Upgrade Kubernetes Cluster

This page outlines the process for upgrading the Cluster Control Plane and Node Pool for Terraform-based infrastructure deployments.

When to upgrade

Each new release of the Terraform script artifacts may specify a different version of the Managed Kubernetes Service for each supported cloud provider: AWS (Amazon Web Services), ACS (Azure Cloud Services) or GCP (Google Cloud Platform).

Because Kubernetes version upgrades can introduce breaking changes, simply re-running deploy-cluster.sh may fail or produce unexpected results. To avoid this, you should follow a targeted upgrade process that focuses specifically on updating the Control Plane and Node Pool, without affecting the rest of the infrastructure.

You can also use this process if you need to upgrade the Cluster version independently of the Terraform artifact version.

Upgrade the Control Plane and Node Pool

1. Open the Terraform variables.tf file for your relevant cloud provider (for example, AWS, ACS, or GCP) and update the cluster_version variable to match the Kubernetes version you are upgrading to.

   AWS

   ACS

   GCP

   console

   Copy

   vi terraform/aws/variables.tf
   variable "cluster_version" {
   ....
   default = "1.32"
   }                                        

   console

   Copy

   vi terraform/azure/variables.tf
   variable "cluster_version" {
   ....
   default = "1.32"
   }

   console

   Copy

   vi terraform/gcp/variables.tf
   variable "cluster_version" {
   ....
   default = "1.32"
   }

2. Run the manage-cluster.sh script.

   Linux

   Windows

   console

   Copy

   ./scripts/manage-cluster.sh

   console

   Copy

   .\scripts\manage-cluster.bat

3. Authenticate to cloud provider.

   AWS

   ACS

   GCP

   console

   Copy

   ./scripts/terraform.sh authenticate                                

   console

   Copy

   ./scripts/terraform.sh authenticate

   console

   Copy

   ./scripts/terraform.sh authenticate

4. Run the terraform init command.

   console

   Copy

   ./scripts/terraform.sh init cluster

5. Switch to the directory of the relevant cloud provider.

   AWS

   ACS

   GCP

   console

   Copy

   cd terraform/aws                                

   console

   Copy

   cd terraform/azure

   console

   Copy

   cd terraform/gcp

6. Export the required environment variables to run Terraform.

   AWS

   ACS

   GCP

   console

   Copy

   export TF_VAR_cluster_name=$CLOUD-$ENV
   export TF_VAR_region=$REGION                                

   console

   Copy

   export TF_VAR_cluster_name=$CLOUD-$ENV
   export TF_VAR_region=$REGION

   console

   Copy

   export TF_VAR_cluster_name=$CLOUD-$ENV
   export TF_VAR_region=$REGION
   export TF_VAR_project=$PROJECT
   zones=$(gcloud compute zones list --filter="region:${TF_VAR_region}" --format='value(name)' | sort | head -1)
   export TF_VAR_zones="[\"${zones}\"]"

7. Run the terraform taint command to mark the Node Pool for replacement.

   AWS

   ACS

   GCP

   console

   Copy

   terraform taint module.kx-aws.module.eks.module.eks_managed_node_group[\"default\"].aws_eks_node_group.this[0]
                                       

   console

   Copy

   terraform taint module.kx-azure.null_resource.upgrade_default_node_pool

   This command is not applicable to GCP.

8. Run the terraform apply command to upgrade the Control Plane and Node Pool.

   AWS

   ACS

   GCP

   console

   Copy

   terraform apply -target=module.kx-aws.module.eks.module.eks_managed_node_group[\"default\"].aws_eks_node_group.this[0] -target=module.kx-aws.module.eks.aws_eks_cluster.this[0]                                

   console

   Copy

   terraform apply -target=module.kx-azure.azurerm_kubernetes_cluster.cluster -target=module.kx-azure.null_resource.upgrade_default_node_pool

   console

   Copy

   terraform apply -target=module.kx-gcp.module.gke.google_container_cluster.primary -target=module.kx-gcp.module.gke.google_container_node_pool.pools[\"gcp-demo-default-node-pool\"] (assumes cluster name is `gcp-demo`)