Deploy Amazon EKS to a new or existing VPC (AWS CLI)

You can use the AWS Launch Wizard CreateDeployment API operation to deploy Amazon EKS. To create a deployment, you must provide values for various specifications. Specifications are a collection of settings that define how your deployment should be created and configured. A workload will have one or more deployment patterns with differing required and optional specifications.

Note

If you want to use the Clone deployment action on your deployment, you need to create your deployment in the Launch Wizard console.

Prerequisites for deploying Amazon EKS with the AWS CLI

Before deploying Amazon EKS with the AWS CLI, ensure you have met the following prerequisites:

• Install and configure the AWS CLI. For more information, see Install or update to the latest version of the AWS CLI.

• Complete the steps in the previous section titled Set up. Some deployment patterns have requirements that must be met for a deployment to be successful.

Create an Amazon EKS deployment with the AWS CLI

You can create a deployment for your Amazon EKS application using the CreateDeployment Launch Wizard API operation.

To create a deployment for Amazon EKS using the AWS CLI

1. List the available workload names using the ListWorkloads Launch Wizard API operation.

   The following example shows listing the available workloads:

   aws launchwizard list-workloads --region us-east-1
   {
       "workloads": [
           {
               "displayName": "Remote Desktop Gateway",
               "workloadName": "RDGW"
           },
           {
               "displayName": "MS SQL Server",
               "workloadName": "SQL"
           },
           {
               "displayName": "SAP",
               "workloadName": "SAP"
           },
           {
               "displayName": "Microsoft Active Directory",
               "workloadName": "MicrosoftActiveDirectory"
           }
           ...
       ]
   }

2. Specify the desired workload name with the ListWorkloadDeploymentPatterns operation to describe the supported values for the deployment pattern names.

   The following example lists the available workload patterns for a given workload:

   aws launch-wizard list-workload-deployment-patterns --workload-name EKS --region us-east-1
   {
       "workloadDeploymentPatterns": [
           {
               "deploymentPatternName": "EKSExistingVpc",
               "description": "Example description.",
               "displayName": "ExampleDisplayName",
               "status": "ACTIVE",
               "workloadName": "EKS",
               "workloadVersionName": "2024-05-03-00-00-00"
           },
           ...
       ]
   }

3. Use the workload and deployment pattern names you discovered with the GetWorkloadDeploymentPattern operation to list the specification details.

   The following example lists the workload specifications of a given workload and deployment pattern:

   aws launchwizard get-workload-deployment-pattern --workload-name EKS --deployment-pattern-name EKSExistingVpc --region us-east-1
   {
       "workloadDeploymentPattern": {
           "deploymentPatternName": "EKSExistingVpc",
           "description": "Example description.",
           "displayName": "ExampleDisplayName",
           "specifications": [
               {
                   "description": "Enter an SNS topic for AWS Launch Wizard to send notifications and alerts.",
                   "name": "AWS:LaunchWizard:TopicArn",
                   "required": "No"
               },
               {
                   "description": "When a deployment fails, your provisioned resources will be deleted/rolled back by default. If deactivated, the provisioned resources will be deleted when you delete your deployment from the Launch Wizard console.",
                   "name": "AWS:LaunchWizard:DisableRollbackFlag",
                   "required": "No"
               },
               {
                   "allowedValues": [
                       "true",
                       "false"
                   ],
                   "description": "Cloud Watch Application Insights monitoring",
                   "name": "SetupAppInsightsMonitoring",
                   "required": "Yes"
               },
               ...
           ]
       }
   }

4. With the workload specifications retrieved, you must provide values for any specification name with a required value of Yes. You can also provide any optional specifications you require for your deployment. We recommend that you pass inputs to the specifications parameter for your deployment as a file for easier usage.

   Your JSON file's format should resemble the following:

   {
     "ExampleName1": "ExampleValue1",
     "ExampleName2": "ExampleValue2",
     "ExampleName3": "ExampleValue3"
   }

5. With the specifications file created, you can create a deployment for your chosen workload and deployment pattern.

   The following example creates a deployment with specifications defined in a file:

   aws launch-wizard create-deployment --workload-name EKS --deployment-pattern-name EKSExistingVpc --name ExampleDeploymentName --region us-east-1 --specifications file://specifications.json