Oracle Cloud Infrastructure Documentation

Removing a Node from a Cluster

Remove a node from a Big Data Service cluster.

Note

When you remove a node, Big Data Service decommissions the node before deleting it. However, if the decommissioning process doesn't finish in the 40-minute time limit, the request fails. Time taken to decommission a node depends on the number of blocks in the node that need to be moved. Therefore, we recommend that you decommission a node before removing it from the cluster.

To decommission a node in Ambari, do the following:

  1. Access Apache Ambari.
  2. From the side toolbar, under Hosts, and then select the worker node to be decommissioned.
  3. From the list of components installed on that host, select Data Node.
  4. the Action icon, and then select Decommission.
  5. If the host has Node Manager Region Server or HttpFS installed, then decommission them.
  6. From the side toolbar, under Services select HDFS. Before moving to the next step, ensure the following:
    • In the Components section, DATANODES STATUS > Decommissioning is 0.
    • In the Service Metrics section, BLOCKS > Under Replicated is 0.
    Verify node decommission in Ambari
  • To remove a node from a cluster from the OCI Console, complete the following:
    1. On the Clusters list page, select the cluster that you want to work with. If you need help finding the list page or the cluster, see Listing Clusters in a Compartment.
    2. On the Cluster details page, select the name of a node you want to work with.
    3. From the Actions menu (Actions Menu) for the node, select Delete node.
    4. In the Delete node panel, enter the name of the node to delete.
    5. Select Force delete even when decommissioning fails to delete the node even when decommissioning of the node fails.
    6. Select Delete.
  • Use the oci bds instance remove command and required parameters to remove a single node from a cluster.
    oci bds instance remove --bds-instance-id <bds_instance_id> --cluster-admin-password <cluster_admin_password> --node-id <node_id> [OPTIONS]

    For a complete list of flags and variable options for CLI commands, see the Command Line Reference for Big Data.

  • Use the RemoveNode operation to remove a single node from a cluster.