node run on the same machine. This article describes how to create the Kubernetes control node on your machine on a private network.

Prerequisites

Set up your primary machine as described in Set up machine.

Step 1: single machine configuration parameters

You can generate the parameters you need to create a single machine cluster using the following command:

New-AksEdgeConfig -DeploymentType SingleMachineCluster -outFile .\aksedge-config.json | Out-Null

This command creates a configuration file called aksedge-config.json that includes the configuration needed to create a single-machine cluster with a Linux node. The file is created in your current working directory. See the following examples for more options for creating the configuration file.

See Deployment JSON configuration for a detailed description of the configuration parameters.

The key parameters for single machine deployment are:

• DeploymentType: This parameter defines the deployment type and is specified as SingleMachineCluster.
• The Network.NetworkPlugin by default is flannel. This is the default for a K3S cluster. If you're using a K8S cluster, change the CNI to calico.
• You can set the following parameters according to your deployment configuration as described here: LinuxNode.CpuCount, LinuxNode.MemoryInMB, LinuxNode.DataSizeInGB, WindowsNode.CpuCount, WindowsNode.MemoryInMB, Init.ServiceIPRangeSize, and Network.InternetDisabled.

Step 2: create a single machine cluster

1. You can now run the New-AksEdgeDeployment cmdlet to deploy a single-machine AKS Edge cluster with a single Linux control-plane node:

New-AksEdgeDeployment -JsonConfigFilePath .\aksedge-config.json

Step 3: validate your cluster

Confirm that the deployment was successful by running:

kubectl get nodes -o wide
kubectl get pods -A -o wide

The following image shows pods on a K3S cluster:

Step 4: Add a Windows worker node (optional)

If you want to add a Windows node to an existing Linux only single machine cluster, first create the configuration file using the following command:

New-AksEdgeScaleConfig -ScaleType AddNode -NodeType Windows -outFile .\ScaleConfig.json | Out-Null

This creates the configuration file ScaleConfig.json in the current working directory. You can also modify the Windows node parameters in the configuration file to specify the resources that need to be allocated to the Windows node. With the configuration file, you can run the following command to add the node the single machine cluster:

Add-AksEdgeNode -JsonConfigFilePath .\ScaleConfig.json

Example deployment options

Create a JSON object with the configuration parameters

You can programmatically edit the JSON object and pass it as a string:

$jsonObj = New-AksEdgeConfig -DeploymentType SingleMachineCluster
$jsonObj.User.AcceptEula = $true
$jsonObj.User.AcceptOptionalTelemetry = $true
$jsonObj.Init.ServiceIpRangeSize = 10
$machine = $jsonObj.Machines[0]
$machine.LinuxNode.CpuCount = 4
$machine.LinuxNode.MemoryInMB = 4096

New-AksEdgeDeployment -JsonConfigString ($jsonObj | ConvertTo-Json -Depth 4)

Create a simple cluster with NodePort service

You can create a simple cluster with no service IPs (ServiceIPRangeSize set as 0):

New-AksEdgeDeployment -JsonConfigString (New-AksEdgeConfig | ConvertTo-Json -Depth 4)

Allocate resources to your nodes

To connect to Arc and deploy your apps with GitOps, allocate four CPUs or more for the LinuxNode.CpuCount (processing power), 4 GB or more for LinuxNode.MemoryinMB (RAM), and to assign a number greater than 0 to the ServiceIpRangeSize. Here, we allocate 10 IP addresses for your Kubernetes services:

{
  "SchemaVersion": "1.5",
  "Version": "1.0",
  "DeploymentType": "SingleMachineCluster",
  "Init": {
    "ServiceIPRangeSize": 10
  },
  "Network": {
      "NetworkPlugin": "flannel"
  },
  "User": {
      "AcceptEula": true,
      "AcceptOptionalTelemetry": true
  },
  "Machines": [
       {
         "LinuxNode": {
              "CpuCount": 4,
              "MemoryInMB": 4096
          }
       }
   ]
}

You can also choose to pass the parameters as a JSON string, as previously described:

$jsonObj = New-AksEdgeConfig -DeploymentType SingleMachineCluster
$jsonObj.User.AcceptEula = $true
$jsonObj.User.AcceptOptionalTelemetry = $true
$jsonObj.Init.ServiceIpRangeSize = 10
$machine = $jsonObj.Machines[0]
$machine.LinuxNode.CpuCount = 4
$machine.LinuxNode.MemoryInMB = 4096

New-AksEdgeDeployment -JsonConfigString ($jsonObj | ConvertTo-Json -Depth 4)

Create a mixed workload cluster

You can create a cluster with both Linux and Windows nodes. You can create the configuration file using the command:

New-AksEdgeConfig -DeploymentType SingleMachineCluster -NodeType LinuxAndWindows -outFile .\aksedge-config.json | Out-Null

Once the configuration file is created, you can deploy your cluster using the following command:

New-AksEdgeDeployment -JsonConfigFilePath .\aksedge-config.json

Next steps

• Deploy your application.
• Overview
• Uninstall AKS cluster