Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
ON THIS PAGE
 

Edit Cluster Nodes

Use the information provided in this topic to edit cluster nodes.

Add a Node

You can edit an existing and operational Paragon Automation cluster and add additional nodes to the cluster. The node can be either a primary or worker node.

However, note that if your existing cluster is configured with a single primary node, you cannot add an additional primary node using this procedure. To add an additional primary node to a single primary node cluster, you must reinstall the whole cluster.

To add new node(s):

  1. Prepare the node and ensure it meets all the cluster node prerequisites as described in Prepare Cluster Nodes.
  2. Log in to the control host.
  3. Edit the inventory file with the IP address or hostname of the new node. Add the node to the required cluster node group (primary or worker).
  4. Execute the ./run -c config-dir deploy command.

    You can also limit the execution of the command to deploy only the new node by using the ./run -c config-dir deploy -l node-address command.

Remove a Node

You can edit an existing and operational Paragon Automation cluster to remove one or more nodes from the cluster.

To remove node(s):

  1. Access kubectl.

    The main interface in the Kubernetes cluster is kubectl which is installed on a primary node. However, you can also access the Kubernetes API from any node that has access to the cluster, including the control host. To use a node other than the primary node, you must ensure that you copy the admin.conf file and set the kubeconfig environment variable, or you can use the export KUBECONFIG=config-dir/admin.conf command.

  2. Execute the following kubectl commands to remove a node.
  3. Log in to the control host. Update the inventory file to delete the node so that the inventory file also reflects the change. However, do not execute the ./run -c config-dir deploy command again.

Replace a Node

You can replace a node with another node in an existing Paragon Automation cluster.

To replace a node with another node:

  1. Prepare the node and ensure it meets all the cluster node prerequisites as described in Prepare Cluster Nodes.
  2. Access kubectl.

    The main interface in the Kubernetes cluster is kubectl which is installed on a primary node. However, you can also access the Kubernetes API from any node that has access to the cluster, including the control host. To use a node other than the primary node, you must ensure that you copy the admin.conf file and set the kubeconfig environment variable, or you can use the export KUBECONFIG=config-dir/admin.conf command.

  3. Execute the following kubectl commands to remove a node.
  4. Log in to the control host.
  5. Edit the inventory file with the IP address or hostname of the new node. Add the node to the required cluster node group (primary or worker). Also, delete the node you replaced.
  6. Execute the ./run -c config-dir deploy command.

    You can also limit the execution of the command to deploy only the new node by using the ./run -c config-dir deploy -l node-address command.

In some cases, if a node fails, you can rebuild a replacement node using the same IP address as the failed node.

To redeploy the same node as a replacement, perform the following steps:

  1. Prepare the node and ensure it meets all the cluster node prerequisites as described in Prepare Cluster Nodes.

  2. Log in to the control host.

  3. Execute the ./run -c config-dir deploy command again.

    You can also limit the execution of the command to only the redeployed node by using the ./run -c config-dir deploy -l node-address command.

Related Documentation