Last modified March 11, 2020

gsctl delete nodepool

The gsctl delete nodepool command deletes a node pool.

Deleting a node pool means that all worker nodes in the pool will be drained, cordoned and then terminated.

In case you are running production workloads on the node pool you want to delete, make sure that there is at least one other node pool with capacity to schedule the workloads. Also check whether label selectors, taints and tolerations will allow scheduling on other pool’s worker nodes. The best way to observe this is by manually cordoning and draining the pool’s worker nodes and checking workload’s node assignments, before issuing the delete nodepool command.

Note: Data stored outside of persistent volumes will be lost and there is no way to undo this.

Usage

The command is called with the cluster and node pool ID as the only argument, separated by a slash.

Example:

gsctl delete nodepool f01r4/op1dl

Here, f01r4 is the cluster ID and op1dl is the node pool ID.

You can also use the cluster’s name for identifying the cluster:

$ gsctl delete nodepool "Cluster name"/op1dl

A confirmation will be required to finally delete the node pool. To suppress this confirmation, add the --force option.