This page explains how to update and delete an existing Dataproc cluster.
Update a cluster
You can update a cluster by issuing a Dataproc API clusters.patch request, running a gcloud dataproc clusters update command in a local terminal window or in Cloud Shell, or by editing cluster parameters from the Configuration tab of the Cluster details page for the cluster in the Google Cloud console.
The following cluster parameters can be updated:
- the number of standard worker nodes in a cluster—see Scaling clusters
- the number of secondary worker nodes in a cluster— see Preemptible VMs
- whether to use graceful decommissioning to control shutting down a worker after its jobs are completed
- adding or deleting cluster labels
Cluster update error
Dataproc issues an ERROR_DUE_TO_UPDATE
error when a cluster
update operation fails. This error places the cluster in a fail-safe state to
avoid data loss, and restricts further cluster operations: jobs can continue
to be submitted to the cluster, the cluster can be deleted, but further
cluster update operations are not permitted.
Recommendation: After receiving an ERROR_DUE_TO_UPDATE
on a cluster,
recreate the cluster
by exporting the cluster configuration to create a new cluster, then delete the
failed cluster. If recreating the cluster fails, contact
Cloud Customer Care for help in restoring the cluster to a
RUNNING
state.
Delete a cluster
You can delete a cluster via a Dataproc API clusters.delete HTTP or programmatic request, using the Google Cloud CLI gcloud command-line tool locally in a terminal window or in Cloud Shell, or from the Google Cloud console.
gcloud command
To delete a Dataproc cluster, use the gcloud CLI gcloud dataproc clusters delete command locally in a terminal window or in Cloud Shell.
gcloud dataproc clusters delete cluster-name \ --region=region
REST API
Use the Dataproc clusters.delete API to delete a cluster.
Console
Open the Dataproc Clusters page in the Google Cloud console. Select the cluster by checking the box to the left of the cluster name, then Click Delete to delete the cluster.
