You can make basic configuration changes to your deployed Google Kubernetes Engine (GKE) resources by editing and redeploying the associated blueprint. This approach helps ensure consistent, repeatable deployments and supports managing your cluster's configuration as code. Using blueprints simplifies maintenance and promotes best practices for infrastructure management.
In this document, you learn how to reconfigure an active GKE cluster that Cluster Toolkit manages. You can update specific configurations, such as resizing a node pool. For fundamental architectural changes, such as switching to a new VPC or changing the scheduler, you must redeploy your cluster. For more information about Cluster Toolkit, see Cluster Toolkit overview.
Limitations
The following changes can't be made by editing the blueprint:
- You can't switch from a zonal to a regional GKE cluster. To switch from a zonal to a regional cluster, you must create a new cluster.
- You can't update the GKE version by using the Cluster Toolkit. To update the GKE version, you must change the version using the Google Cloud console or Google Cloud CLI.
Reconfigure the cluster
For GKE clusters, you can update the cluster as follows:
- Edit the deployment blueprint to update the supported parameters. For a list of supported parameters, see Supported parameters.
- Run the
./gcluster deploy <blueprint-yaml-file> -wcommand to redeploy the environment.
Supported parameters
When you update the blueprint, the following parameters can be modified.
| Parameters | Description |
|---|---|
autoscaling_total_min_nodes and autoscaling_total_max_nodes in a nodepool |
Use this parameter to update the minimum and maximum number of nodes that exist within an autoscaling nodepool. |
static_node_count in a nodepool |
Use this parameter to adjust the static number of nodes in the nodepool. You can also replace this parameter with autoscaling_total_min_nodes and autoscaling_total_max_nodes to switch your cluster from a static to an autoscale nodepool. |